Does “Corporation X” follow the hacker way?

Got a query from a journalist today working on a major story about a certain large corporation that’s been much in the news lately. Seems the corporation’s founder has been talking up his organization’s allegiance to “the hacker way”, and she not unreasonably wanted my opinion as to whether or not this was complete horse-puckey.

So as not to steal the lady’s thunder, I won’t reveal the identity of Corporation X. I will, however, repeat a version of my answer with its identity lightly obscured – because I think these are questions we should ask any corporation that talks like that.

I answered:

I have never used Corporation X’s website myself. Therefore, rather than directly opining about what Corporation X espouses, I’ll suggest some criteria your readers can apply for themselves.

As a user of Corporation X’s site, do I have control of the data Corporation X keeps about me? Concretely: can I examine and modify that data using tools of my choosing which are built for my needs?

Does Corporation X act as though I own my on-line life, or as though it does? Concretely: Can I control what data it shares with other users, with advertisers, and with business partners?

Does Corporation X behave like a tool in my hand, or a firehose designed to spew at me in accordance with other peoples’ agendas? Concretely: can I write my own client to present a filtered view of the Corporation X data stream, or have other people do that for me?

No prize for guessing that “the hacker way” is to give control to the individual, to respect his or her privacy, to create tools for autonomy and liberty, and to encourage creative re-use of software.

And yes, one of the most basic questions is “Does Corporation X publish the source code of its user-facing software in a form that can easily be understood, modified, and reused?” Because if the answer to that question is “no”, it is very unlikely that users will or ever can have the control of their on-line lives that they deserve.

UPDATE: Now that the article has been published, I can confirm that “Corporation X” is indeed Facebook.

484 comments

  1. *cough*facebook*cough*

    Their discussion of how they follow “the hacker way” has already made the news. And I don’t use it either.

  2. > I won’t reveal the identity of Corporation X

    Yeah, it’s not even close to being hard to guess. I suppose that’s deliberate.

  3. Caveat: what does “user-facing” mean these days? View source? Do I care about the schema of the databases?

    It’s a good list, though.

  4. At one of my last companies, all of our web GUI pages were structured that if you passed in a parameter to the query the result you got back was effectively the same data, but as a JSON dump. This was used internally by some of our own implementation and tools, and allowed the whole website to be scripted. Written by hackers, used by everybody. Its also good practice, too, because it allows for the separation of UI and data. As long as the UI logic is separate, you can validate the entire data-processing component through automated testing, leaving only the UI processing code to be tested manually as that actually requires a human to look at and see if things are working correctly.

    Good design leading to better quality and testability.

    1. >Good design leading to better quality and testability.

      Damn. Garret, can you email me more details? That sounds like a good case study for the second edition of TAOUP.

  5. > That sounds like a good case study for the second edition of TAOUP.

    Which leads to the first OT question:
    Have you got AW interested in publishing a second edition?

  6. Garrett, I’d like to know more also. I’m building a history site and if someone could grab the data and play with it, this would make it very easy to build things like maps and timelines.

  7. While the company in question is certainly not doing a good job at applying the “hacker way”, there is a positive side to this: their use of the term “hacker” is much closer to how we in the hacker community define it than to how the media has corrupted the term in the past. Said company has given us an opportunity to have a discussion with the media about the true meaning and values of being a “hacker”. I say we take it.

    1. >Said company has given us an opportunity to have a discussion with the media about the true meaning and values of being a “hacker”. I say we take it.

      Yup. I’m already on the job.

  8. Fecesbook goes out of their way to tighten the clamps on your eyeballs all the time. They will not rest until they have a direct feed into your brain that cannot be shut off.

    In the old days, you used to be able to fetch their data stream as an RSS feed, parse it using whatever tools you wanted, and display it however you wanted. This allowed you to have your daily dose of high school drama queens displayed to you along with whatever else you follow on the Internet. About a year ago they shut that off. They want you to use their web site directly. I was about ready to quit Fecesbook by then anyway, so I did.

    Let’s go ahead and stab all of the Google-bashers, because although it’s trendy to bash Google they really are doing better than anyone else out there.

  9. In fact, ESR, some of us turn down FB interviews for the reasons you expressed above..

    But to be fair and honest..Google is closer..but not fully there yet..

    1. >Google is closer..but not fully there yet..

      Agreed. The Data Liberation Front is a very large step in the right direction.

  10. Eric,

    Apparently you missed that Comscore released results today

    Total Smartphone Subscribers
    Google 44.8% 47.3% +2.5
    Apple 27.4% 29.6% +2.2
    RIM 18.9% 16.0% -2.9
    Microsoft 5.6% 4.7% -0.9
    Symbian 1.8% 1.4% -0.4

    http://www.comscore.com/Press_Events/Press_Releases/2012/2/comScore_Reports_December_2011_U.S._Mobile_Subscriber_Market_Share

    Quoting: “Wake me up when Apple’s smartphone growth rate exceeds Android’s smartphone growth rate in the comScore numbers.”

    Exceeds? No. But damned close to parity.

    1. >Exceeds? No. But damned close to parity.

      So after all that exulting about Apple’s sooper dooper whizbang record quarter with fireworks and angelic hosannas Android is still increasing its lead.

      ROFLMAO. And I’m laughing at you, not with you, fanboy. How’s that heaping helping of crow taste?

      “Damned close to parity.” Maybe that’s what they’ll chisel on the iPhone’s grave marker.

  11. > But to be fair and honest..Google is closer..but not fully there yet..

    Though Google is slower on it’s feet, and FB is Google’s enemy #1 right now.

    Facebook’s engagement is shifting to mobile. Their entire business model relies on web. Their app ecosystem has been crashing as a result, and display advertising is likely to follow.

    Facebook is filled with smart people. They understand the trends better than we do.

    From Facebook’s S-1, 4 of their top 5 Risk Factors relate to the web-to-mobile shift and the problems inherent with that:

    #2 – “We generate a substantial majority of our revenue from advertising. The loss of advertisers, or reduction in spending by advertisers with Facebook, could seriously harm our business;”
    #3 – “Growth in use of Facebook through our mobile products, where we do not currently display ads, as a substitute for use on personal computers may negatively affect our revenue and financial results;”
    #4 – “Facebook user growth and engagement on mobile devices depend upon effective operation with mobile operating systems, networks, and standards that we do not control;”
    #5 – “We may not be successful in our efforts to grow and further monetize the Facebook Platform;”

    Most of the insane/easy growth (in profit as well as valuation) has already been squeezed out. I think this explains why the IPO exit is happening now. (Gets the investors whole.)

    There is still a huge opportunity for Facebook to grow revenue/profit if they switch to using personal data to make well-directed ads they can push to sites using Facebook Connect and/or if they can roll out a solid first-party Facebook-phone. They could quite possibly fork Android and do to Google in the phone market what Amazon is doing to the tablet market.

    And they’d be crazy not to try.

    Facebook will never grow revenue to 10x (matching Google) by continuing to sell display advertising on its own site. However, it knows a lot about its users (and keeps them logged in), so it definitely has potential to serve highly targeted or viral campaigns everywhere else on the web and mobile apps, yielding a much better version of adsense.

    This is the real reason Google is getting desperate about Google+, they are worried about FB launching its own hyper targeted ad network. Google has context and tons of advertisers, FB has context and deep demographics and quite a lot of advertisers.

    Let the deathmatch begin.

  12. > Got a query from a journalist today working on a major story about a certain large corporation that’s been much in the news lately.

    Since everyone understands “Corporation X” is Facebook, it would be more interesting if you released the name of the reporter, or at least her publication. I, for one, am curious what she might have done to your quotes.

  13. Finally, Android, with 47.3%, is not yet at 50%, (forecast for last Oct) and won’t be next month, either.

  14. > ROFLMAO. And I’m laughing at you, not with you, fanboy. How’s that heaping helping of crow taste?

    Roasted.

    What’s your problem, Eric? Does this all hit to close to home? Has anyone ever told you that you’re quite rude?

    Look at your own data. 2.5% growth has been normal for Android. Apple has never grown so fast according to the same data. Except for August, Apple has never grown even half as fast as it just did.

    Wall Street is finally waking up to the juggernaut that is Apple. Everybody loves a winner.

    Did you look at Amazon’s results? (a huge loss, more to come) Google’s? (also a huge loss) Motorola? (a f***ing huge loss).

    Meanwhile Google is beset on all sides by those who seek its undoing.

    The worm has turned. The fat is in the fire. Who knows what will happen now?

    You can rail all you like, but while you froth at the mouth, you’re missing the larger picture.

    http://yfrog.com/mm1grp

    1. >Has anyone ever told you that you’re quite rude?

      Yes, and I revel in it. You begged for rudeness. I’ve been listening for weeks to you and the other besotted Apple fanboys wax all snotty about how 4Q2011 was going to be the quarter where it all turned around because the 4S is the platinum-plated hammer of the gods that will smite Android with great smitingness, uh huh uh huh. So now I am going to gloat at your FAIL and the fact that once again, my predictions were correct. I shall gloat obscenely. See me gloat.

      Heed the lesson and try being a bit less of an overbearing ass in the future. I can be surprisingly polite to people who aren’t overbearing asses.

  15. > So after all that exulting about Apple’s sooper dooper whizbang record quarter with fireworks and angelic hosannas Android is still increasing its lead.

    I refuse to believe that you’re really this blind.

    In the first quarter that iPhone is available on 3 of the 4 major characters, it turns in a growth rate 88% of Android’s.

    Sprint announces results next Weds. It will be interesting to see how fast they sold iPhone.

    Of the top 5 OEMs, only Apple grew. Samsung, by far the largest Android OEM had ZERO growth.
    Samsung 25.3% 25.3% 0.0
    LG 20.6% 20.0% -0.6
    Motorola 13.8% 13.3% -0.5
    Apple 10.2% 12.4% 2.2
    RIM 7.1% 6.7% -0.4

  16. His problem, Larry, is that you’ve continually made extraordinary claims about Apple’s sales and yet you came in her exultant about numbers that don’t match your hype.

  17. Returning different views of the same data on a website is baked into the mindset of the Ruby and Rails communities. The Rails framework makes it ridiculously easy to code REST-ful sites where the url http://mysite.com/resources/.html returns an HTML page, http://mysite.com/resources/.json returns a json view of the data and http://mysite.com/resources/.xml returns an xml view of the data. Turning this functionality on is literally two lines of code if the default output of the built-in json and xml generators is acceptable.

    Many startups using Ruby and Rails are using this ability to expose the site’s data to third parties.

    I am not as familiar with other languages and frameworks, but pretty much all the modern ones have bought into this kind of exposure at one level or another. Whether or not a company chooses to use it is another matter.

  18. @esr

    >“Damned close to parity.” Maybe that’s what they’ll chisel on the iPhone’s grave marker.

    You almost got through it without saying something silly.

    It’s fine to be happy with Android’s market share, but you should really stop making predictions about the demise of the iPhone and Apple. It’s not going to happen, and your credibility suffers.

  19. By the way, on the subject of ‘Corporation X’, it’s incredible to me that anybody believes a single word that comes out of CEO X’s mouth. His ‘the hacker way’ bullshit is beyond belief considering his track record.

    And yet the tech press laps it up.

    1. >And yet the tech press laps it up.

      I’m not going to complain. It’s what they call a “teachable moment”.

  20. To be fair, the comscore numbers do reflect growth for Apple that almost any other company would kill for — a 21% increase in the number of US subscribers carrying Apple phones over a three month period. Also to be fair, with the 3 month averaging (possibly weighted) that comscore does, we might not understand the full impact of Christmas buying for another couple of months.

    However, the net add in Android users over that same 3 month period was 7 million, compared to the net add in Apple users of 5 million. Compare the 5 million number to the 12 million Apple handsets sold at AT&T & Verizon (are Sprint numbers even out yet?) in the last quarter and even with comscore’s reporting lag it becomes probable that a lot of the Apple faithful (a) upgraded and (b) didn’t sell or give their old handsets away (kept for a backup device?).

    1. >it becomes probable that a lot of the Apple faithful (a) upgraded and (b) didn’t sell or give their old handsets away (kept for a backup device?).

      I’m going to blog about these numbers with some analysis, probably later tonight. Let’s have the discussion on that thread.

  21. By the way, on the subject of ‘Corporation X’, it’s incredible to me that anybody believes a single word that comes out of CEO X’s mouth. His ‘the hacker way’ bullshit is beyond belief considering his track record.

    I think you mean CEO Z amirite?

    And he’s half-right, but only half-right. His version of “the hacker’s way” is, loosely speaking, agile methodologies, which Eric correctly notes were pioneered in hackerdom. Compared to most places that produce software, Corporation X is a hacker’s paradise. (Try working on a government contract sometime, all the documentation you are required to produce and meetings you are required to sit in on before a line of code is even written. Be enlightened, and despair.)

    Perhaps the crucial point of the hacker’s way is openness and transparency, something even the White House pays lip service to nowadays (though not much of any other kind of service). Openness and transparency will be woven into Corporation X’s culture shortly after a nor’easter strikes Hades. Ergo, despite the closeness, Corporation X still don’t get the cigar.

  22. >Quoting: “Wake me up when Apple’s smartphone growth rate exceeds Android’s smartphone growth rate in the comScore >numbers.”

    >Exceeds? No. But damned close to parity.

    What’s going to be interesting is next month. Can they make two in a row, or as I suspect will happen, this is a ‘one in a row’ trend.

    Jim

  23. >Finally, Android, with 47.3%, is not yet at 50%, (forecast for last Oct) and won’t be next month, either.

    The difference between 49.8% and 50% is way smaller than the noise level for this kind of data. I will accept that android reached 50% market share at about the time this data represents. Which I believe is a month or six weeks ago?

  24. @Jeff Read

    Compared to most places that produce software, Corporation X is a hacker’s paradise.

    Maybe so, but as somebody who has had the distinct displeasure of working on third-party apps for the Facebook platform (something I absolutely will never do again, despite the stupid amounts of money you can charge for such services) their software architecture is about as far from the Unix ideal of small, well-documented, elegant programs that work well with and use a common input/output standard as you can possibly get. My God, I shudder at the memory of all the crap I had to go through.

    I don’t want to talk about it any more.

  25. @Jim Hurlburt

    The difference between 49.8% and 50% is way smaller than the noise level for this kind of data.

    49.8%? Did I miss something? What is this number?

    1. >49.8%? Did I miss something? What is this number?

      I suspect it’s just a typo, and what Jim meant is that the difference between 47-whatever% and 50% is within the noise level for these measurements.

      He’s right, but also wrong. The actual number 50% is magic, in this context, for psychological reasons. We won’t see a tipping point until the number people see is above 50%.

  26. > Which I believe is a month or six weeks ago?

    At least 6 months later than Eric predicted.

    > 49.8%? Did I miss something? What is this number?

    47.3% (comScore’s marketshare for Android in the Dec ’11 report) + 2.5% (the growth over that same period).

    1. >At least 6 months later than Eric predicted.

      No, we’re only at about three months of plus time now. My original linear extrapolation said later October.

      And I was careful to note that without a generative theory of why the rise looked linear we shouldn’t actually trust the extrapolation – I actually wrote a mini-tutorial on how not to get fooled using it as an example.

  27. “Android is still increasing its lead.”

    It’s lead decreased from 18.2% (46.9% – 28.7%) to 17.7% (47.3% – 29.6%).

    “The difference between 49.8% and 50% is way smaller than the noise level for this kind of data. I will accept that android reached 50% market share at about the time this data represents. Which I believe is a month or six weeks ago?”

    Not sure where you’re grabbing that 49.8% number, but since 29.6% is closer to 30% than 47.3%, I presume it’s safe to claim that Apple has already surpassed 30% as well. It’s unclear to me how a platform with 50% share has such dynamic network effects and cost efficiencies as to cause the decline of a platform with 30% share, 90% approval rating, and 60% of the entire markets profits. Nor do I see an analog to Windows vs. Mac.

  28. edit: …”but since 29.6% is closer to 30% than 47.3%” is to 50%…, “I presume…”

  29. 47.3% (comScore’s marketshare for Android in the Dec ’11 report) + 2.5% (the growth over that same period).

    Isn’t that 2.5% growth from September to December? Three months?

    I just don’t quite understand why we are adding 2.5% to the December numbers and deciding that represents the situation a month ago.

    By the way, when they say ‘3 months ending December 2011’ does that mean ending at the start of December, or the end of December?

  30. “And yes, one of the most basic questions is “Does Corporation X publish the source code of its user-facing software in a form that can easily be understood, modified, and reused?” Because if the answer to that question is “no”, it is very unlikely that users will or ever can have the control of their on-line lives that they deserve.”

    I find this statement too ideologically bent as to have much merit, if any: if the ultimate goal is control of personal data, I don’t see any need for source code to achieve that goal.

  31. @esr

    He’s right, but also wrong.

    OK, but if the argument is that 47.3% could actually be 50%, then presumably it could also be 44.6%.

    The actual number 50% is magic, in this context, for psychological reasons. We won’t see a tipping point until the number people see is above 50%.

    Sorry if you have explained this before, but is this analysis based on data from another analogous market? That is, do we have an example of another market where share for a particular product has suddenly exploded after that product’s share goes over 50%?

  32. “Android is still increasing its lead.”

    Ah, presumably you are using September’s #s and measuring over the 3 months rather than looking at monthly sequential growth. My apologies.

  33. > What’s going to be interesting is next month. Can they make two in a row, or as I suspect will happen, this is a ‘one in a row’ trend.

    Well, if it’s a single-time event, then Eric will be sure to crow about it. Eric also wants to suggest that comScore’s data has more veracity than the numbers out of AT&T, Apple, and Verizon’s 10-Q reports, which can’t be true.

    comScore’s data is potentially (quite) useful for trend analysis, but that’s about it. Android hasn’t grown their market share at more than 2% during the past year.

    Looking at “U.S. smartphone market share by platform, change per month” on http://www.catb.org/esr/comscore/, we find:

    Android:
    Jan: 2.50%, Feb: 1.80%, Mar: 1.70%, Apr: 1.70%, May: 1.70%: Jun: 2.00%, Jul: 1.70%, Aug: 1.90%, Sep: 1.10%, Oct: 1.50%, Nov: 0.60%, Dec: 2.5%

    iPhone:
    Jan: -0.30, Feb: 0.50%, Mar: 0.30%, Apr: 0.50%, May: 0.60%, Jun: 0.00%, Jul: 0.40%, Aug: 0.30%%, Sep: 0.10%, Oct: 0.70%, Nov: 0.60%, Dec: 2.2%

    and then calculate Android/iPhone:
    Jan: ???, Feb: 3.6x, Mar: 5.6x, Apr: 3.4x, May: 2.83x, Jun: , Jul: 4.25x, Aug: 6.3x, Sep: 11x, Oct: 2.14x, Nov 1.0 (parity), Oct: 1.13x

    Last month, Eric wrote, “This month’s comScore results are out, with no drama. Android continues to out-grow Apple by about 2:1 while other players dwindle towards insignificance.”

    And it’s easy to see that Android hasn’t outgrown Apple by 2:1 since October. Who is ‘crowing’ now? Who’s statements are full of #FAIL?

    We were talking about marketshare, right?

    This month comScore’s data shows that Android is no longer continuing “to out-grow Apple by about 2:1”, though Eric is only dismissive of anyone suggesting that this is an interesting event. It’s more interesting to look at what has been happening since October. Android, which had been outpacing Android by a huge margin (again, this all according to the comScore data), is now at functional parity with Apple’s growth.

    We are left only to ask, “What happened in October?”

  34. > We won’t see a tipping point until the number people see is above 50%.

    Well yeah, that has to be true by definition. The question is ‘where above 50%?”

    It’s not 100%, and it’s probably not 50.5%, either.

    So you’ll need to provide a sharper figure (something I predict you won’t actually do.)

    1. >So you’ll need to provide a sharper figure (something I predict you won’t actually do.)

      Well predicted! Now, let’s talk about the difference between humility and timidity.

      I’m not going to say “tipping point will happen at percentage foo” because I don’t think I have a rational basis for knowing that. But I’m not timid – if I thought I had such a basis, I’d ship that number.

      I have a pretty good predictive record. I’m not always right, but I’m right more often than I’m wrong, and I’m frequently right about large predictions that matter. The formula for this kind of success is simple: lots of domain knowledge, epistemic humility, and fearlessness. You have to be very aware of the limits of your knowledge, but utterly unafraid of speaking up about where your logic takes you even when you know “conventional wisdom” will consider it crazy and ungrounded.

      Another aspect of being fearless is refusing to be baited, either by your own ego or by needling from others, into making predictions for which you don’t have a sound rational basis. This is me refusing to be baited.

  35. @Larry Yelnick

    Well yeah, that has to be true by definition.

    No, it doesn’t. There is nothing in the ‘definition’ of a tipping point that says it has to come after 50%.

  36. “That is, do we have an example of another market where share for a particular product has suddenly exploded after that product’s share goes over 50%?”

    I refer to it as the generative theory of why? Because the magician himself says so. No, I don’t think ESR has ever presented a rational argument or analog for why 50% is magical.

  37. > My original linear extrapolation said later October.

    What you said was: “50% market-share crossover for Android still looks like being sometime in October 2011.” #FAIL

    and

    “Accordingly, I no longer think there is any plausible scenario under which Android fails to achieve over 50% smartphone market share in the U.S. and worldwide this year.” #FAIL

    and

    “Android share growth continues to bucket along at about 2% a month, while Apple’s shows no increase in the latest figures.” (Aug 7, after which Android didn’t see 2%/month until this past month.) #FAIL

    I could continue, but that would be rude.

  38. > No, it doesn’t. There is nothing in the ‘definition’ of a tipping point that says it has to come after 50%.

    The market has clearly tipped at some share prior to 100%.

  39. @Larry

    >> No, it doesn’t. There is nothing in the ‘definition’ of a tipping point that says it has to come after 50%.

    >The market has clearly tipped at some share prior to 100%.

    Sorry, but I don’t understand what this means.

  40. “And I was careful to note that without a generative theory of why the rise looked linear we shouldn’t actually trust the extrapolation – I actually wrote a mini-tutorial on how not to get fooled using it as an example.”

    And that entire post, while admittedly & nobly acknowledging your lack of knowledge, appeared to focus on suggesting that the growth was likely greater than linear, opting to mention but seemingly disregard the likely true reason for linear growth (smartphone growth is constrained by production capacity) which others have been stating for a rather long time.

  41. Moreover, you posited:

    “Under these conditions, the entire Android army would have to be afflicted by some Android-specific design or execution failure for sales growth to go sublinear.”

    Of course, there are numerous reasons for sublinear growth besides some design or execution failure, but that doesn’t matter. It gets better, you then posited:

    “The Android Army is not going to be hit by some kind of huge gotcha like a massive design failure or patent TRO.”

    While this is true (I presume you don’t think a huge gotcha emerged over the last 3 months), you were completely wrong to state that it would not go sublinear.

    So it looks like you got fooled in your post about not getting fooled and/or you think there is some massive design or execution failure at play with the Android platform.

    1. >So it looks like you got fooled in your post about not getting fooled and/or you think there is some massive design or execution failure at play with the Android platform.

      You misread that paragraph. It was one of several examples of generative theory intended to illustrate the point that a linear (or any other) regularity in a graph is meaningless without such a generative theory. I didn’t actually commit to believing any of them; that wasn’t the point. Thus, “Such a theory needn’t be super-elaborate to make useful predictions. For example, …”. And that’s what the theory you’re pointing at was – a tutorial example.

      See also: “If that’s true, my linear prediction will be too conservative. Alternatively, growth might be about to nose over into the saturation part of a logistic curve. There’s simply no way to know this from the data; you have to have a model underlying your curve-fit, an independently confirmed theory about the future.”

      My whole point in that post is that we didn’t have a model underlying the curve fit, and we therefore shouldn’t be fooled by the apparent linearity. I never lost sight of this point; apparently some of the rest of you did.

  42. And I apologize in advance for suggesting any intention where you were clearly speaking hypothetically despite a particular bent coming through in your statements in the More Fun with Statistics post. I can see how you can dismiss anything and/or everything in that post as a hypothetical examination of forming a theory.

    Putting aside what I said in my two previous comments and what appears to be your beliefs underlying that post, the 4 premises you chose to support your linear growth and 50% market share in Oct prediction held more or less true (although I would quibble with all of them) and yet the model did not bear out. That would suggest to me: you are unable or not good at identifying key factors affecting the dynamics of this market.

    1. >That would suggest to me: you are unable or not good at identifying key factors affecting the dynamics of this market.

      And yet I have a record of predictions about it that were both bold and successful. Including my most recent one. When that stops happening, I will consider that your suggestion might be worth more than a fart in a windstorm.

  43. @Tim F.

    you are unable or not good at identifying key factors affecting the dynamics of this market.

    Tim, you are making this too personal. If your aim here is to embarrass Eric or somehow extract a retraction then you are wasting your time.

    Yes, Eric’s prediction of October was wrong, and he has admitted that. So far it is a little bit wrong, we will see if it becomes extremely wrong.

    However, the interesting thing here is not to point the finger and laugh. That is boring. The interesting thing would be to ask if there is some underlying reason for the prediction not holding up, or whether it is just noise.

  44. Or rather… I apologize in RETROSPECT since I didn’t refresh quickly enough. I didn’t misread: I understand that on the surface you were seeking out unknowns to preserve intellectual honesty, but it’s easy enough to read the subtext. Now I’m just trying to reconcile these statements:

    “I have a pretty good predictive record. I’m not always right, but I’m right more often than I’m wrong, and I’m frequently right about large predictions that matter. The formula for this kind of success is simple: lots of domain knowledge, epistemic humility, and fearlessness.”

    …with…

    “My whole point in that post is that we didn’t have a model underlying the curve fit…”

    Maybe you should have a model? Maybe you should correctly identify the factors underlying whether or not your prediction will be correct? Maybe you shouldn’t irrationally pin your predictions on 50% market share… of the US only… of smartphones only as a basis for a theory for platform success.

    Some of us do have a model. Don’t blame your lack of one when your predictions fail.

    1. >Maybe you should have a model? Maybe you should correctly identify the factors underlying whether or not your prediction will be correct? Maybe you shouldn’t irrationally pin your predictions on 50% market share… of the US only… of smartphones only as a basis for a theory for platform success.

      You’re confounding several different issues here. I don’t need to have a generative model of why marketshare changes to predict consequences of it changing. There’s a difference between having theory about when share will pass 50% and having theory about what will happen when it does.

      And “irrational”? Really? Your ignorance about tipping-point effects in technology markets is profound if you can say that and mean it. Might be I’m wrong from here to Andromeda, but I’m not maintaining a position any economic historian would find even peculiar, let alone irrational.

  45. “Tim, you are making this too personal. ”

    I don’t see how saying the only factor which could make Android growth go sublinear is a massive design or execution failure on their own part is good observation of the factors affecting the “Smartphone Wars” nor do I see how saying so is getting “personal.” Unless you interpret “accurate” or “on the nose” as personal.

  46. @Tim F.

    Your conclusion, which I quoted in my post, directly attacks Eric. It attacks his analytical ability. That is personal.

    Just stick to attacking arguments and you will be better off.

  47. >Said company has given us an opportunity to have a discussion with the media about the true meaning and values of being a “hacker”. I say we take it.

    On the other hand, maybe we should pick our battles.

    John Q. Public thinks “hacker” means a criminal. Why fight it? Instead let’s call ourselves techies or geniuses or tricksters or something else, and move on.

  48. “And yet I have a record of predictions about it that were both bold and successful.”

    I’m not seeing it. Where you claim bold, I see widely held beliefs. Where you see successful, I see you wiggling away from your own statements, admitting you have no underlying model to base your predictions on, and conceding that much of your prediction (the key 50% market share figure) has no rational basis. I’ve literally watched you move the goalpost on timing about 3 or 4 times over the last 5 months… and then claim that you’ve never had a time element to your predictions. I’ve had a massively difficult time getting you to simply acknowledge what you are or are not predicting. I don’t even have a clue as to what you think is your most recent prediction is, never mind if it was correct.

    1. >I’m not seeing it.

      OK, let’s take some samples. Did you predict when Android’s market share would pass Apple’s many months in advance and get it right to within weeks? Did you predict that the much heralded end of the AT&T exclusive would fail to even speed-bump Android’s growth? Did you predict the trends away from skins and locked bootloaders? Did you predict that Apple would lose not just the market share but the technology lead in features like 4G, and explain why?

      I did. In all these cases I was bucking conventional wisdom at the time. But I knew why I thought these things would happen, and I showed my work. And we have at least one commenter (Daniel Franke) who reports having bet on my predictions in the stock market and made money.

      Now, do yourself a favor and shut up until your record is not embarrassing by comparison.

  49. “I don’t need to have a generative model of why marketshare changes to predict consequences of it changing.”

    No, but you haven’t just been predicting the consequences of changes. You have actually been predicting that nothing would alter the current dynamics of the market.

    “There’s a difference between having theory about when share will pass 50% and having theory about what will happen when it does.”

    Yes, and so far, it looks like you’ll be wrong on both counts.

    “And “irrational”?”

    Yes, I find it utterly irrational to predict a change in the dynamics of the smartphone market based on the arbitrary figure of 50% US market share of smartphones. I have not seen you make a rational argument on behalf of said belief. No, I do not think that finding that figure irrational shows any ignorance of tipping point effects in technology markets.

  50. “Did you predict when Android’s market share would pass Apple’s many months in advance and get it right to within weeks?”

    Yes, I predicted it very early. I do not recall how accurate it was.

    “Did you predict that the much heralded end of the AT&T exclusive would fail to even speed-bump Android’s growth?”

    Yes, I predicted it would take more than a year to see any substantial effect.

    “Did you predict the trends away from skins and locked bootloaders?”

    No, because there is not a trend away from skins.

    “Did you predict that Apple would lose not just the market share but the technology lead in features like 4G, and explain why?”

    I did not predict they would lose a market share lead because they never had one. By the time they eclipsed Nokia and RIMM, Android had already long surpassed them. I did not predict Apple would lose a technological lead because they never had one; Apple has always been willing to lag on “must have” technology features. (Seriously, when did Apple lead in either market share or technology that you were able to predict they would lose it?) I have ALWAYS said that neither market share or “technological leads” are highly relevant to Apple’s potential for success.

    1. >Yes, I predicted it would take more than a year to see any substantial effect.

      Admittedly that puts you ahead of the many fanboys who were screaming INSTANT DOOM at me. A lot of people in the trade press agreed with them.

      >No, because there is not a trend away from skins.

      Incorrect. The really obtrusive brand skins like the Sprint/NASCAR tie-in are all dead (I explained the economics of this last year). Later, the carriers figured out that users don’t like their skins either and began thinning them. One – HTC, I think – actually announced it was retiring the branding elements of its skin. At least some ICS phones that aren’t developer models won’t be skinnned at all.

      I think your last paragraph is extremely evasive.

  51. > it’s incredible to me that anybody believes a single word that comes out
    > of CEO X’s mouth. His ‘the hacker way’ bullshit is beyond belief considering
    > his track record.

    The only thing that I can ever think about when I look at Zuckerburg’s mouth is that I want to remove that smart-ass grin from it using a baseball bat.

  52. First we have this:

    > I’m not seeing it.

    Then this.

    > Now, do yourself a favor and shut up until your record is not embarrassing by comparison.

    Oh man. Haven’t you two ever watched a pair of competent politicians face off? Each one is capable of cherry picking his record to to play to his chosen constituencies, as well as cherry picking his opponent’s record in the same way. Most of them don’t even think they are being dishonest when they do the former. They are convinced they represented their constituents well.

    Neither one of you can possibly win this duel of remembered predictions. You are both too competent.

    I’ve have excellent success with this repeated prediction: Wait for the other shoe to drop.

    Yours,
    Tom

  53. @Tim F:

    > the likely true reason for linear growth [is that] smartphone growth is constrained by production capacity

    This could possibly be true globally; it’s very unlikely that it’s true domestically. To me, domestic smartphone growth seems more stunted by carrier policy than anything else:

    For contract plans, if you have a smartphone, they make you get a (not cheap) data plan, even if you don’t want it and are willing to pay up front.

    For prepaid, they have a limited handset selection (e.g. no iPhone or Galaxy) and you aren’t allowed to BYOD.

  54. @Patrick Maupin:
    “To me, domestic smartphone growth seems more stunted by carrier policy than anything else. For contract plans, if you have a smartphone, they make you get a (not cheap) data plan, even if you don’t want it and are willing to pay up front. For prepaid, they have a limited handset selection (e.g. no iPhone or Galaxy) and you aren’t allowed to BYOD.”

    Yes, and as more and more of the relatively price-insensitive customers acquire smartphones, this factor will become more and more important. Getting that last chunk of laggards into the smartphone world will absolutely require economics that look more like the pre-paid dumbphone market (cf. Tracphone). At that point, I expect Apple will be getting 0% additional, incremental share (which is not to say that it necessarily won’t be profitable) and Android close to 100%, because all of the remaining potential customers will be highly price-sensitive.

    A lot of the controversy in these smartphone threads will continue until RIM is sucked dry, which can’t be that far away. The question is not when they will go to zero, but when the company will fold due to insufficient profitability (in other words, expect a discontinuity, not a straight-line extrapolation to zero.) There is so little share of the WP7 and anyone else that at that point, we will be looking at a straight iPhone/Android battle. Then, and only then, will we have a near zero-sum gain where Android’s gain is Apple’s loss, and vice versa.

    That’s the point I plan to break out the popcorn…

  55. watch out with the ‘tool in my hand’ question. Facebook the website definitely behaves as a firehose, but the answer to “can i write my own client” is a fairly solid ‘yes.’

  56. Oh snap!

    http://www.reuters.com/article/2012/02/03/idUS201473221420120203

    Semantic hashing fail.

    In the past I’ve mentioned that the meaning of the word “hacker” has undergone irreversible shift, such that even geek outlets use the term to refer to a malicious attacker rather than a builder of cool stuff. We (by “we” I mean, roughly, “people who use ‘hacker’ in the canonical sense”) know what we’re referring to when we use that word amongst ourselves, but now to the great wide world out there it only has one meaning which is something skeery.

    We’re gonna have a bitch of a time getting that word back. Most of us have just let it go. If anyone can stake a claim on its meaning in the wider culture it’s someone like Zuckerdouche, who’s kinda like one of us but FAMOUS! No, seriously, he had a movie made about him and everything, that’s why he’s an authority and a representative for our community. (Whuffie amongst us works differently than whuffie amongst people who follow the Kardashians.)

    Even so, he’s got a tough row to hoe when the media drop turds like this on him.

  57. > I think your last paragraph is extremely evasive.

    Pot, Kettle, Black.

    > At least some ICS phones that aren’t developer models won’t be skinnned at all.

    Possibly true (some minor Chinese or Korean ODM could just install the Holo theme and be done), but none from HTC, Samsung, Motorola, LG, etc. will ship sans skins.

    What is true is that Google has announced that all ICS devices with Android Market installed must have the stock Android 4.0 “Holo” (Nexus) theme/skin/layout installed, so the owner/user can always select it. http://android-developers.blogspot.com/2012/01/holo-everywhere.html

    This basically means that the look and feel of 3rd-party Android skins, such as Samsung’s TouchWiz, HTC’s Sense UI, and others won’t integrate as deep into 3rd-party apps unless developers design them to do so. Which, of course, won’t happen, so now Android will look even less consistent.

    It may drive HTC to give up Sense, Moto to give up Blur, and Samsung to give up TouchWiz, but that will take some time.

  58. esr:
    > I can be surprisingly polite to people who aren’t overbearing asses.

    Provided that they lose the argument and you win it.

    You have not been polite to me, and I don’t think I have been an overbearing ass. You called me a racist and a Nazi.

    These days, when someone loses an argument, he cannot call the winner a “mother fucking bastard”, because that sounds kind of archaic, old fashioned, and anachronistic. Such words are no longer hurtful, nor do they have power. Cursing by genitals is like cursing by “God’s Wounds”

    And he cannot call the winner “gay”, because even though the word is hurtful and has power, we are all required to pretend it is not hurtful and does not have power.

    So the loser calls the winner a “racist”.

  59. @James A Donald:
    I thought you actually were a racist. I mean you do think that the races are statistically different from each other and that its morally acceptable to statically discriminate between them in hiring and such? I mean, even if you are correct (I don’t think you are) you would still be a racist under the strict definition of the term.

  60. Not even Google can fit this bill and for good reason: if it is for free, you are not the customer, your attention and data is the product for other customers. But who likes to sell an autonomous, wriggling product, and more importantly: who can? I do think following the hacker way, while providing service free of charge, and still being profitable, is extremely difficult and borders on the impossible.

    Of course hackers would rather pay some monthly fee to use a service that follows the hacker way. The only problem is that only hackers would.

    Well actually… that’s a good idea. A for-fee premium membership following the hacker way and a free basic membership not following? Only problem is in that case it becomes obvious this is a privilege, not a right, not even morally or idealistically, let alone legally.

    1. >I do think following the hacker way, while providing service free of charge, and still being profitable, is extremely difficult and borders on the impossible.

      Supposing you are right, so what?

      Hacker values are what they are. If it’s impossible for a corporation to act in accordance with them, then everybody involved should simply accept that and Corporation X shouldn’t pretend to be doing something they can’t do.

      Whether we actually live in that kind of universe is an interesting but separate question.

  61. @Tim F.
    “Yes, I find it utterly irrational to predict a change in the dynamics of the smartphone market based on the arbitrary figure of 50% US market share of smartphones.”

    Numerology is a strong force in marketing: There is a reason they market everything at $…99.99, including cars (at 9,995 euro, really).

    When Android officially breaks 50% marketshare in the USA, they will be the mainstream platform, and all the others will be alternative platforms for special needs (the old, flashy, ignorant consumers). At that point many people will simply buy an Android phone because everybody buys one.

    And you get that effect at 51% and not at 49%. And it does not matter whether the real numbers are 40% or 60%, this is all about public perception. I expect Apple and MS to frantically redefine the market to keep Android from attaining >50% marketshare.

    Furthermore, it is rather childish to try to shame a prophet for missing the mark by a quarter out of two years. It is the mark of a bad loser.

    1. >Furthermore, it is rather childish to try to shame a prophet for missing the mark by a quarter out of two years. It is the mark of a bad loser.

      In the interests of scrupulous fairness, I’m actually going to semi-disagree with this. There’s a principled case for “shaming” me about getting the timing wrong if you have a rational argument that the error is fundamental rather than superficial – that is, my causal model is wildly wrong, rather than simply having some play in the joints or failing to account for something exogenous and unexpected that nobody else saw coming either.

      Now, in the case at hand, nobody here has actually made such an argument. But in principle someone could. Here are some examples of how principled attacks might begin:

      Eric, your premise that Android collects large market advantages because multiple vendors are able to offer devices embodying a wider range of bets about what people actually want is wrong because…

      Eric, your premise that the presence of multiple systems integrators in the Android market increases supplier willingness to invest in capacity rampups (thus lowering prices relative to an integrator monopsony) is wrong because…

      Eric, your premise that the smartphone market has a stromg winner-take-all dynamic due to every consumer’s desire to not get stuck in an ecosystem other than the eventual winner’s is wrong because…

      The underlying reason I frequently sound contemptuous of the Apple fanboys on here is that they don’t engage my causal model on this level. Instead it’s “Worship teh Jobs awesomness!” or “ZOMG Apple record quarter!” ranting as though that’s even relevant. Makes it hard not to dismiss them as idiots.

    2. >Numerology is a strong force in marketing: […] When Android officially breaks 50% marketshare in the USA […] many people will simply buy an Android phone because everybody buys one.

      The argument the above is excerpted from is correct in every detail, and obvious not only from the incentives at play but from the history of several different prior technology markets. So when I hear someone say that my interest in the 50% threshold is “irrational”, my reaction is “you must have severe brain damage”.

  62. Looked at eric’s comscore page
    http://www.catb.org/esr/comscore/

    The whole conundrum about the marketshare numbers is caused by an “end-of-year” 6.5M spike in smartphone sales. Everybody but MS profited, but Apple captured an extraordinary slice of the pie.

    I think the game changed and the conversion of feature/dumb phones to smartphones will take off like wildfire this year. The war entered a new phase.

  63. “end-of-year” 6.5M spike in smartphone sales.

    Should be

    “end-of-year” 6.5M spike in new smartphone users

    There seems to have been an unexpected spike/increase in demand for smartphones at the end of 2011.

    The availability of new and shiny models on offer at the right time might have determined “marketshare” more than the actual platforms (e.g., iPhone 4S). That might be one reason MS fared so badly: The Lumia was too little too late to benefit.

    1. >The availability of new and shiny models on offer at the right time might have determined “marketshare” more than the actual platforms (e.g., iPhone 4S).

      This is one of the theories I’ve been contemplating. I hadn’t written about it because I have not figured out what testable prediction it could lead to.

  64. Eric’s discussion of “bold humility” (slight paraphrase) amply repays the time invested in reading this comment thread. Another anecdotal example for the advantages of attaching comments to threads: they do bring out points that weblogs respnding to other weblogs rarely do.

    Cathy wrote: “until RIM is sucked dry, which can’t be that far away.”

    Maybe; that seems to be the opinion most often voiced. At an executive level, RIM is in a disastrous state and demand for their high-end products is as lame as they are. But demand for their workhorse products holds up, especially outside North America and there is still considerable loyalty to the brand. I think the RIM drama can go on for quite some time.

    So, was company X in {RIM, Apple}, or did Larry pull this comment thread off course? I used the CUPS website just yesterday: that’s an Apple website, right, like llvm.org? Pretty hacker friendly, I thought.

    1. >But demand for their workhorse products holds up, especially outside North America and there is still considerable loyalty to the brand.

      What you say is true, but I don’t think it’s going to save RIM. One reason is that the company seems to be out of ideas. They made a huge jump in a disasterously costly direction by going all-in on their own proprietary OS rather than becoming an Android OEM, and their last couple of major product introductions have been design botches at the point-and-laugh level.

      I think the December uptick in RIM’s userbase numbers tells us something interesting about the market, though. I’ll blog about that shortly.

  65. @esr
    “I hadn’t written about it because I have not figured out what testable prediction it could lead to.”

    A “backward” prediction would be that new and shiny offers available in the USA in around October (I think), would have taken the majority share of the spike in sales later on.

    Especially, when the sales follow the new models (press/advertisement coverage?) more than the platforms.

    However, I do not think that avenue would be worth the effort. A better way is to plot sales against the number of (new) models on offer.

  66. “Eric, your premise that the smartphone market has a stromg winner-take-all dynamic due to every consumer’s desire to not get stuck in an ecosystem other than the eventual winner’s is wrong because…”

    Something like…

    The key network effect for a cell phone is the ability to communicate with other telephones. Texting and voice communications are #1 and #2, in some order. Email is #3. The biggest reason Eric’s wrong about network effects is because he’s pretending OS matters to cell phones as much as it mattered to PCs in the 80s and 90s.

    Two bonus questions there — why did Microsoft’s agreement to release Office on Macs save Apple’s hide, and why was RIM’s secure text messaging so important? The answer to the second question explains both the Skype acquisition and iMessage. One reason why I know Eric isn’t a good analyst in this field was his failure to notice the significance of iMessage; he didn’t even acknowledge what Apple was trying to do. Time will tell if Apple succeeded, of course.

    That was two posts ago, but oddly enough you ignored it. You did find time to engage with a comment Larry made two comments afterwards.

    1. >The biggest reason Eric’s wrong about network effects is because he’s pretending OS matters to cell phones as much as it mattered to PCs in the 80s and 90s.

      I’m not “pretending” anything. Don’t be unnecessarily offensive, it does nothing for your argument.

      Ah, how times change. I remember when every Appledolator’s favorite reason #1 or #2 the Android would never get traction was positive network effects from the app market. I guess the new Party line is that the app market has no gravitational effects. Noted.

  67. Dear me, it’s plenty hard to filter through all the comments and find the ones _relevant to this post_.

    In any case:

    >>”On the other hand, maybe we should pick our battles.”

    I believe we’re in a difficult spot, culturally, for the hacker identity. The incorrect use of the word ‘hacker’ has been left to marinate for far too long, and we might just be too late to ‘save’ the meaning of the term.

    No, no.
    I think we need to choose a different strategy to make people know what _we_ want the word hacker to mean.
    I will dwell on this, and maybe write a blog post on the subject. @eric: would it be acceptable to link to my post (if I write it and don’t just comment) here? seeing as it’d be on topic?

  68. @Bryant:
    “Texting and voice communications are #1 and #2, in some order. Email is #3. The biggest reason Eric’s wrong about network effects is because he’s pretending OS matters to cell phones as much as it mattered to PCs in the 80s and 90s.”

    Good points. You might be right.

    Take PCs today. #1 Internet, #2 email, #3 watching movies, #4 writing stuff, #5 spreadsheets. Arguably, #1-3 are platform independent. #4&5 can be handled very well by Libre/OpenOffice for 99% of the people. This breaks down when games and employers are put into consideration. These two favor Windows 10:1. Which gets down to Apple’s market share on PCs.

    Gaming will be a big differentiator on mobile OS’. Everything else is available on every platform, from social networks to banking to messaging and video calling. But getting a large portfolio of games will be a problem for “minority” platforms.

  69. @Patrick
    >For prepaid, they have a limited handset selection (e.g. no iPhone or Galaxy) and you aren’t allowed to BYOD.

    This is not true. I recently switched to a prepaid plan BECAUSE I was able to BYOD. While part of your statement is true (I went to the iPhone-less major carrier) I think the BYOD part is more important. People just don’t know about it- I had to do careful research online to discover that the carrier would in fact activate *any* compatible device, they just never said so anywhere. Possibly because they want to sell a new device, but I think they’re shooting themselves in the foot. They should be shouting from the rooftops “BYOD party!”

    But unfortunately this carrier is pretty good at self-foot-shooting. Shame.

  70. @Bryant:
    “Texting and voice communications are #1 and #2, in some order. Email is #3. The biggest reason Eric’s wrong about network effects is because he’s pretending OS matters to cell phones as much as it mattered to PCs in the 80s and 90s.”

    Now the other side: Production.

    What can a mobile handset producer do when she wants to launch a new phone:
    1 Make an iPhone? Nope
    2 Make a Blackberry? Nope
    3 Make an Android phone? Yes
    4 Make a Windows phone? Yes

    So, on the production side OS really does matter.

    For instance, Nokia could not produce iPhones nor Blackberries. So Nokia had to chose between rolling their own (MeeGo), getting the Free option Android (free as in Beer and Speech, MS would not have bothered them with patents), and WP7. They went into bed with the one paying the largest bribe. And that was also the one most desperate. And they now find out exactly why MS were so desperate.

  71. @Cathy:

    You are right (and we have discussed previously) that most remaining dumbphone customers are price-sensitive. I wouldn’t go so far as to say all of them, at least for some level of price-sensitivity. Take me, for example. I’m somewhat price-sensitive on monthly services. I pay way too much for cable with static IPs, so if I really want something, not so much. But cell is a different story. I pay $15 every 90 days for voice at the moment. I’d be willing to pay $700 out of pocket for the right phone, but not $50/month to go with it. I believe there are others like me, so Apple could pick up some of the customers who don’t want large monthly bills…

  72. @Greg:

    Good to know. Thanks. Which particular device did you bring to T-Mobile? Did you use their $30 new-subscriber-only monthly 4G?

  73. @Patrick

    My pleasure. :)

    I bought a G2. I’m very happy with it. I wanted one really badly when it was new, a year later it’s still competitive, which is amazing. I was really worried, I had to find forum posts from people who had experience with Tmobile prepaid plans. TMobile sells activation kits on their website, for ~$5 you get a SIM and an activation code. Lets you activate anything.

    And yes I’m on the $30 monthly 4G. I have yet to go over the voice minutes, but I am averaging 3GB/month data usage. I figure I’m getting my money’s worth. That plan should be a magnet for price sensitive and/or cautious late adopters.

    1. >I’m very happy with [the T-Mobile G-2]. I wanted one really badly when it was new, a year later it’s still competitive, which is amazing.

      Seconded on both counts. It’s what I carry.

  74. @Bryant:

    I think there is some truth to what you say about what customers want to do with their phones. But Angry Birds seems to be high on the list as well, and as Eric already pointed out several columns ago, the smartphone is the “eater of gadgets.” Any time an individual can carry one fewer devices, it’s a win. The devices vary by individual (MP3 player, GPS, flashlight), but anything that can replace them all will do well.

    In any case, as Eric points out in a comment on this post, Apple’s network effects were supposedly going to doom Android to being stillborn. Winter starts to make the case about production, but doesn’t really carry through. There are huge production/distribution ecosystem network effects that allow good phones to be delivered cheaply. (This is part of why I’m not sure Nokia is down for the count yet — they still have a little time to recover.)

    But while we’re discussing ignored comments from two posts ago (rather than several months ago on the “eater of gadgets” post), I already pointed out that “Another [network effect] is more users = more places to turn for answers to questions = more users.” This is not trivial, as any of us who do tech support for friends and relatives should immediately intuit.

  75. “The argument the above is excerpted from is correct in every detail, and obvious not only from the incentives at play but from the history of several different prior technology markets.”

    I do not find the incentive to change to the 50+% team obvious at all (iOS will still present a larger/equal app ecosystem, larger/equal developer base, larger ecosystem of peripheral devices, userbase of 100+ million, numerous network effects from content and service offerings, network effects across form factors and Mac OS X (when relevant), 90% device satisfaction, 75% of the market’s profits — an update from my conservative 60% figure earlier, released yesterday by IDC and comaprable to Horace Dedieu’s analysis)

    However, what are these examples of “several different prior technology markets” where achieving 50% or greater market share was the direct cause of smaller but entrenched competitors to shed marketshare? I ask honestly, in hope that in examining these examples we can learn more. I can’t think of any, but I certainly allow for the possibility that my memory is failing.

  76. Also, yes, I find this fundamentally irrational and unsupported: “When Android officially breaks 50% marketshare in the USA, they will be the mainstream platform, and all the others will be alternative platforms for special needs (the old, flashy, ignorant consumers). At that point many people will simply buy an Android phone because everybody buys one.”

    Android is ALREADY the mainstream platform; all others are ALREADY alternative platforms. The notion that consumer’s buying decision will fundamentally change to selecting on market share alone, at the expense of all other factors, when transitioning from 47% versus 29% market share to 51% versus 32% market share irrational and unsupported — not at all analogous to the thoroughly examined effects of $x.99 pricing.

  77. @Tim F.
    “Android is ALREADY the mainstream platform; all others are ALREADY alternative platforms.”

    What changes at 50% is that the news outlets will start printing that Android is mainstream.

    Upto now, they are writing about the “Smartphone wars”. When Android crosses that psychological barrier, they will start writing about how Android won the war. The News is all about winners and losers. >50% is the winner. The news outlets will shift attention to the Tablet wars.

    The news outlets are a bunch of morons you say? Who am I to contradict you?

  78. About hacking, the hacker way, and Android:

    Soon You Will Be Running Android On Your PCs
    http://www.muktware.com/news/3275/soon-you-will-be-running-android-your-pcs

    I had a discussion with Linus Torvalds about Android, he told me, “Every time there is a fork, and I think actually forks are good things, it means somebody sees a need and a technical reason to do something different from the standard kernel. But most forks are failures. They find that the things they needed were not actually worth doing and as a result most forks die. Some forks are successful — Google with Android has really been very successful. Now, we are talking about how to say ‘Hey you were right, we were wrong, you were successful doing something right’. We are talking about how to merge the good parts.”

  79. Larry Y:
    Amazing complete divergence from the topic of the post… And you did it w/o being tagged “off-topic”… well played.

    Jeff R:
    I think the problem with letting go of the word (“undergone an irreversible shift”) hacker is this: Where does that leave us then? The term “linguistic slide” is often used to describe this phenomenon.

    Activists on either side of the table and up and down understand the importance of preserving the terms that we use to describe our tribe when set against the backdrop of making the culture-at-large understand “who we are”. Softening terms (Black->”Aftican American”-> ??? … “Indian”–>”Native American”–>??) just leave us with more vague ways of describing ourselves, and a weaker affiliation with the tribe. It is why many rappers (or is it “rap artists”?) took back the term that comes in the list just before “Black”. And no, I don’t have any problem with using the PC terminology, but I don’t want it forced on me anymore than I want it forced on someone else (namely the target of the label to begin with).

    It becomes a question of what we choose to call ourselves, and saying “Fuck no, I’m not a {whatever watered down computer programmer term you wish to insert}! I’m a hacker. I don’t break into your systems to fuck them up, I build things. ”

    ~Dave

  80. “What changes at 50% is that the news outlets will start printing that Android is mainstream. ”

    I don’t find any absence of media coverage declaring Android the majority and/or mainstream smartphone OS NOW.

  81. @Tim F.
    “I don’t find any absence of media coverage declaring Android the majority and/or mainstream smartphone OS NOW.”

    Indeed, but there comes a point when that is not even declared by many, but simply assumed by ALL media coverage.

  82. “Indeed, but there comes a point when that is not even declared by many, but simply assumed by ALL media coverage.”

    So… they are already saying it but this psychological effect that negates all other purchasing factors will kick in when they STOP saying it? Which media outlet isn’t “assuming” 47 is a bigger number than 29?

  83. Well, Tim F. you keep validating my notions of your competence, and you are working with esr and Winter to elevate discussion here above it’s normally high level. You seem to be arguing that Apple is successfully aiming at the BMW / Mercedes segment of the market, so to speak. This harmonizes with what esr has said for a long time. What I am seeing is that you and esr agree more than you disagree.

    Yours,
    Tom

  84. While I was briefly enthusiastic about the word “hacker”, I have returned to the older word “tinker” to describe myself. Fortunately, “tinker” has mostly shed the negative connotations (they were variations on “thief”) it had in the 19th century and before. Besides, I am at least as much interested in actual, physical tools as software, and “hacker” never did catch on very well there. Nor am I the only one, see Freedom to Tinker
    … is your freedom to understand, discuss, repair, and modify the technological devices you own.
    for example.

  85. @Tom DeGisi: I disagree with your assessment. No, I do not think Apple is “aiming” for BMW/Mercedes segment. If I recall correctly, both have about 3% market share in the US for example: much more closely analogous to Mac or desktop Linux. Heck, even as % of ALL mobile phones, iPhone has already far surpassed such levels. I would say that I’m arguing for a market that can support equally strong platforms as measured by market share and several other factors. I do not see anything, presumably, “niche” about 30% market share.

    Yes, I HAVE argued that the iPhone could still lose significant share and still remain a viable and highly successful competitor. I have not argued that that is going to happen or is an aspiration of Apple’s however.

    I am unsure if I have stated it here before, but I will now: I will wager that iPhone will not risk significant share loss (which I interpret as, closer to the the 5% of total market share on an annualized basis — to account for quarterly shifts due to product transtions) of the smartphone market UNTIL we reach a true zero sum game (i.e. 80-100% of all mobile phones are smartphones worldwide, and there is no longer “easy” growth in the market). At which point, it only becomes a risk, not an inevitability. Prior to true zero sum game status, Apple is more likely to continue small but measured share gains.

  86. Eric, your premise that Android collects large market advantages because multiple vendors are able to offer devices embodying a wider range of bets about what people actually want is wrong because…”

    Eric, your premise that the presence of multiple systems integrators in the Android market increases supplier willingness to invest in capacity rampups (thus lowering prices relative to an integrator monopsony) is wrong because…

    Ah, you’re wrong on the second point because of the first point. Multiple vendors (or even a single vendor) with a wide range of different products does not enjoy the same economies of scale that production of a smaller cohesive product line. This is why Apple has a significant advantage in supply chain economics. They choose to use that advantage to maintain both high ASPs and high margins. That’s probably wise given their profit in comparison to competitors.

    Sure, Android phones are similar but the panels differ in size, technology, price, quality, etc. The CPU and GPUs come from multiple vendors, etc.

    Apple buys more silicon than anyone else and it’s largely homogeneous. And they’re willing to dump billions into purchases to guarantee supply availability and priority delivery. The capacity rampups seen in the smartphone industry appear to me to be largely in Apple’s favor. They’re paying for tooling. They’re paying for bulk purchases. They’re getting the best prices and the first delivery of many components.

    The first point is potentially wrong if it turns out that only Samsung and random Chinese companies (evidently not including HTC given their guidance) can make money off Android. Then the rapid market share gains could reverse in favor of something else companies can differentiate enough to make money off of.

    Profit does matter if you are trying to model OS platform adoption. If Samsung is wolfing down most of the Android profits then the ones losing are going to do something else besides hemorrhage money. This can be things like lock into their own ecosystem like Sony Xperia with exclusives like FIFA to full up forks like the Kindle Fire to jumping ship to WP.

    This is where I differ from Winter. He believes that Nokia is doomed by going WP. I believe that Nokia was doomed going either Meego or Android. There’s no indication that they would have executed an Android approach any faster or better than WP and my impression is Samsung would have eaten their lunch anyway. By getting preferential treatment from MS for going “all in” they have some kind of advantage vis a vis other WP device manufacturers. The reality is that they’re probably just doomed but less doomed than going Android.

    Eric, your premise that the smartphone market has a stromg winner-take-all dynamic due to every consumer’s desire to not get stuck in an ecosystem other than the eventual winner’s is wrong because…

    This is potentially wrong because you are overestimating the number of apps and services the average consumer actually has or uses. Tech geeks like us have a lot more lock in than Grandma. There’s also the potential that the Android ecosystem is less sticky than iOS because there are many more ad driven free apps in that ecosystem. I have a good amount of money invested in iOS apps. I’m not as likely to leave for the WP or Android platforms because of the $$ value but honestly if the user experience was compellingly better it’s not that big a deal.

    Most folks are used to buying a new phone every 2 years. And I would imagine many first time smartphone users are still largely using them as a feature phone with a browser, netflix and a couple free games.

    Eh, I believe that the Android family will have the largest share just like the Symbian family did. But I think you’re going to have the same kinds of split between UIQ and Series 60/80/90 which will make app development annoying. Even more annoying than today and by design by the device makers. Some are going to deliberately diverge from Google standards to implement a competitive advantage vis a vis other Android makers.

  87. I am unsure if I have stated it here before, but I will now: I will wager that iPhone will not risk significant share loss (which I interpret as, closer to the the 5% of total market share on an annualized basis — to account for quarterly shifts due to product transtions) of the smartphone market UNTIL we reach a true zero sum game (i.e. 80-100% of all mobile phones are smartphones worldwide, and there is no longer “easy” growth in the market). At which point, it only becomes a risk, not an inevitability. Prior to true zero sum game status, Apple is more likely to continue small but measured share gains.

    Eric likes to say that Android has been eating iOS’s lunch. Except that it hasn’t. It’s been eating BB, WinMo and Symbian lunch the whole time and there’s more share for them to lose.

    In a zero sum game I doubt Android will make much of a dent in iOS market share unless Apple becomes unable to ramp production further. If something happens to A6 or A7 production I can see iPhone market share tanking. I doubt Tim Cook would let that happen though.

    1. >Eric likes to say that Android has been eating iOS’s lunch. Except that it hasn’t. It’s been eating BB, WinMo and Symbian lunch the whole time and there’s more share for them to lose.

      I guess you haven’t been paying attention. Both I and Patrick Maupin have been saying the latter, repeatedly. I thought early in 4Q2011 that it looked like we finally had moved into the beginning of zero-sum competition between Android and Apple, but I’m more doubtful of that now for reasons I will explain in an upcoming post.

  88. “There seems to have been an unexpected spike/increase in demand for smartphones at the end of 2011. The availability of new and shiny models on offer at the right time might have determined “marketshare” more than the actual platforms (e.g., iPhone 4S).”

    I don’t think you can ignore the fact that the economy started to recover in 2011. This has made a significant chunk of consumers more willing to spend money on discretionary purchases.

    The joke at my workplace is that we live in “the bubble”, i.e. everyone in the room makes $100K+ and has secure jobs, and demand for our skills is high enough that there has been a lot of turnover lately as people leave for better opportunities and promotions. Despite the high level of unemployment and housing issues, the reality is that a good chunk of the upper middle class is doing very well. I think this is where the Apple sales are coming from; iPhones are more common than Android in my “bubble”.

    At the same time, there are more and more mobile-only households that use phones as their only voice and Internet access. So even those of modest incomes may be shifting existing dollars toward better mobile hardware and software.

  89. @Eric
    There are still 1.5B feature and 3B dumb phones out their waiting to be converted. There still is plenty of room.

  90. “I thought early in 4Q2011 that it looked like we finally had moved into the beginning of zero-sum competition between Android and Apple, but I’m more doubtful of that now for reasons I will explain in an upcoming post.”

    I’m not sure most of us need it explained that a market that has only saturated roughly 30% of its market potential is no where near a zero sum game.

    1. >I’m not sure most of us need it explained that a market that has only saturated roughly 30% of its market potential is no where near a zero sum game.

      It’s not that simple. There isn’t any “the” market in this sense, but a couple of different ones partly walled off from each other by differing levels of consumer price sensitivity and requirements. Could well be that the market consisting of “people willing to pay the premium for a smartphone” saturates, leading to zero-sum Apple-vs.-Android competition, while the unconverted dumbphone market is still huge.

      Case in point: my wife Cathy, who despite living with me and my Android is still by choice using a $30 dumbphone. She doesn’t want the extra bulk. Not all premiums are in the purchase price.

  91. “There are still 1.5B feature and 3B dumb phones out their waiting to be converted. There still is plenty of room.”

    In units, yes. But this is irrelevant to smartphone share.

  92. “I’m not sure most of us need it explained that a market that has only saturated roughly 30% of its market potential is no where near a zero sum game.”

    Again, smartphone share is a zero sum game by definition. And when it gets down to only two players with meaningful share, it’s a straight fight for share between them. It’s perfectly possible to grow sales units and profits while losing share.

    For someone reason, this keeps getting forgotten in this debate, and people end up talking past each other.

  93. “Again, smartphone share is a zero sum game by definition.”

    No, by definition it is NOT: any one player can continue to add share without stealing it from another player because of the massive transition from dumb/feature phones to smartphones.

    “And when it gets down to only two players with meaningful share, it’s a straight fight for share between them.”

    No, again, it is not a zero sum game no matter how many players are significant or not.

    “It’s perfectly possible to grow sales units and profits while losing share.”

    Absolitely, have never lost sight of this whatsoever.

  94. “It’s not that simple.” Sure, but it is certainly as simple as zero sum game has been reached when Android has 50% market share of US smartphones though. It is certainly as simple as we are no where near close to zero sum game status.

    And, generally, while I agree that some of the world’s market will be foreclosed to partipication in smartphone usage, I disagree that it’s appropriate to claim “saturation” any time in the 30-75% percent range or as long as some providers are still experiencing greater than 100+% growth.

  95. Another off topic question: what do you think of the lightsquared vs GPS dogfight, either laying aside the crony capitalism issues or, not…

    1. >Another off topic question: what do you think of the lightsquared vs GPS dogfight, either laying aside the crony capitalism issues or, not…

      [Insert blistering profanity here]

      Yeah, “crony capitalism” is a good description of the LightSquared clusterfuck. If this were a Republican administration the press would be screaming about corruption and influence peddling, and rightly so.

  96. EDIT: “but it is certainly as simple as zero sum game has [NOT] been reached when Android has 50% market share of US smartphones though.” But I suppose that should be obviously my view at this point.

  97. @Greg:

    > And yes I’m on the $30 monthly 4G.

    I was eyeballing that. It was a bit worrying that if you inadvertently let it expire, then you can’t get back on that plan. Does their automatic top-up work OK? Also, I didn’t see how much minutes cost if you go over 100.

  98. @Tim F.:

    > No, by definition it is NOT: any one player can continue to add share without stealing it from another player.

    You seem to be missing Cathy’s point here. Total marketshare always equals 100%, making any marketshare gains or losses zero-sum, even though there is enough untapped market for all players to increase total unit sales.

  99. @william

    You might find this interesting. The Open Access Movement is disorganized; this must not continue. Among other things, he is discussing the OSI and pointing out that Open Access needs an equivalent organization and set of licenses. And particularly that “non-commercial” limitations are not acceptable.

    I would argue that OSI is about the least useful organizational template for Open Access and that Creative Commons is a far better model.

    And one reason is that “non-commercial” limitations fit under the umbrella and there is a well defined range of options from very permissive to much less so that content creators can use. This is how they solved the problem of:

    “Cannot agree on what “open access” means in practice” and “Appears to be composed of factions which while they agree on some things disagree on enough others to make this a serious problem.”

    That OSI has a very narrow definition of open source that is only slightly looser than the FSF is a defect, not a “great achievement”. It ignored a large body of open source that existed in the academic world that was largely tagged as NC. Leading to tiresome arguments about open source vs shared source vs whatever. In the CC world it’s all Commons and the tags tell you pretty much what the intent is.

    Here’s a suggestion: why not be inclusive vs exclusive and not buy into the zealotry of the one true definition of open? It seems to work well for the CC world. If you want CC-BY-NC then use CC-BY-NC. Or even CC-BY-ND if that floats your boat.

    And his assertion that when he finds an Open Source program he knows what he is getting is BS given the amount of misunderstanding of licensing issues even among open source developers much less normal corporate developers. Some of this confusion is by design *cough*FSF*cough*.

    1. >And one reason is that “non-commercial” limitations fit under the umbrella

      NC limitations are a set of time bombs waiting to go off. There’s a reason OSD disallows them, and it’s a brutally practical one: there’s no bright-line definition of what “commercial use” is. In the presence of NC restrictions, reasoning about your rights is just too difficult. In fact, the nascent open-access community has already tripped over this issue – it’s a major reason they have factions.

      Someday one of the CC-NC edge cases is going to blow up in some well-intentioned person’s face. You think the FSF’s virality is bad and confusing? I agree…but CC-NC model is, in terms of chilling effects and potential legal disruption, much, much riskier.

      That this is so has been acknowledged to me by a CC cofounder. I lit into him about the problem, and he stopped me with “You’re right.” As my jaw dropped to the floor, he admitted that CC-NC represents a strategic gamble – they’re hoping a court will pin some kind of definition and defense in place for them before some well-heeled predator exploits the absence of a bright line.

      The open-access folks are struggling with the realization that they need to avoid this. And they’ll do it, I think, eventually by writing something pretty much isomorphic to the OSD. I’ve offered to help.

  100. I see…

    Anyway, when I first caught wind of this, the crony capitalism angle wasn’t yet visible; and I still wondered about the technical aspects. Particularly without an FCC to moderate the competing interests of the GPS vendors and consumers vs the owners of the spectrum.

    (Side note : WTF was the Lightsquared people thinking, that any amount of crony capitalism would let them override the combined forces of the .mil and the FAA once the interference issue came up)

  101. @esr

    I guess you haven’t been paying attention.

    Actually I have been paying attention which is why I made that statement. Do you REALLY want me to go back and find all the places you’ve stated that phrase. I can but it would tedious and of not much value other than to count coup.

  102. “You seem to be missing Cathy’s point here. Total marketshare always equals 100%”

    No, I’m disputing a complete and utter distortion of the definition of zero sum game. This is the equivalent of saying: anything expressed as a percentage represents a zero sum game. This is a false and useless statement. If I use the same numbers, not expressed as a percentage, it is quite clearly not a zero sum game.

    “…even though there is enough untapped market for all players to increase total unit sales.”

    Which makes it not a zero sum game.

    I don’t think it gets any simpler than this: “If the total gains of the participants are added up, and the total losses are subtracted, they will sum to zero. In contrast, non-zero-sum describes a situation in which the interacting parties’ aggregate gains and losses are either less than or more than zero.”

  103. @Tim. F:

    You are mostly agreeing with Cathy, while acting like you violently disagree with her.

    My take is slightly different — I think it’s close to a zero sum game, but in handsets, not in smartphones. (Handset shipments expected to grow around 2% this year.) But don’t take my word for it — ask Nokia or LG or HTC.

  104. Ian, Argent, its hard to figure out where the stupidity in the LightSquared situation most lies. Whether it was that the LightSquared management thought that a legalistic approach to the spectrum allocation would prevail over the practical (their argument is that the GPS spectrum useage being broader than the actual allocated GPS spectrum is too-bad for the device manufacturers and the installed base) or whether they just thought that they could buy off the administration / FCC.

    Combine the stupid letter that the LightSquared CEO sent to Senator Grassley and its probably arrogance and stupidity in equal proportions.

  105. @esr, I think that your point about the “non-commercial” having no bright-line definition cannot be emphasized enough. Analogous to the copyright “Fair Use” doctrine utter uselessness.

  106. “You are mostly agreeing with Cathy, while acting like you violently disagree with her.”

    No, I am not. I understand that a market expressed as a percentage will always equal 100%. I disagree profoundly and fundamentally with the bizarre notion that this, thusly, can be called a zero sum game.

    If Cathy wants to state that smartphone market share summed across manufacturers or platforms, including others, always equals 100% — i.e., merely restate the definition of a percentage, she is free to do so. However, if she wants to call that a zero sum game, I will certainly disagree with her.

    “My take is slightly different — I think it’s close to a zero sum game, but in handsets, not in smartphones.”

    I do not think that is a “slight” difference, and I fully agree: the market for all mobile handsets is, for all intents and purposes (excepting for the possibility of slight market growth and/or even contraction as virtually negligible), a zero sum game. This is not true of the smartphone market.

  107. Unfortunately, Open Access means way different things to different people:

    http://www.si2.org/?page=69

    Which (btw) is not really all that open to access.

    When it comes to academic open access, I agree that non-commercial only is a non-starter. I guess I don’t really understand the issues though. Since scholarly articles aren’t functional like code, what’s wrong with “copy this all you want but don’t change it” and letting fair use deal with the normal small extracts?

  108. @Tim F.:

    > I disagree profoundly and fundamentally with the bizarre notion that this, thusly, can be called a zero sum game.

    But the way you expressed your fundamental, profound disagreement was to make the statement that “any one player can continue to add share without stealing it from another player” which seemingly leads to the even more bizarre conclusion that market share doesn’t have to add up to 100%.

  109. @SPQR:

    I disagree that the fair use doctrine is useless. When we hear about it, it is usually edge cases of either somebody pushing too hard and copying too much, or somebody else trying to copyright 25 words. But if you take a step back and look around at all the cases where articles quote verbatim from other articles, I think you have to agree that the system actually works pretty well. This is a feature of copyright that I would not willingly give up (part of the public bargain) and yet have no idea how to make the line sharper, other than repeated applications of court cases.

  110. “But the way you expressed your fundamental, profound disagreement was to make the statement that “any one player can continue to add share without stealing it from another player” which seemingly leads to the even more bizarre conclusion that market share doesn’t have to add up to 100%.”

    No, I see how you can quibble with my usage of the word “share” in this statement, but I do not think it’s not abundantly clear what my point is: A percentage of a whole is not inherently a zero sum game.

  111. Lightsquared was apparently marching to the same plan ever since 2005 (Bush administration) and Obama apparently lost money on lightsquared:

    http://www.usatoday.com/news/elections/2007-03-08-2531343349_x.htm

    And AT&T and other opponents are excellent at using crony capitalism.

    So, I chalk it up to technical naivete, combined with ham-handed attempts at using money to influence the political process. A repeat of the broadband powerline fiasco. There may be a smoking gun that says somebody inside the government was really behaving extra badly, but I have yet to see it. More like the normal “he’s my really good constituent — isn’t there anything you can do for him?” sort of plaintive plea, AFAICT. Not great behavior, but probably not enough pressure to cause a serious malfunction, despite the billions on the line.

  112. @Tim F.:

    > I see how you can quibble with my usage of the word “share” in this statement,

    But, from my perspective, the whole thing is a quibble.

    > A percentage of a whole is not inherently a zero sum game.

    And another quibble. Multiple parties each vying for some percentage of a whole is practically the definition of a zero sum game. I understand your point that (in this case) the market can easily be grown, but there are lots of examples of people using “zero sum” to relate to “market share.” I offer one:

    http://www.fool.com/investing/international/2011/03/08/china-pc-share-you-cant-always-get-what-you-want.aspx

  113. “But, from my perspective, the whole thing is a quibble.”

    No, I think calling market share % the equivalent of a zero sum game a semantic game to remove all value to the meaning of the phrase “zero sum game.” If you are not talking about all future growth being at the expense of a competitor, then it is pointless to use the phrase.

    “Multiple parties each vying for some percentage of a whole is practically the definition of a zero sum game.”

    No, it is when the pie can no longer increase or decrease in size but not in all instances.

    “I understand your point that (in this case) the market can easily be grown, but there are lots of examples of people using “zero sum” to relate to “market share.” I offer one:”

    I don’t accept anything written at the Motley Fool as proof or even demonstration of basic competence.

    Again, removing both the phrasings of market share % or zero sum game:

    Is there massive room for all players to continue to increase their sales/user base? Yes

    Is all growth and/or lose by one competitor at the expense of another competitor at this stage? No.

    These are the salient, non-quibble-able points.

  114. > These are the salient, non-quibble-able points.

    And on those points, I think that you, I, Cathy, esr, and lots of other readers probably agree.

  115. No, I think calling market share % the equivalent of a zero sum game a semantic game to remove all value to the meaning of the phrase “zero sum game.” If you are not talking about all future growth being at the expense of a competitor, then it is pointless to use the phrase.

    I agree that all future growth in share is at the expense of a competitor’s share. Growth in the market does not change that. Not all growth in sales is necessarily at the expense of a competitor’s sales, but that’s a different issue.

    And when your interest is not in being profitable, but in having open standards that can’t be locked out — in short, members of the hacker community — it’s share that matters, not sales.

  116. @Tim F.
    “Is there massive room for all players to continue to increase their sales/user base? Yes
    Is all growth and/or lose by one competitor at the expense of another competitor at this stage? No.
    These are the salient, non-quibble-able points.”

    @Patrick
    “And on those points, I think that you, I, Cathy, esr, and lots of other readers probably agree.”

    Yes, I agree.

  117. @Tim F:
    >No, by definition it is NOT: any one player can continue to add share without stealing it from another player because of the massive transition from dumb/feature phones to smartphones.

    Wrong. Any one player can continue to add share without stealing *user base* from another player, but if Android gains 5% market share, that 5% has to come from some other player(s), regardless of whether it’s a static 100 million user market in which it gains 5 million users when a competitor loses 5 million users, or whether a 100 million user market grew by 10 million users and it captured all of those new users while its competitors gained nobody.

    In either case, Android gains 5% share, and its competitors, between them, lose 5%.

    User base is not a zero sum game.
    Share is.

  118. “And on those points, I think that you, I, Cathy, esr, and lots of other readers probably agree.”

    Not when ESR thinks we’d be at that stage sometime between Sept/Oct, now, and some indeterminant time in the near-term future, no, we don’t.

  119. “Wrong. Any one player can continue to add share without stealing *user base* from another player…”

    I’ve already addressed that, in the rush to respond to numerous comments, I used “share” poorly in that statement.

  120. >“Wrong. Any one player can continue to add share without stealing *user base* from another player…”

    I noticed that. I read through the comments and then posted without realizing how long it had been since I refreshed the page. My bad.

  121. @Tim F.:

    > Not when ESR thinks we’d be at that stage sometime between Sept/Oct, now, and some indeterminant time in the near-term future, no, we don’t.

    And here I thought you deliberately worded things so carefully we could all agree.

    Yes, there is massive room for all players to continue to increase their sales/user base, especially on a global basis. But not all players will equally be able to take advantage of this because they don’t have anything the customers want, and in any case this is a state of affairs that might not last for too long (especially on a domestic basis), because some of the players who do have things the customers want have huge manufacturing capacities.

    No, not all growth and/or loss by one competitor is at the expense of another [smartphone] competitor at this stage. But some of it is, even now, unless you think that Android couldn’t have sold more phones absent Apple last quarter.

  122. Oops. And now I quoted the wrong part of your post. I’m not awake today, it seems:

    I was responding to:

    >I’ve already addressed that, in the rush to respond to numerous comments, I used “share” poorly in that statement.

    not to:

    >“Wrong. Any one player can continue to add share without stealing *user base* from another player…”

  123. @Patrick
    >I was eyeballing that. It was a bit worrying that if you inadvertently let it expire, then you can’t get back on that plan.
    >Does their automatic top-up work OK? Also, I didn’t see how much minutes cost if you go over 100.

    I’m in my 3rd month on the plan. I haven’t used the auto refill option yet, though I’m considering it. I once got distracted and forgot to refill on time and had no service for part of one day, once I refilled I was back online in minutes. I understand that to expire for real you have to go 90 days without service. Even then I think you can get back in, you just have to start all over from the beginning with a new activation kit and a new phone #. (Treat it like a completely new line of service.) Voice minute overage is $0.10 per minute which I think is pretty decent, it’s the same per minute rate as Tmobile’s cheapest pay-as-you-go option.

  124. “No, not all growth and/or loss by one competitor is at the expense of another [smartphone] competitor at this stage.”

    And? That still doesn’t make it a zero sum game, it still doesn’t make any sense of ESR claiming 50% share in US smartphones represents the entry to a zero sum game, etc…

    “But some of it is, even now, unless you think that Android couldn’t have sold more phones absent Apple last quarter.”

    I’m certainly not certain that they definitely would have.

    1. >And? That still doesn’t make it a zero sum game, it still doesn’t make any sense of ESR claiming 50% share in US smartphones represents the entry to a zero sum game, etc…

      That is not a claim I have ever made; those are different thresholds. There will be a point at which Android-Apple competition becomes zero-sum. There will be a point at which Android goes over 50%. There’s no reason those have to happen at these same time, and they could happen in either order.

  125. Sorry about all the kind of off topic cross conversation, but it’s at least educational.

    I should add, there are a couple of drawbacks to being on prepaid. One is that prepaid customers don’t benefit from TMobile’s roaming agreements- the coverage you get is their own network and that’s it. Fortunately I’m in a very good coverage area, but that’s not the case for everyone. Another drawback, a little thing but annoying- Tmobile doesn’t let prepaid customers do any call forwarding. So using Google Voice for voicemail won’t work properly, it won’t be fully seamless. (I’m not a big talker, I don’t have Tmobile vmail even configured. I give out my Google Voice # anyway, people can leave a message and it works out the same as if they texted me.)

  126. @Tim F.:

    “No, not all growth and/or loss by one competitor is at the expense of another [smartphone] competitor at this stage.”

    And? That still doesn’t make it a zero sum game, it still doesn’t make any sense of ESR claiming 50% share in US smartphones represents the entry to a zero sum game, etc…

    What do you mean, “and”? I was essentially agreeing with you here: “Is all growth and/or lose [sic] by one competitor at the expense of another competitor at this stage? No.” before adding a caveat. Or wasn’t that you? It’s interesting to see your response to your own sentiment. I think you’re conflating and confusing a few different issues here, but I’ll let others address that if they care to.

    I’m certainly not certain that [Android vendors] definitely would have [sold more last quarter absent the iPhone].

    It’s hard to separate out the very existence of the iPhone (which certainly helped grow the market) from its effects last quarter, but I certainly believe that more Androids would have been sold last quarter if, for example, Apple had stumbled on their 4S release. A look at the comscore data makes that (to me) fairly obvious.

  127. Patrick, originally, I had quoted you in one large snippet first, but then I broke it up to address your second point more directly. So I wasn’t questioning my statement of what a zero sum game is… I was questioning your attempt to add an “and BUT…” A zero sum game + a but = NOT a zero sum game. I apologize if my edit made it sound like I was questioning my own statement; I was questioning the foreseeable “but, but, but…”

  128. Patrick, with respect my friend, I disagree. And I disagree for a couple of reasons. The Fair Use doctrine is a difficult thing to intentionally rely upon. You cannot find bright lines and even where you can find such, you may not be able to practically defend them. I look at things from a different point of view as a practicing attorney, when my clients ask me if such and such is a “fair use”, I have to tell them that I don’t have bright lines and if I did, they may not be able to defend their use against a copyright owner plaintiff. See how many people that Righthaven conned out of money who thought they had a fair use. What finally defeated Righthaven was that a defendant finally got institutional backing to pay for an effective defense.

  129. @SPQR:

    Ahh, I (think I) misunderstood. You’re arguing against reliance on fair use rather than the concept, because copyright law is unforgiving, and for any particular defendant, the odds of prevailing at a reasonable cost suck. So, in retrospect, I agree that this is something that a good license should address.

  130. Like Soylent Green, Facebook is made of people. People who share little quips, changing status, and photos. They’re hooked into a social graph.

    Google, on the other hand, is made of search.

    Both are ripe for disruption in the transition of the mass market to smartphones.

    The question will become, will Google continue to fund the very expensive / low ROI Android as revenues continue to fall? And if Google does reduce focus on Android, how can it tip the market, or, if the market has tipped, how can it prevent being disruption?

    The open source (-ish) nature of Android isn’t going to help.

  131. I’m not “pretending” anything. Don’t be unnecessarily offensive, it does nothing for your argument.

    Ah, how times change. I remember when every Appledolator’s favorite reason #1 or #2 the Android would never get traction was positive network effects from the app market. I guess the new Party line is that the app market has no gravitational effects. Noted.

    None of that in any way actually addresses the point I raised. The rhetorical trick where you classify me as something I’m not, then claim you can ignore my point because people in that group aren’t intelligent? I know that one even if your fans don’t. It is, however, a great way to drive off people who’re interested in thoughtful disagreement… and then you can sneer more about how nobody ever thoughtfully disagrees around here.

    1. >None of that in any way actually addresses the point I raised.

      You denied that network externalities are a significant factor in the smartphone market. I pointed out that this is inconsistent with previous claims by Apple partisans that network externalities from apps would smother Android. You are an Apple partisan. So how is pointing this out non-responsive, again?

      And if you don’t want to elicit rudeness, don’t accuse me of “pretending” to positions. If you’re not prepared to believe that I present my beliefs and my reasoning honestly, to hell with you – I actually do want to “drive you off”.

  132. When I hear about mobile computing, I think about reversing all the hazards introduced by sitting at a desk all day. However, when I see other people using them it seems to be used as a way to bring work/Internet out into non-desk situations. I would like a computer where I can move while I calculate. Something that lets me go for a 2 hour walk to nowhere in particular while I’m programming. Does anything like this exist in the mobile space?

  133. >There seems to have been an unexpected spike/increase in demand for smartphones at the end of 2011. The availability of new and shiny models
    >on offer at the right time might have determined “marketshare” more than the actual platforms (e.g., iPhone 4S).

    There’s also the fact that the holidays are Apple’s go time for almost any and all of their portables. Which leads me to an interesting question, does anyone remember a time before the iPhone and Android when every day people would get cell phones for christmas? Did it happen during the RAZR boom? Cell phones as gifts always struck me as weird when I was selling them. “Here, have a brand new $300 electronic device that you must now pay $70+ per month to use.”

  134. “Facebook is made of people. People who share little quips, changing status, and photos. They’re hooked into a social graph. Google, on the other hand, is made of search. Both are ripe for disruption in the transition of the mass market to smartphones.”

    Yes, I completely agree.

    “The question will become, will Google continue to fund the very expensive / low ROI Android as revenues continue to fall? And if Google does reduce focus on Android, how can it tip the market, or, if the market has tipped, how can it prevent being disruption?”

    Um, Android is the tool that Google has developed to help prevent it from being disrupted. That’s why they made the investment in the first place. It may or may not work, but calling Android “ROI” ignores its strategic nature in helping to shield its key revenue-generating business from an Apple-driven disruption.

  135. And yes, one of the most basic questions is “Does Corporation X publish the source code of its user-facing software in a form that can easily be understood, modified, and reused?” Because if the answer to that question is “no”, it is very unlikely that users will or ever can have the control of their on-line lives that they deserve.

    Can I just say that this sounds exactly like something RMS would say?

    1. >Can I just say that this sounds exactly like something RMS would say?

      Go right ahead. I don’t mind when RMS and I reach similar conclusions, even if the ethical basis of our reasoning is somewhat different.

  136. Doc Merlin Says:
    > I thought you actually were a racist. I mean you do think that the races are statistically different from each other and that its morally acceptable to statically discriminate between them in hiring and such? I mean, even if you are correct (I don’t think you are) you would still be a racist under the strict definition of the term.

    Well if that is the definition of racist, then all sane people are racists, but esr also accused me of believing that Obama is subhuman because of his race, which would seem to imply a more stringent definition of “racist”.

  137. esr,

    The open-access folks are struggling with the realization that they need to avoid this. And they’ll do it, I think, eventually by writing something pretty much isomorphic to the OSD. I’ve offered to help.”

    You may want to seek compatibility with the Definition of Free Cultural Works, if not endorse that def. outright. CC-BY, CC-BY-SA and the “CC-0” license+PD dedication are compliant, and the DFCW seal appears on their “license deed” pages. Wikimedia Commons uses the DFCW as a guideline for acceptably-licensed content.

    1. >You may want to seek compatibility with the Definition of Free Cultural Works, if not endorse that def. outright.

      I’ll endorse that, all right. When I read the preamble I made a mental list of several “gotchas” – places that experience with the Open Source Definition taught me traps and dangerous corner cases await. (Prohibition of NC restrictions was one, but not the only one.) The rest of the DFCW covered them all.

      Obviously, somebody has already read the Open Source Definition and done the right kind of key-transposition so I don’t have to. Excellent!

  138. >> And yes, one of the most basic questions is “Does Corporation X publish the source code of its user-facing software in a form that can
    >> easily be understood, modified, and reused?” Because if the answer to that question is “no”, it is very unlikely that users will or ever can >> have the control of their on-line lives that they deserve.

    > Can I just say that this sounds exactly like something RMS would say?

    Right, I noticed that too and then in the comments I read esr saying that GPL’s “viral” clauses (copyleft) are essentially bad. I knew there is a rift between Free Software advocates and esr but to this day I do not understand why exactly. If you’re an open-source advocate why wouldn’t you want the software released as open-source to remain that way and prevent the situation when somebody just takes the codebase and makes it closed source/proprietary which makes us all lose the benefits it provided us – be it the freedom or technical soundness.

    Eric, is there a kind of summary post I could read to understand your problems with Free Software? As I tried to search your blog but there are only bits and pieces all over the place it seems. If there is a post explaining all the areas and points you find objectionable that would be great – or if such a post does not exist you may consider writing one to make your position more clear to those of us who do not understand why would anybody (who doesn’t have a paycheck or profits at stake) object to Free Software and GPL. Thanks.

    1. >I knew there is a rift between Free Software advocates and esr but to this day I do not understand why exactly.

      I’ll give you a condensed version, trying to be fair to the other side.

      I judge the propaganda of “free software” to be so incompetent that it’s actually damaging to the hacker community – alienating to potential allies, and encouraging of needless fears.

      I reject the “free software” premise that proprietary software is morally wrong and people involved with it are evil. Instead, I argue that proprietary software development should be abandoned because, except in a few special cases which I have discussed, it leads to inferior engineering and bad results. (See “more effective propaganda”, above.)

      I reject the GPL. It’s an anti-solution to a non-problem. My reasoning is too complex to fit in a blog comment; see The Economic Case Against The GPL for discussion.

      I have urged hackers to cooperate with the free market and corporate development shops – to use capitalism as an instrument rather than regarding it as an adversary. In fairness, the FSF claims not to be anti-market – but almost nobody who has read their propaganda actually believes this.

      Rather than merely theorizing on these points, I spent years (roughly 1997-2004) running a public propaganda campaign so successful that Corporation X now thinks it can woo Wall Street Investors by talking up “the hacker way”. The Free Software Foundation’s more hardened zealots (including, on a bad day, RMS himself) cannot forgive me for this. Their charge is that I hijacked RMS’s revolution.

      My response is “Yes, I did. Somebody had to. Your implementation was seriously buggy.”

      RMS and I differ more about means than ends. He is an idealist who thinks action follows thought, and thus the most important thing is that people have the right ethical premises so right action will follow. I think this is exactly backwards for the 93% of the population that isn’t Myers-Briggs type XNTX; therefore, I teach them good selfish greedy reasons to do the right thing and expect that will cause them to eventually talk themselves into thinking in the way I would prefer. Which is pretty close to what RMS would prefer, except for the moralizing.

    1. >Mrf. The DFCW makes me twitch in its Stallmanite redefinition of “free”.

      Jay, with all due respect, you are sometime excessively twitchy about such things. This is both your friend and the theorist of open source telling you this.

  139. > calling Android “(low) ROI” ignores its strategic nature in helping to shield its key revenue-generating business from an Apple-driven disruption.

    Please explain how Google is going to earn back the investment in Android. We’re talking well over over $14B with the $12.5 billion all cash acquisition of Motorola.

  140. “You denied that network externalities are a significant factor in the smartphone market. I pointed out that this is inconsistent with previous claims by Apple partisans that network externalities from apps would smother Android. You are an Apple partisan. So how is pointing this out non-responsive, again?”

    Nah. I posited a reason why network externalities might not be as important as you’ve claimed. If you don’t know the difference between an assertion backed by an argument and a denial, you’re not a very good writer — but you are, and I suspect you do know the difference.

    But that’s irrelevant. The value of my argument doesn’t depend on me. Shit, I’ll even give you a pass on that bit where you classify me as an Apple partisan. Your syllogism looks like this:

    1. Some statements made by Apple partisans about Apple products are incorrect. (Some A are B.)
    2. Argument Y is a statement made by an Apple partisan about an Apple product. (Y is an A.)
    3. Therefore, argument Y is false. (Y is B.)

    Or maybe you mean to assert:

    1. All statements made by Apple partisans about Apple products are incorrect. (All A are B.)

    But if that’s the case, then this statement: “The underlying reason I frequently sound contemptuous of the Apple fanboys on here is that they don’t engage my causal model on this level,” isn’t entirely accurate. Perhaps you meant to say it’s because we’re always wrong. If that’s the case, I wish you’d been more clear. I wouldn’t have wasted time presenting a counter-argument to you; I can always get a reasonably interesting discussion with Winter and Patrick.

    1. >Or maybe you mean to assert:

      I mean to assert that you, like most Apple partisans, breezily switch from asserting A to not-A without noticing the contradiction between past and present behavior. Network effects are super-important and dominating, until they’d be favorable for Android when they’re not.

      Apple fanboys aren’t always wrong. They’re occasionally right in the same way a stopped clock is occasionally right – by accident, rather than because there’s a robust causal connection between reality and their world models (see “religious”). The underlying premise is always APPLE IS TEH AWSUM!!!

  141. It’s quite possible that I’m excessively twitchy about Stallmanites redefining “free”. Still, doesn’t redefining “freedom” to mean something that’s definitely not free make the libertarian in you cringe?

    1. >Still, doesn’t redefining “freedom” to mean something that’s definitely not free make the libertarian in you cringe?

      It would if I thought the DFCW were doing that. But they seem to me to be doing a pretty good job of homing in on “free as in speech”. Perhaps part of the reason for this is that it’s more difficult to misinterpret the phrase “free culture” than the phrase “free software”.

      English-speakers aren’t likely to think of “culture” as something that can be bought or sold in the way that specific expressions of it such as movies or music tracks can. So while the phrases “free movies” or “free music” are ambiguous due to gratis/libre confusion, “free culture” is not. This changes the semantic frame around the DFCW.

  142. “That is not a claim I have ever made; those are different thresholds. There will be a point at which Android-Apple competition becomes zero-sum. There will be a point at which Android goes over 50%. There’s no reason those have to happen at these same time, and they could happen in either order.”

    You have said:

    “Since early 2011 I’ve been writing that the U.S. smartphone market didn’t look like its players were in a zero-sum game. In particular, both Android and Apple were spared having to compete head-to-head in each others’ core markets by the huge volume of dumbphone conversions and customers bailing out of RIM and Microsoft. But I also expected this to change in 3Q2011 as the U.S. smartphone market neared saturation; this change was hinted at in the June numbers and I think the September ones show it arriving right on schedule.”

    “In a recent post I projected that serious zero-sum competition between Android and Apple wouldn’t begin until at least 3Q2011 after Android has chewed up the market share of all the soft targets – RIM and Microsoft, most notably. Kantar’s non-US figures give us a foretaste if what that’s going to be like, and they suggest that I may actually have overestimated Apple’s ability to retain share under Android pressure. It looks as though the disruptive collapse of iPhone has arrived in Europe.”

    Coincidentally, you predicted Google would have 50% of the smartphone market in 3Q2011, but it was statistically clear that only 40% of the US would have smartphones.

    Of course, you then moved the goalposts:

    “I thought the market would have transitioned to a seriously zero-sum game by mid-1Q12.”

    Coincidentally, this is when you now think Google will achieve 50% of the market. Again, the US smartphone market still will not be close to saturation.

    If this all mere coincidence, and you think there is some other arbitrary reason to determine the smartphone market saturated (zero sum game) but it just coincidentally should happen around the time that you keep saying something else should happen, I apologize but I still disagree with WHEN and WHAT you are calling a zero sum game. The market isn’t close to saturated.

    Now that you say these are not linked events but they are both likely to occur near-term, maybe even the same quarter, but they could now occur in completely different orders (Really? Now you think we could potentially enter a zero sum game BEFORE Android has 50% of the market?) I disagree with you even more.

  143. “Rather than merely theorizing on these points, I spent years (roughly 1997-2004) running a public propaganda campaign so successful that Corporation X now thinks it can woo Wall Street Investors by talking up “the hacker way”.”

    I’m fairly certain the market would be valuing Facebook at a $100 Billion with or without “the hacker way” and your propaganda.

  144. Tim F.: I agree that the valuation of Facebook is independent of their supposed embracing of “the hacker way”. That wasn’t ESR’s point. His point was that Facebook thinks they can improve their valuation by doing so. As far as the Open Source INitiative propaganda campaign is concerned, that is indeed a success.

  145. “His point was that Facebook thinks they can improve their valuation by doing so. As far as the Open Source INitiative propaganda campaign is concerned, that is indeed a success.”

    Firstly, I am certain that Facebook/Mark Zuckerberg have a very different definition for “the hacker way” than ESR’s. Secondly, I specifically disagree with the notion that FB citing “the hacker way” (something they’ve done for their entire history) has much to do with improving their valuation or thinking it can improve their valuation rather than simply being the story of the company for nearly the last decade. I highly doubt any FBers close to the IPO and their underwriters are trying to make an OSI-derived argument with investors, at least in comparison to the weight of their financial and market share data.

    1. >I highly doubt any FBers close to the IPO and their underwriters are trying to make an OSI-derived argument with investors, at least in comparison to the weight of their financial and market share data.

      Shows what you don’t know. In fact such arguments are made routinely as part of due-diligence reviews in software-related IPOs (and mergers, too). I know this because I’m sometimes asked to consult by the investment banks putting the deals together.

  146. @libreman:

    If you’re an open-source advocate why wouldn’t you want the software released as open-source to remain that way and prevent the situation when somebody just takes the codebase and makes it closed source/proprietary which makes us all lose the benefits it provided us – be it the freedom or technical soundness.

    This is a semantic obfuscation I see quite often on people arguing the side of the GPL. When I release free software under a permissive license, nobody can “make it closed and [make] us all lose the benefits it provide[s].”

    I find this argument infuriatingly specious. If somebody makes a copy of my software and adds stuff to it, they haven’t altered the availability of my software.

    I have no objection against people using the GPL, if that’s what it takes to make them feel comfortable adding their efforts to the commons, and I am extremely grateful for those who add to the commons. I quite happily use GPLed programs such as Linux, GCC, Inkscape, etc. Nonetheless, there are cases where I, personally, much prefer a non-GPL solution. These cases have to do with when I spend a lot of time learning the guts of a package (e.g. a library) and making it part of my personal “toolbox” or “arsenal” (depending on context). A GPLed component is not nearly as flexible there as a non-GPLed component. The worst case I saw of this was in a newsgroup where somebody slapped a GPL license on a 10 line snippet of not very well written code. That tail’s not wagging this dog.

  147. “>I highly doubt any FBers close to the IPO and their underwriters are trying to make an OSI-derived argument with investors, at least in comparison to the weight of their financial and market share data.

    Shows what you don’t know.”

    And I find it laughable to suggest that an OSI-derived argument is more important to FB investors than their financials — which is what I said — or that a page and a half out of an almost 200 page document is more important than the remainder of the doc, etc.

    1. >And I find it laughable to suggest that an OSI-derived argument is more important to FB investors than their financials

      Yes, you are ignorant. Certainly the financials are important. So is careful evaluation of strategic business risks. Vulnerabilities related to proprietary software, open-source software, and IP liabilities are among these. I’m also often asked to give “state of the market” appreciations when a company’s business plans project that it will need to either compete with open-source software or rely on the open-source community. Questions about the subject company’s relations with the open-source community, and their reputation in it, do get asked.

  148. He didn’t say “more important”. You didn’t either. The argument is made, and that’s enough: people are making the argument to investors as part of the overall package.

    Please stop making silly arguments like this. You are making those of us who use and appreciate Apple products look bad.

  149. @esr:

    You think the FSF’s virality is bad and confusing?

    Actually, I think “viral” is a bad term for this. I came up with (IMO) a better term a few years ago. Perhaps if others here agree with my analysis, we can spread this meme a bit farther. A virus can infect two organisms in contact, but that’s not really what’s going on here.

    The GPL is designed to win license competitions. If you view the license as part of the DNA of a piece of software, then whenever two packages “breed” (are combined) the resultant package will always have the GPL if either of the source packages did. In attempting to draw a biological parallel, many have equated the GPL to a virus, but this analogy fails miserably. The “selfish gene” analogy has much to recommend it, however:

    http://en.wikipedia.org/wiki/The_Sel…lfish.22_genes

    It’s an interesting exercise to extend the analogy to show how some of the GPL gene mutations (from v2 to v3) attempted to increase the fitness of the GPL license gene by allowing it to mate with even *more* licenses (and still always come out on top), yet one side effect of the v2 -> v3 mutation is that the mutated license can no longer interbreed with any software with the original v2 license, unless that other software also contains the “or later version” clause gene.

    So now there are two completely incompatible selfish gene FOSS licenses in the ecosystem. These selfish license genes always propagate whenever the host software mates with other software, but in order to have that genetic advantage, they simply avoid allowing their host software to mate whenever they couldn’t be the dominant license gene of the resultant package. One side effect of this is that the two major GPL variants are unable to mate with each other.

  150. Patrick: the word you’re looking for is “dominant”.

    I think the term “viral” is perfectly apt, but then that’s because I’m the one who came up with applying it to the GPL.

  151. >I judge the propaganda of “free software” to be so incompetent that it’s actually damaging to the hacker community – alienating to potential allies, and encouraging of needless fears.

    While being a Free Software advocate (open-minded to other alternatives, e.g. not a zealot) and after watching a few of rms videos I would have to agree on this count. It’s a marketing/PR disaster, that much is clear – I mean, the guy “browses” the Internet by mailing a link to some server that then mails a cached version of that website back to him and he looks at it the day after, the guy is out of touch with the real world unfortunately. And that’s the reason he thought it was a good idea to call the software that he hoped to become mainstream and get companies on board “free software” – who in their right mind would attempt to sell software called “free”, it short-circuits the brain from the get go and without a working economic model it isn’t going anywhere.

    That said I have to take issue with the second count as being a PR disaster doesn’t say anything about the validity of the underlying principles/philosophy.

    >I reject the “free software” premise that proprietary software is morally wrong and people involved with it are evil.

    Well, so you do not find morally objectionable for a software to have backdoors (like Windows and so many others), preventing the user to install software of his choice on *his own device* (DRM), removing content from users devices without consent of said user (Kindle), not allowing recorded media to be taken out of a device preventing perfectly legitimate fair use cases like political commentary etc. (TiVO), special backdoors for “law enforcement” or as appropriately called oppressive regimes to snoop on you and so on?

    Do you really not see anything morally wrong with that? Or do you simply not think about the very real *social consequences* of software? All of the above is what can be solved by open-sourcing the sw – I know of no OSS that would behave in a similar malicious ways but it is pervasive in the world of proprietary ones, at least the popular and widely used ones.

    As for that “people involved with it are evil”, I do not remember rms ever saying something like that, at least not in such a general way you’re using it – if you want to be “fair to the other side”, it would be a good idea not to use overblown claims like that. As for the final product being “evil”, I would probably subscribe to that myself in some instances ;) but that doesn’t necessarily imply that all the individual people making it are “evil”.

    >I reject the GPL. It’s an anti-solution to a non-problem.

    >I have urged hackers to cooperate with the free market and corporate development shops – to use capitalism as an instrument rather than regarding it as an adversary. In fairness, the FSF claims not to be anti-market – but almost nobody who has read their propaganda actually believes this.

    I’ll read your blog post but for now my take on this is that FSF and Free Software advocates (such as myself) are not regarding capitalism as an adversary, that’s in my opinion mostly a projection of people who think they know best how capitalism *should* look like – mainly if it’s not sold for a price under the penalty of punishment if not then it’s not capitalism, that’s bollocks!

    A model based on voluntary donations or paying for services as opposed to the software itself is as much capitalism as anything else. The problem is that companies do not reject or hesitate about GPL because it would be “anti-capitalistic” or anything like that, they reject it because they plan to impose something on their users they know they’re not going to like (backdoor, DRM, snooping features etc. etc.) and they do not want those users to have the ability to simply remove these unwanted “features” which is obviously considered unethical by the FSF. There is no other reason to reject the GPL, is there? You’re a open-source advocate, do you see any legitimate reasons a company should reject the open-source model that are not extremely narrow considering the actual current landscape?

    So I currently see it this way: GPL is not anti-business, but rather todays businesses are anti-ethical.

    But I’m open to being persuaded otherwise if I can be shown legitimate reasons of why a company should reject open-source while not planning to do anything fishy to it’s users. Do you know of any such reasons?

    The rest of your post I mostly agree to and see your point. I just think you were solving a real problem with inappropriate solution. The problem was PR/marketing in communicating that “Free Software” in not anti-business … so you “solved” it by getting rid of the very principles that defined it instead of communicating them better. Well, that’s understandable if you do not identify with those principles and so find them expendable, I just find hard to believe that this is actually the case and you really do not see any moral/ethical dimension to the free/proprietary problem. I think you rather choose to ignore it or believe it should be ignored – not that it does not exist.

    1. >Well, so you do not find morally objectionable for a software to have backdoors (like Windows and so many others), preventing the user to install software of his choice on *his own device* (DRM), removing content from users devices without consent of said user (Kindle), not allowing recorded media to be taken out of a device preventing perfectly legitimate fair use cases like political commentary etc. (TiVO), special backdoors for “law enforcement” or as appropriately called oppressive regimes to snoop on you and so on?

      I do find these things objectionable, and frequently speak out against them. But…

      1. They are much more specific than “proprietary vs. free”. You can take a strong position against them while not buying into the “all software should be free” premise.

      2. I don’t think the language of moral outrage is very helpful in these situations even if what you feel is legitimate moral outrage. If you want to persuade, you need to do it by appealing to your listener’s interests and concerns rather than trying to impose your interests and concerns on them.

      >But I’m open to being persuaded otherwise if I can be shown legitimate reasons of why a company should reject open-source while not planning to do anything fishy to it’s users. Do you know of any such reasons?

      Yes, I do. You should read my discussion of some of these exceptional cases in The Magic Cauldron.

  152. Realized I didn’t expound on why I think viral is inaccurate.

    If I have two pieces of software near each other — on my computer, on a CD, whatever — the GPL won’t jump from one to the other. It’s only if I somehow link the two pieces of software together that the GPL is attached to the aggregate. But even then, the GPL still doesn’t attach to the non-GPLed version when distributed by itself.

    Having said that, “viral” is accurate insofar as it refers to the meme — the idea of using GPLed software — passing from programmer to programmer, and how it’s hard to stop once you’ve started using it on a particular piece of software.

    But if you consider each program to be an organism, and the license to be part of the DNA of the program, I think the selfish gene analogy works better.

  153. @libreman:

    The fact that you think the sole problem with RMS and the FSF is “marketing” shows that you disagree completely with everything esr just said, and the fact that you think he hasn’t thought about the consequences of his position shows you haven’t read very carefully at all. The fact that you think that what he said means that he’s OK with backdoors in proprietary software shows that you aren’t paying attention and/or are very rude, and the fact that you don’t think that RMS would ever declare proprietary software people evil shows you aren’t paying attention, like to sweep things under the rug, and/or don’t know how to use google.

    You’re obviously a zealot, despite your protestations to the contrary, and in complete accordance with your chosen moniker. I will attempt not to feed your trolls any further, but please feel free to come back with original and well-considered arguments.

  154. Patrick: The GPL is a legal virus that contaminates everything it touches. There are viruses that require more intimate contact than simply being next to each other to propagate. (HIV, anyone?)

    libreman:

    The problem is that companies do not reject or hesitate about GPL because it would be “anti-capitalistic” or anything like that, they reject it because they plan to impose something on their users they know they’re not going to like (backdoor, DRM, snooping features etc. etc.) and they do not want those users to have the ability to simply remove these unwanted “features” which is obviously considered unethical by the FSF. There is no other reason to reject the GPL, is there?

    Wrong, and highly revealing of the Stallmanite bias.

    Companies avoid infecting their software with the GPL because a) their lawyers tell them they can’t guarantee they can protect themselves and their competitive advantages in the GPL world, and b) they aren’t interested in becoming the next poster child for the Stallmanite zealots to attack just because they miss dotting an i or crossing a t and therefore don’t do everything in the Stallman-approved manner. Neither of these requires anything in the way of nefarious motives.

    The GPL is a terrible license, in either incarnation, to anyone who’s not interested in joining the Stallmanite utopia. Thoroughly upright companies will nevertheless either choose software distributed under some other license or develop their own rather than get sucked into that particular pit of quicksand.

  155. @Jay Maynard:

    The GPL is a legal virus that contaminates everything it touches.

    Contamination only occurs via propagation. A preexisting copy of software doesn’t magically become infected. The software has to mate with a (possibly null) piece of software with the GPL, and the offspring contains the GPL.

    In any case, I think we agree on the essentials as to how it happens and how pernicious it is, and I agree that “viral” is an extremely catchy way to describe the process, but I still find it inaccurate.

  156. libreman: One other note:

    But I’m open to being persuaded otherwise if I can be shown legitimate reasons of why a company should reject open-source while not planning to do anything fishy to it’s users.

    Open source != GPL. It’s entirely possible to be open source and stay far far away from the GPL.

  157. “You are making those of us who use and appreciate Apple products look bad.”

    What does my argument with the significance of ESR’s meaning of “the hacker way”, while admittedly coming from one the central figures in the development, promotion, and success of the OSS movement, to the financial success and IPO of Facebook have to do with Apple?

  158. I agree almost fully with ESR on the FSF issue, the biological metaphor for GPL, etc. I also happen to think, that while both movements are important, they are sometimes less biologically/evolutionarily “robust” as commercial software. I don’t think my views on the GPL are as strong as Jay Maynard’s — I just find it even less robust than more permissive licenses.

    I am using biological/evolutionary “robustness” fairly abstractly, but the closest sense would be: 1. top of the food chain in ecosystem, 2. most immune or resistent to predation, health risks, other threats to existence, 3. most “complex,” “sophisticated,” and/or “intelligent” lifeform (this is where it gets more abstract).

    FS and OS software is q

  159. Oops.

    FS and OS software is quite important and significant. But it’s more often plankton or bacteria (both hugely significant) rather than a primate, octopus, or shark.

  160. Patrick Maupin,

    Sprint has a couple of Android phones you can buy from them without a contract. 35$ a month for 300 minutes and unlimited everything else. Phones are $149 and $199,

    Yours,
    Tom

  161. Also, when I speak of “robustness,” I am not merely speaking of the code and its license: I am referring to the intersection of the “packaged” end product software and the business or organization of development resources creating the software. So you can end up with a robustness food chain roughly like: khtml < Opera < Mozilla < WebKit < Epiphany < Chromium < Safari (including iOS) < Firefox (possibly still ahead of both Safari and Chrome, but I think this is temporary) < Chrome & Android Browser < IE (Still, unfortunately).

  162. >I do find these things objectionable, and frequently speak out against them. But…
    >1. They are much more specific than “proprietary vs. free”. You can take a strong position against them while not buying into the “all software should be free” premise.

    Of course you can – it just doesn’t make a lot of sense, the question is do you have a better solution for those problems than using Free Software (open-source)?
    You said: “I reject the “free software” premise that proprietary software is morally wrong”

    Well, the reasoning of FSF advocate such as myself goes like this: backdoors are morally objectionable (you agree) -> backdoors are readily found in proprietary software where they’re hard to detect and remove (your position?) -> proprietary software is also morally objectionable because it facilitates former morally objectionable practices (you disagree). The solution: Use Free Software/open-source where detection and removal is a piece of cake (your solution: unknown)

    So clearly the disconnect is in the middle where I would like to know whether/why you do not accept that statement. It is quite natural to me to consider something that facilitates morally objectionable practices which are associated with it and seen in practice to be associated with it to be morally objectionable themselves – we may technically say they’re not, that it’s just the practices themselves but that’s just semantics and playing with words – clearly the FSF is opposing proprietary sw because it facilitates unethical practices and it’s those practices it wants to eliminate by the means of FS, whether you also call the sw itself unethical is irrelevant and nit-picking, is it not? That’s clearly not the point, the point is to get rid of the unethical, morally objectionable practices which is accomplished by using and advocating Free Software only. I do not know what your solution to the same problem would be. Let the market sort it out? Well, I am the market and I’m sorting it out by using Free Software, YOU are the market and the people reading this blog are the market – how are you going to sort it out?

    >2. I don’t think the language of moral outrage is very helpful in these situations even if what you feel is legitimate moral outrage. If you want to persuade, you need to do it by appealing to your listener’s interests and concerns rather than trying to impose your interests and concerns on them.

    This is just half of the possible solution and therefore not entirely correct. Yes you could appeal to listener interests however those interests are not fixed and immovable like you imply (you’re doing the same with markets etc. where you act like you were just an observer unable to change them which is clearly incorrect) so you could also make him realize that Free Software is in his interest by explaining why – most people just do not understand the connections. I’d never become a FS advocate if rms was trying to “appeal to my interests” instead of explaining and make me realize why those interests are wrong. In fact we would never get anywhere in anything if everybody did this, we would probably still have slavery if somebody didn’t come along, found it morally objectionable and persuaded people of the same.

    This IMO reveals a deeper difference between you and rms – you like to adapt to whatever the situation is and act in essence like an observer, predicting things based on what you believe others will do as if you were outside the system unable to influence the direction yourself, you ‘go with the flow’ essentially. While rms is a mover, he is trying to influence/change the direction people are moving in by persuading them of the reasons why they should, perhaps a little too much so ;) and he doesn’t care what “the majority” thinks.

    This manifests itself in your writings – you never seem to mention the option of actually influencing the world … like in this instance you completely omitted the option of making the person revise his interests. You rather opted for you changing yours by appealing to his as if they were rigid and unchangeable. You do the same with “the markets” where you talk about them as if they were somehow separate from you and ..well.. actual people, failing to account for a possibility that if the FSF ideals start to appeal to a large portion of the population “the market” will fundamentally change – it’s the people who make the market, not the other way around. If you persuade people you change the market – it seems you’ve completely given up on that possibility.

    And that is exactly what FSF is trying to do – change the market. Not to “appeal to its existing interests” as they consider them misguided – the solution to that is to persuade them by logic that their interest are in fact misguided not to resign your own principles to “appeal to them”.

    1. >Of course you can – it just doesn’t make a lot of sense, the question is do you have a better solution for those problems than using Free Software (open-source)?

      Your logic is broken. Of course the best overall solution to these problems is open source, and that’s the one I work towards because it’s the best solution for a lot of other problems you didn’t mention, too. But those particular problems are only arguments for the necessity of open source if there are in principle no other solutions.

      Take intrusive DRM, for example. It could be effectively prevented by a legal ban, because violations would be easy to detect and cause actionable harm. So, while open source is sufficient to solve this problem, it’s not strictly necessary. Which in turn means you cannot argue deontically from “DRM is evil” to “closed source is evil” – the relationship is underconstrained. Logic fail!

      To put it most generally: the premises “B is evil” and “A prevents B” do not imply “~A is evil”. You have to strengthen the second premise to “B cannot be prevented by means other than A” before you get a syllogism that actually works. (This kind of reasoning is studied under the name “deontic logic”.)

      >I’d never become a FS advocate if rms was trying to “appeal to my interests” instead of explaining and make me realize why those interests are wrong

      That’s because you have an XNTX personality, like RMS and like me. That’s fine, except that about 93% of the population and effectively all of the non-hackers we need to persuade aren’t wired like us. When you throw complex, abstract moral arguments at that 93% their brains shut down and they get stressed out. This is not their fault; they are simply not equipped to process abstractions as well as you are. Appealing to fear, greed, vanity, or desire for increased social status, by contrast, actually works.

      >you ‘go with the flow’ essentially.

      Heh. You don’t know me very well if you think that. I’m as stubborn and determined to change peoples’ behavior as RMS. The difference is that he argues as though every human on the planet is a sort of less-bright version of himself. I know better, and accept human beings as they actually are, and am as a result a much more capable propagandist.

      >you completely omitted the option of making the person revise his interests.

      Yes. This option is doomed. People who attempt it pretty much invariably fail and stir up resentment.

  163. The FSF has been trying to change the market for 25 years. How much success have they had?

    You will never get anywhere by trying to convince your customers they’re wrong.

  164. >The FSF has been trying to change the market for 25 years. How much success have they had?

    They’d HUGE success, it’s quite possible Linux wouldn’t be even here without rms and GNU, the internet would be completely different place without the freedom it provided to its infrastructure. Not Stallman nor Torvalds, two of the most influential people who made this revolution happen, were motivated by financial returns or “economy” and couldn’t give two farts about “the markets” which seems to be all that esr is talking about and analyzing therefore he is missing the only deciding factor as proven by history to actually matter – motivation by ideals, principles and philosophy behind ones actions.

    I’ve read NOTHING about that here as a factor. All is about businesses, markets and what makes “economic sense” … all of which has proven impotent at achieving any game-changing events in a way Free Software or Linux has shown. All of businesses using open-source/Free software are free-riders riding on the software made possible by following a certain set of principles and ideals NOT any business model. And it is my conviction that if these ideals evaporate so will the “open-source” for all practical purposes as companies will “open-source” only what is convenient to them and doesn’t threaten their established practices preventing any meaningful disruption in the direction of more competition and therefore more and better sw/services. The market is going to stagnate as a result and the only things “open-sourced” will be those that don’t actually matter much.

    They’ll gravitate towards “open-core” transforming open-source into a kind of new shareware model – try it free, pay for premium features! It’s already happening and without GPL to prevent it it’s going to be a growing problem.

  165. @libreman:

    You’ve got it exactly backward. The internet greatly enabled the spread of GPLed stuff, not the other way around. Torvalds might have chosen a different license absent the GPL, or BSD, which was escaping its legal cloud just as Linux started taking off seriously, might have been ascendant. Either way, the vacuum of freely available software would have been filled.

  166. >You should read my discussion of some of these exceptional cases in The Magic Cauldron.

    I’ve read it and I do not see anything about preventing backdoors, anti-features like DRM, secret snooping etc. I see only decisions based on whether it makes “economic sense” (e.g. money can be made off of it) to go open-source but that was not the issue – it may make perfect economic sense to introduce DRM and/or backdoor, that does not make it morally acceptable, right? Have I missed it? If so could you direct me to the place where I can find considerations for such cases?

    You acknowledged that these practices are morally objectionable and you oppose them so I’d like to know what is your idea of preventing them from happening other than making the software open-source.

  167. @libreman:

    If we’ve gotten to the point where the best argument for the GPL is it makes it more difficult (note: it doesn’t make it impossible) to insert backdoors, then:

    a) you still haven’t shown any GPL superiority to permissively licensed software; and
    b) this feature is most important in the OS, because the OS can alert you if apps are trying to be bad.

    But I’ve never seen a serious claim that Linux is more secure than OpenBSD.

  168. I’ve got it exactly backward? Or you’re trying very hard not to understand what I’m trying to say? So it’s not Linux that enabled every other to set up a webserver with a wordpress running on it democratizing the publishing? Is it not the reason I can set up such a system in an hour right now at no cost at all without being at mercy of some company? That is what I have meant …

    Sorry, but I will not respond to any more of these posts other then esr as it seems he is the only one with which a meaningful conversation is possible – I have to give credit there Eric as you are reasonable enough to actually address the relevant points and be constructive even when you do not agree, that’s not seen often these times. I’ve become quite used to irrational reactions like that of other participants calling me a Stallmanite or whatever and being a zealot merely for revealing the apparent secret that I support his ideas of FS (no shit Sherlock) and now Patrick equating my specific historical examples and circumstances of how things transpired to his fantasy speculation of what would surely happen if not for Stallman and Torvalds. It’s very frustrating and it’s making me lose focus so I’ll not respond anymore, sorry.

    You should really examine who is the biased one and a “zealot” as it is clear it’s not contained to rms supporters, but again to your credit you do note when you believe your supporters go overboard in their claims. So props for that ;)

    I also have to say that your approach is a polar opposite of rms who seems unable to consider other perspectives … from my own experience. So while a do see merit in some of his ideas about software, I also seriously dislike his personality (well, that’s probably an understatement – I actually hate it) and as I already pointed out I too as yourself believe it may do (or already has done) more harm to Free Software/open-source than good.

    So I’m looking forward to continuing this conversation …

  169. @Patrick

    OK, last reply :)

    If we agree that the best way to make it more difficult (nit-picking much?) to introduce anti-features is to use open-source then we agree that all software should be open-source (obviously to prevent such anti-features from getting in) so that makes “permissive” licenses counter-productive as the only reason you would use such a license is to allow for proprietary implementations which we just agreed are undesirable as they could introduce said anti-features and make them hard to detect and rectify.

    So the only license that makes sense in such an environment is copyleft license such as GPL to assure that nobody can insert such anti-features and keep them hidden – if we allowed that, the whole exercise of open-sourcing it was meaningless as anyone can get around the transparency it provides by simply shipping his own modified version without releasing the sources.

    As an example I can not regard Android I get on my new phone as open-source software (except the kernel) because there could be any number of undisclosed modifications for which there are no sources released and therefore no way to easily detect and possibly change them. Case in point – Carrier IQ so this is not just hypothetical you see.

    But now really, I do not have the time to respond to every comment directed at me so I’ll wait for esr to comment.

  170. @libreman

    You’re entertaining, and a nice counterpoint to the Apple dudes we usually get. I especially like how you claim you aren’t a zealot, then proceed to fake or exhibit ignorance of anything beyond an FSF worldview, immediately try to re-frame the topic to match that worldview, and throw a minor hissyfit when people refuse to acknowledge it. Good times. Keep posting and giving me a chuckle.

    My $0.02, from my perspective your idea of ‘freedom’ is one that removes my own in order to combat some amorphous maybe-threat. If I could easily, fully expunge GNU’s overwrought crap from my linux boxen but keep the supports-everything kernel, I would. And please, don’t try the whole ‘if it weren’t for GNU there wouldn’t be Linux’ argument. It assumes too much.

  171. @jsk:

    > And please, don’t try the whole ‘if it weren’t for GNU there wouldn’t be Linux’ argument.

    Too late.

  172. “Sorry, but I will not respond to any more of these posts other then esr as it seems he is the only one with which a meaningful conversation is possible…”

    You’ve got to be kidding. Patrick is one of best, most level-headed posters on this board. He posts both facts and opinions, and always keeps the line between them clear.

    But it’s your choice to whom you respond.

  173. “Not Stallman nor Torvalds, two of the most influential people who made this revolution happen, were motivated by financial returns or “economy” and couldn’t give two farts about “the markets” which seems to be all that esr is talking about and analyzing therefore he is missing the only deciding factor as proven by history to actually matter – motivation by ideals, principles and philosophy behind ones actions.”

    Anyone who believes this about Linus needs to read some of the many interviews he has given and posts he has made around the Net. He is very much a pragmatist in the ESR mold, not an ideologue like RMS. Check out his use of BitKeeper for years as one obvious example.

  174. @Cathy
    Don’t let Linus’ declared “pragmatism” fool you into thinking he is not a die hard idealist.

  175. As an example I can not regard Android I get on my new phone as open-source software (except the kernel) because there could be any number of undisclosed modifications for which there are no sources released and therefore no way to easily detect and possibly change them. Case in point – Carrier IQ so this is not just hypothetical you see.

    Actually Carrier IQ is a brilliant example… but not of the point you are trying to make.

    Yes Carrier IQ was put onto phones and even on this “non-free” software people found it and publicised it.
    They reverse engineered what it was doing, teased apart its secret sauce and brought its workings into the light of day.
    Most importantly it was very quickly being disavowed by business.

    You don’t need to scare off all your potential converts with the GPL to stop people doing bad things to their customers. All you need to do is publicise the bad things and let the market (and people’s eagerness to believe the worst of business) do the work for you.

    RMS’s seemingly insane (from their perspective) rants are never going to make that kind of difference.

    1. >Actually Carrier IQ is a brilliant example… but not of the point you are trying to make.

      JonCB is correct. Let A = “software and hardware is fully open” and B = “CarrierIQ remains deployed”, and add the predicate E(x) = “x is evil”. In formal deontic phrasing we interpret E(x) as OB(~x), that is “it is obligatory that not x”. Than A -> ~B, but the inference OB(~B) -> OB(A) does not follow. To show this is true, we sketch a proof using a Kripke-style possible-worlds interpretation. For this purpose we need to introduce the predicate IM(x), “it is impossible that x”, interpreted as “x is false in all possible worlds”.

      Now: OB(x) -> OB(y) only if IM(x & ~y) – that is, it is impossible to satisfy obligation x without satisfying y. Suppose M = “market pressure kills CarrierIQ”. What JonCB points out that we live in an M-world in which ~A & ~B. Set x = ~B and y = A. We have ~B & ~A, therefore ~IM(~B & ~A), therefore OB(~B) -> OB(A) is falsified

      What I was pointing out earlier is if B = “intrusive DRM is deployed” and if there is any proposition P for which there there is a P-world in which ~A & ~B, then ~IM(~B & ~A); therefore OB(~B) -> OB(A) is falsified. Or, to put it in less formal language, “If there is any way to wipe out intrusive DRM without a fully open-source software and hardware stack, than it cannot be the case that the obligation to wipe out DRM implies the obligation to have a fully open-source software and hardware stack.”

      And that is more formal logic than I’ve done at one go since 1977. Kind of nice to know I can still hack this stuff, actually.

      1. I wrote: “Kind of nice to know I can still hack this stuff, actually.”

        Then I took my morning shower, and got to thinking about modality and Kripke semantics, and had an insight. I think it might be possible to reason rigorously about deontic relations like “the obligation to do x nearly implies the obligation to do y” with a maneuver very like the one I just performed, moving the discussion to modal logic via possible-worlds interpretation – then applying a sort of measure-theoretic approach over possible worlds. “Nearly implies” might unpack to something like “implies except over a subset of measure zero”.

        Neat!

  176. @esr:
    “To put it most generally: the premises “B is evil” and “A prevents B” do not imply “~A is evil”. You have to strengthen the second premise to “B cannot be prevented by means other than A” before you get a syllogism that actually works. (This kind of reasoning is studied under the name “deontic logic”.)”

    Thats not enough. You must also argue that B is less evil than A.

  177. >Your logic is broken. Of course the best overall solution to these problems is open source, and that’s the one I work towards because it’s the best solution for a lot of other problems you didn’t mention, too. But those particular problems are only arguments for the necessity of open source if there are in principle no other solutions.

    No, it’s not actually broken – it’s just that what you correctly pointed out as a necessary condition is not stated explicitly but as you correctly observe it’s implied and I’m well aware of that.

    >only arguments for the necessity of open source if there are in principle no other solutions

    This is actually not accurate for several reasons:

    1. I’m not talking about some absolute necessity of open-source, I’m saying that it is the best and only reasonably feasible solution to the outlined problems
    2. It is the correct solution to pursue not “if there are in principle no other solutions” but if there are not any reasonably feasible solutions comparable with going open-source (in other words if there is a hypothetical solution that is 10x as hard to achieve as OSS then it can not be really considered a solution at all as we might just go OSS instead and not bother)

    And as you may have guessed I’m convinced that OSS is the only reasonable solution on the plate and there is no other reasonably feasible in comparison – it was implied in my previous statements.

    >Take intrusive DRM, for example. It could be effectively prevented by a legal ban, because violations would be easy to detect and cause actionable harm. So, while open source is sufficient to solve this problem, it’s not strictly necessary. Which in turn means you cannot argue deontically from “DRM is evil” to “closed source is evil” – the relationship is underconstrained. Logic fail!

    I’m a little stunned by this paragraph. Is a person of your convictions actually proposing a series of government bans instead of free-market solution of exhorting a market pressure by the means of OSS? OK, now that I’ve recovered the problem is this – you’re talking in absolutes, this is not what I’m doing (or FSF, rms or anyone else), nobody is arguing that OSS is “strictly necessary” and believe you know this so why do you frame it in this way?

    Let’s talk about the real world, what FSF is saying is that in the current frame of reference and environment it does follow to argue from “DRM is evil” to “closed source is evil” as there is no other reasonably feasible solution to that problem comparable with that of OSS. btw can we drop the charged word “evil”? It’s not used by rms, the FSF or myself but rather by those trying to create a hyperbole which I hope is not you. Can we just say that DRM is bad for you and therefore closed source is bad for you? Thanks

    Also, if I take your above paragraph as if you actually meant it then you’re proposing a series of government bans as a solution to these problems that OSS can solve also as a venue to pursue instead of using OSS. Is that your actual belief and official position? Are you going to recommend that course of action to people who are concerned about those problems?

    As if you’re not then I do not see the point of even mentioning it as we both can see that such an option is there but can not feasibly be or even should be pursued. So in case you were talking in strict hypotheticals, can we get back to the real world please? ;)

    >That’s because you have an XNTX personality, like RMS and like me. That’s fine, except that about 93% of the population and effectively all of the non-hackers we need to persuade aren’t wired like us.

    There are several very deep issues that are wrong with this kind of reasoning. The most obvious and direct one is that you do not need to convince anywhere near a majority to succeed. American revolution was won by 3% of the colonists (http://sipseystreetirregulars.blogspot.com/2009/02/what-is-three-percenter.html) and I doubt that using and advocating for OSS is anywhere near as challenging as winning a revolutionary war.

    It’s always the persistent minority that decides the future so while it’s possible you could convince “only” 7%, that’s also all you need because those are the people that actually matter and decide the direction the society goes in. Those 93% just as they can not think abstractly (your words and numbers) can not also become leaders for the same reason, they’re almost by definition followers and are going to go whenever social peer pressure takes them. Would you object to such an assessment?

    I feel it’s fair considering historical examples. Google (or even Apple or MS) “population” is minuscule compared to the whole population but they control almost 50% of mobile OS market so your reasoning is clearly inadequate in this instance. You do not need such numbers that you imply at all.

    >Heh. You don’t know me very well if you think that. I’m as stubborn and determined to change peoples’ behavior as RMS.

    If that’s so I did not notice that in your writings, it may be as you said that you believe it’s futile to try so you resign and now do not even acknowledge it as an option. I believe that is a mistake and you’re underestimating the power of this dynamic. Determined minority working on the basis of a set of strong principles can have great effect (if their PR isn’t utterly incompetent that is)

    >Yes. This option is doomed. People who attempt it pretty much invariably fail and stir up resentment.

    No, this is the only thing that stopped PIPA/SOPA and is hindering ACTA as just one example, as I said you’re underestimating the power of this dynamic.

    1. >I’m a little stunned by this paragraph. Is a person of your convictions actually proposing a series of government bans instead of free-market solution of exhorting a market pressure by the means of OSS?

      No, and you’re being dimwitted if you think so. I’m simply demonstrating that your claim about entailment of obligation is false. Your logic is sloppy; you’re letting the conclusions you desire to reach distort your reasoning.

      > btw can we drop the charged word “evil”?

      Fine. RMS definitely uses the word “immoral”. We can change every occurrence of “evil” in this discussion to “immoral”, but doing so will not fix the bugs in your logic.

      >It’s always the persistent minority that decides the future so while it’s possible you could convince “only” 7%, that’s also all you need because those are the people that actually matter and decide the direction the society goes in.

      Wrong. Because in this case, the “persistent minority” you need to persuade is the people who write the checks that present a demand pattern to software developers. This is not the same minority you can persuade with RMS’s arguments – in fact, if you try that, the check-writers will either shut you out or attack you. Which explains why the FSF has always struggled, mostly unsuccessfully, to exert even the most marginal influence outside the hacker community itself.

      >No, this is the only thing that stopped PIPA/SOPA and is hindering ACTA as just one example, as I said you’re underestimating the power of this dynamic

      You misunderstand what stopped SOPA and PIPA. We didn’t succeed there by making a moral argument, we succeeded by appealing to peoples’ fears that they would lose control of their lives and of resources valuable to them. Declaiming that “x is immoral” doesn’t work on the 93%; pointing out that “x will injure you” does.

  178. >Yes Carrier IQ was put onto phones and even on this “non-free” software people found it and publicised it.

    This is not an argument to use (or not to oppose) non-free software, quite the opposite. It was by chance that it was discovered and it is a exception rather than a rule that it got discovered. This is confirmed by the pervasive use of similar backdoors in a wide spectrum of other non-free software some of which people even know about but are unable to remove – like Windows for example. We know there is abckdoor there, but we’re powerless to remove it – the only (reasonably feasible) solution to that is to use OSS … and to guarantee that that OSS remains OSS and is not hijacked and the risk of backdoor reintroduced it needs to be also GPL. It’s that simple.

    So the proper market response should be to reject proprietary software and use GPL licensed as much as possible. This is perfectly valid free-market solution. But for some reason some free-market zealots don’t consider it such … not because it wouldn’t be free-market solution but because they want to impose their idea of what a market response *should* look like onto others.

  179. to guarantee that that OSS remains OSS and is not hijacked and the risk of backdoor reintroduced it needs to be also GPL. It’s that simple.

    This is exactly equivalent to “fucking for virginity”.

    Open source software cannot be “hijacked”. Software under any license that meets the Open Source Definition will always and forever be available in its pure, unmodified form. Just because Apple uses FreeBSD as the base for OS X has no effect at all on the continued availability of FreeBSD. This has always been true, and Stallmanites have always used your false argument to claim otherwise. It goes back even unto SunOS: if Sun could have caused BSD to stop being available by releasing a proprietary OS, don’t you think they would have?

    What you’re doing is arguing that I shouldn’t have the freedom to choose OS X, additional proprietary code and all. The hell with that. True freedom must necessarily include the freedom to do that which pisses others off, and yet that’s the exact freedom the Stallmanites want to deny me. Hence, it’s self-contradictory, just like “fucking for virginity”. You don’t get to define my freedoms.

  180. >This is exactly equivalent to “fucking for virginity”

    How so? You lost me here …

    >Open source software cannot be “hijacked”.

    Let me explain how I mean that. If I buy an Android phone in a store I supposedly get the open-source OS Android in it – but this is not true by the time I buy the phone – the “Android” loaded in that phone is not open-source anymore in any practical sense. It got “hijacked” and I can not be sure what’s in it as the sources of the modifications aren’t published … just like with proprietary software and we’re at square one.

    That’s what I mean by “hijacked”. With GPL this could not happen – if Android was GPL then I can be sure I and many others do have access to the source code of the *actual software loaded on my phone* not some other software on Google servers that it is just derived from – that mises the point. I do not care what Google has on its servers, I want to know what’s running on *my phone* so I can be reasonably sure it’s not running any anti-features like said backdoor.

    But you’re missing much deeper point. I’m arguing that proprietary software is bad for you as it can not be trusted to not work against your interests and therefore proprietary software as a category should be avoided to prevent the great risk of anti-features like backdoors, DRM, snooping etc. from manifesting. In that kind of framework it doesn’t make sense to “allow” for proprietary derivatives because you believe they’re harmful and risky – from that point of view BSD is more bad than GPL as it allows for the type of software you consider harmful, do you understand that?

    In order to defuse that kind of reasoning of mine (and rms and FSFs) you would need to propose just as or more reasonably feasible method of eliminating said anti-features from proprietary software as practice shows they’re pervasive in it so clearly current method don’t work. On the other side you’d have very hard time finding such anti-features in OSS – therefore we should strive to produce OSS code exclusively and protect it from getting proprietarized ;)

  181. I’m arguing that proprietary software is bad for you as it can not be trusted to not work against your interests

    Stop right there.

    You don’t get to define my interests, either.

    I have the right to decide what is in my own interest. If that means that I trade the possibility of having a backdoor in the software for having that software be usable for my needs, then that is my right, and you do not get to define it away in the name of saving me from myself.

    I don’t need to show that I can prevent backdoors. All I need to show is that your premise – that the risk of backdoors must be eliminated at all costs – is invalid. I categorically refuse to accept that kind of nannyism, and it is highly destructive of true freedom. Hence, fucking for virginity. Your actions that claim to advance freedom in fact destroy it.

  182. >No, and you’re being dimwitted if you think so. I’m simply demonstrating that your claim about entailment of obligation is false.

    If you’re not proposing an actual solution other than going OSS that you would actually recommend then you’re not demonstrating anything (except that you’re more concerned about semantics than real solutions). Let’s summarize:

    1. We agreed some things (which I call anti-features) are morally objectionable
    2. I said that the only working solution to those anti-features like backdoors, DRM and snooping is OSS
    3. You replied that it’s not necessary solution as there are others like a government ban
    4. I reply that I’ve never claimed that is a necessary solution, only that it’s the only reasonably feasible one so unless you can provide a similarly or more reasonably feasible solution the original point stands (and express dismay that your previous proposition was actually serious)
    5. You tell me that I’m dimwitted if I think you meant what you’ve written seriously

    You do not see anything peculiar in the above? :)

    You propose a solution that you do not actually believe is a solution and call me dimwitted? Anyway, I already replied in case you didn’t mean it seriously (which I’ve expected you didn’t, just couldn’t understand why you mentioned it then as a possible solution when you do not actually believe that it’s a solution, eh):

    >As if you’re not then I do not see the point of even mentioning it as we both can see that such an option is there but can not feasibly be or even should be pursued. So in case you were talking in strict hypotheticals, can we get back to the real world please?

    So why have you ignored it? Why do you propose solutions that you do not believe are solutions? it seems to me that you’re trying very hard to hold on to the idea that proprietary software can not be called bad as a category and are willing to break and contradict your own convictions in the process to achieve it (cognitive dissonance?). I believe that if you stopped doing that and looked at things objectively and pragmatically rather than hypothetically you would come to the inevitable conclusion that proprietary software indeed can be called bad as a category considering current state of affairs.

    Because you still did not contradict my original point. SO here it goes again:

    If OSS is the only reasonably feasible solution to the problem of backdoors, DRM, snooping etc. pervasive in proprietary software then it follows that we should pursue OSS and oppose proprietary software (e.g. procliam it “bad” for those reasons) as a solution to that problem. If we consider OSS as the solution and proprietary software as the enabler of said bad practices then it makes sense to use and promote copyleft licenses such as GPL instead of BSD to help keep already open sourced software open and discourage closed software.

    1. >Why do you propose solutions that you do not believe are solutions?

      They’re solutions all right, just rather unpalatable ones. And I point them out to demonstrate that your reasoning is sloppy. This matters, because people who reason sloppily about important things often end up sabotaging themselves – and like RMS, you spend a lot of effort moralizing in a way that is seriously self-sabotaging.

      >If we consider OSS as the solution and proprietary software as the enabler of said bad practices then it makes sense to use and promote copyleft licenses such as GPL instead of BSD to help keep already open sourced software open and discourage closed software.

      And here’s another example of where your reasoning is sloppy. Promoting the GPL doesn’t follow from your premise alone. You need additional premises (which you don’t specify) to go from “improving open source software” to “promoting the GPL” – especially when I’ve already pointed you at an economic analysis showing that the GPL is probably, on net, harmful to the cause of openness.

  183. >You don’t get to define my interests, either.

    Absolutely, if you do not mind backdoors in your software then I’m not talking to you … I’m talking to just people who do mind such backdoors such as esr. Case closed, have a nice day ;)

  184. Absolutely, if you do not mind backdoors in your software then I’m not talking to you

    But by casting it as a question of morals, you are putting yourself in the position of dictating my choices, as surely as the fundamentalist Christian who decries dancing. It’s beyond question that RMS wants to eliminate proprietary software as a choice for all, and your words say you support that goal. I will fight that destruction of true freedom as I would any other destruction of true freedom.

  185. >Fine. RMS definitely uses the word “immoral”. We can change every occurrence of “evil” in this discussion to “immoral”, but doing so will not fix the bugs in your logic.

    Immoral I can accept, that’s very different from “evil”. Just take note that immoral isn’t just an abstract concept but has very real consequences like persecution of politically inconvenient bloggers, whistleblowers etc. (so to put in your framework – people can be ‘scared’ by these consequences if they do not get the ‘abstract’ notion of immorality)

    >Wrong. Because in this case, the “persistent minority” you need to persuade is the people who write the checks that present a demand pattern to software developers.

    You’re shifting arguments. Fist it was that you need to persuade too many people, now it’s that you can not persuade the right people which is also very much different from my actual point that it’s those people that you need to persuade – I said nothing about complexity of that feat. Please be more careful about your arguments … you’re moving into a territory where cognitive dissonance is getting the better of you.

    >You misunderstand what stopped SOPA and PIPA. We didn’t succeed there by making a moral argument, we succeeded by appealing to peoples’ fears that they would lose control of their lives and of resources valuable to them. Declaiming that “x is immoral” doesn’t work on the 93%; pointing out that “x will injure you” does.

    This is very superficial and inconsequential observation – if we can “scare” people into action on issues like PIPA/SOPA and consider it acceptable then we can do the same with free/proprietary software. There certainly isn’t a lack of things to scare people with in the proprietary world ;)

  186. >But by casting it as a question of morals, you are putting yourself in the position of dictating my choices, as surely as the fundamentalist Christian who decries dancing.

    HOW for God’s sakes? Write your own proprietary software if MS goes out of business, nobody’s stopping you … have I ever proposed to ban proprietary software?? I would fight along your side if ever somebody proposed that. I’m getting tired of your irrational fears and projections … I wouldn’t be so irritated by it if it wasn’t so common, some people start hyperventilating at the mere mention of people making decisions based on their morals – they feel threatened by it for some reason, maybe because they do not have any morals themselves and accept anything that makes money (or “economic sense” as they put it) … if that turns out to be baking Jews then let be it, as long as it makes money – they do not apply moral judgment anywhere it seems and quite frankly I’m sick of it.

    (sorry about the hyperbole about the Jews but I needed some super extreme example as it truly seems these people do not apply moral judgment ANYWHERE)

  187. Companies avoid infecting their software with the GPL because a) their lawyers tell them they can’t guarantee they can protect themselves and their competitive advantages in the GPL world, and b) they aren’t interested in becoming the next poster child for the Stallmanite zealots to attack just because they miss dotting an i or crossing a t and therefore don’t do everything in the Stallman-approved manner. Neither of these requires anything in the way of nefarious motives.

    Not least because the GPL and the FSF use a definition of what constitutes a derivative work that would be subject to the GPL using a definition that does not match the actual definitions used in copyright law. I wrote a paper on this topic some years ago when I was teaching CLE courses on Open Source software legal issues that I ought to update one of these days.

  188. Absolutely, if you do not mind backdoors in your software then I’m not talking to you … I’m talking to just people who do mind such backdoors such as esr. Case closed, have a nice day ;)

    The answer is “last week”. That was when I stopped beating my wife.

  189. >If you’re not proposing an actual solution other than going OSS that you would actually recommend then you’re not demonstrating anything

    “Solutions” rarely exist to real-world problems. Reality is too complex, with too many trade-offs and differing interests. This is the point Thomas Sowell repeatedly makes in nearly all of his books, as do many other economics writers. I just realized that this fallacy is probably why nearly everyone believes RMS is really a communist despite his claims to support business and freedom.

  190. >I just realized that this fallacy is probably why nearly everyone believes RMS is really a communist despite his claims to support business and freedom.

    I believe the reason behind that is most likely something completely different, namely that they’re simply reactionary idiots :) or people strategically applying scary labels to scare people off while knowing they have nothing to do with reality. RMS ideas are just as capitalistic as anything … some people badmouth it like that just because they do not like that particular style of capitalism. They like only the style where producer is in the charge, the type where the actual consumer is in charge scares the shit out of them …

  191. @libreman

    You do not disappoint! Keep it up. : )

    My problem with your GPL-solves-everything solution is that it doesn’t. Are you familiar with Reflections on Trusting Trust? The takeaway is that, unless I write, from the bare metal up, my own entire software build stack, I have little additional security with GPL code vs. BSD code vs. proprietary FrobCo code. Unless, of course, I _ASSUME_ nefarious intent from proprietary code, which is what you’re doing and which is flawed. But, even given that assumption, as esr logic’d out above you still have to prove GPL is the better solution than others.

    I haven’t seen that proof yet, just a lot of moralizing and (deliberate?) misunderstanding.

  192. @Cathy:

    You’ve got to be kidding. Patrick is one of best, most level-headed posters on this board.

    Thanks, but in fairness, I have been harsh on libreman, mostly because the way he conducts himself reminds me greatly of the worst attributes of every GPL advocate I’ve ever sparred with on the internet. Case in point:

    William B. Swift said:

    I just realized that [the fallacy of thinking you can fix things] is probably why nearly everyone believes RMS is really a communist despite his claims to support business and freedom.

    My internal translation of this is “that’s interesting. If I understand correctly, William says it’s a terminology problem, where ‘communist’ is the closest, most accurate label people can find in their own internal map for the consequences of the policies advocated by Stallman. That certainly maps well to my own experience.”

    But libreman says:

    I believe the reason behind that is most likely something completely different, namely that they’re simply reactionary idiots :) or people strategically applying scary labels to scare people off while knowing they have nothing to do with reality.

    So either I really misunderstood where William was going with this, or libreman really misunderstood, or libreman understood perfectly and decided to redirect. Any of those three possibilities, multiplied dozens of times over (to account for the number of messages and points in a typical conversation of this type), makes for a really tiresome exchange for all involved.

    Either I’m too stupid to ever be able to understand him, he’s too stupid to ever be able to understand me, or he understands perfectly, but wants to control the populace. I think it’s the third one of those, but it doesn’t really matter.

  193. @Jay Maynard:

    I’m arguing that proprietary software is bad for you as it can not be trusted to not work against your interests

    Stop right there. You don’t get to define my interests, either.

    Jay, you’re completely missing the point. You can program. This makes you dangerous. Just look on the news. How’s that for segueing back to the discussion about the meaning of the word “hacker”?

    The thing is, rms realized this before most governments or the general populace. You are not representative of those whose interests are to be protected. You are on the watch list, to be conscripted and to have your labor appropriated. This gets back to the discussion of whether rms ever used the word “evil” in a particular context. I assure you he did:


    Audience member: […] in this new World, and you’re talking about GPL going over to the next version, how do you see proprietary software businesses making a profit?

    Richard Stallman: That’s unethical, they shouldn’t be making any money. I hope to see all proprietary software wiped out. That’s what I aim for. That would be a World in which our freedom is respected. A proprietary program is a program that is not free. That is to say, a program that does respect the user’s essential rights. That’s evil. A proprietary program is part of a predatory scheme where people who don’t value their freedom are drawn into giving it up in order to gain some kind of practical convenience. And then once they’re there, it’s harder and harder to get out. Our goal is to rescue people from this.

    In an earlier post, libreman attempted to draw the distinction of proprietary software possibly being evil against the distinction of those who create the software being evil. Now you might think a bright third grader could attack this logic, but you have to remember that in libreman’s and rms’s ideal world, there is no such thing as entrepreneurship — control is exerted from the top.

    And this brings us full circle back to William’s observation about why people think rms is a communist.

  194. @Patrick
    > I think it’s the third one of those, but it doesn’t really matter.

    Fully agree. And it is either deliberate or exhibits the subconscious effects of drinking the FSF kool-aid.

    He reminds me of young religious advocates who have not been given enough training to spar with knowledgable heathens, and so rely on underhanded conversational trickery to try to look the better.

  195. @spqr:

    I wrote a paper on this topic some years ago when I was teaching CLE courses on Open Source software legal issues that I ought to update one of these days.

    I would love to see that. Another issue, which I have addressed both in comments on a much earlier blog post and in comments on comp.lang.python, is that the license (v3) in particular is violated regularly, and license violations are even encouraged. If you burn a Ubuntu CD for your friend, and don’t accompany it with a written offer of source, you have just violated the license…

  196. > The takeaway is that, unless I write, from the bare metal up, my own entire software build stack, I have little additional security with GPL code vs. BSD code vs. proprietary FrobCo code.

    But what about, “with enough eyeballs, all bugs are shallow.”?

  197. @Larry Yelnick

    They can have all the profits they want, so long as I’m not forced into Apple’s idea of an ecosystem down the road. Now take your mongering over to a post where it’s on-topic.

    Have I mentioned I have a Chrome killfile extension just for A&D? Guess who’s on it now. *PLONK*

  198. @Larry Yelnick:

    > I see your market share, and raise you profit share

    Perhaps you’ve missed the discussions about this. They only happen about 5 times a quarter, for the last half-dozen quarters.

    > But what about, “with enough eyeballs, all bugs are shallow.”?

    Since when is a deliberate backdoor a bug?

  199. @ Patrick Maupin
    >but you have to remember that in libreman’s and rms’s ideal world, there is no such thing as entrepreneurship — control is exerted from the top.
    And this brings us full circle back to William’s observation about why people think rms is a communist.

    If that refers back to rms:”they shouldn’t be making any money”, I think you have to read that as a reply to the question “how do you see proprietary software businesses making a profit?” — the (rms) answer being there shouldn’t be any proprietary software, and therefore also no businesses producing proprietary software and making money of proprietary software.
    It doesn’t say rms or FSF have a problem with businesses making money of Free Software, or of baking bread for that matter.

  200. @kn
    > It doesn’t say rms or FSF have a problem with businesses making money of Free Software, or of baking bread for that matter.

    Ehh, including ‘baking bread’ there isn’t so good. If you stick with that, the rms ideal would be that no bread baking business should ever have trade secret recipes.

  201. @kn:

    Presumably rms would have a problem with people baking bread using a secret recipe. In any case, you just made my point for me.

  202. >Ehh, including ‘baking bread’ there isn’t so good. If you stick with that, the rms ideal would be that no bread baking business should ever have trade secret recipes.

    The point being that in the answer and reply Patrick Maupin quoted, rms doesn’t talk about anything but making money of proprietary software so his (Patrick Maupin’s) conclusions about anti-entrepreneurship and communism are a non sequitur.

    >trade secret recipies
    >LOL at near-simultaneous replies…

    I don’t presume to know what rms thinks anout making bread or running a bakery.
    I do wonder how this trade secret recipe analogy would apply to Free Software and at the same time not apply to Open Source Software.

  203. @kn
    > I do wonder how this trade secret recipe analogy would apply to Free Software and at the same time not
    > apply to Open Source Software.

    Ideology thing. FSF sez “NO, it is immoral/evil!” OSS sez “Well OK, in THIS situation you might be better doing that, but maybe going open is the better business strategy overall, just sayin’.”

  204. “Have I mentioned I have a Chrome killfile extension just for A&D?”

    I’d like to hear more about this! A&D is a great place, but it really needs killfiles.

  205. OK, the “conversation” degraded into full circle-jerk and shouting of slogans instead of actually addressing the logic I’ve presented of why I believe GPL is the right solution to the problems I’ve outlined.

    To escape that logic you either have to say that you do not care about backdoors, DRM and snooping as Jay Maynard has done (which I suspect he actually does but doesn’t like the inevitable conclusion that realization leads to so just says he doesn’t to avoid conceding that) or you would need to propose a better way of solving those problems in proprietary world which was not proposed by anyone so far.

    The only attempt at that was by esr who proposed a series of government bans but it turns out he wasn’t really serious about that (fortunately as it would be horrendous idea) so we’re at square one and there is NO proposed alternative solution. Instead of addressing that logic people ponder about why some call rms a “communist”, call me names or ascribe all kind of weird labels and personal traits to me.

    This is why I do not often join conversations on the Internet, there is just not a penalty for being wrong or the abuse of logic. I could spend hours demonstrating why a slogan like “No one who supports copyleft is supporting freedom. Copyleft is preemptively coercive” is spectacularly wrong but after those several hours the person would just repeat it and that’s it, wasted time. The positions of people able to support their statements and those just throwing slogans are asymetrical – it takes a lot of effort to substaintiate a statement but it takes few seconds to throw a slogan. This is very frustrating, I’m sure esr knows what I’m talking about.

    I joined the conversation on this site hoping to avoid the kind of nit-picking, irrelevant non-sense and ad-hominem attacks the Internet is famous for hoping to find a group of people that are able to focus on the actual arguments and attack the logic. I guess the Internet is still the Internet even in this corner, the only one who was closest to my expectation of addressing the actual logic is esr so I hope to hear more from him, there is a chance of having a reasonable discussion.

  206. Patrick Maupin Says:

    > Perhaps you’ve missed the discussions about this. They only happen about 5 times a quarter, for the last half-dozen quarters.

    Right, but back then Apple wasn’t taking over 50% of the available profit.
    It takes profit (and cash flow) to run a business, Mr. Maupin.

    > Since when is a deliberate backdoor a bug?

    Are you asserting that it isn’t? Ken thought it was.

    (Enjoying the cold snap in Austin today?)

  207. I do wonder how this trade secret recipe analogy would apply to Free Software and at the same time not apply to Open Source Software.

    OSS says that the closer you move towards openness the more efficiency you get. Under certain circumstances it makes more sense to open source generic components and then add your own “trade secret special sauce” on top (e.g. open source the standard double entry book keeping functions, add your special accounting formulas on top).

    So to extend the analogy, by open sourcing the bread recipe you get feedback as to how to improve your bread recipe and thus end up with better bread than you would have without open sourcing the recipe. Sure you might do Bread with your special herbs and spices on top whereas the group down the road do the same bread but with their own secret cheese and bacon topping. And both are all good. Some other guy does this bread wagon thing that only does plain bread but he charges because he brings it too you.

  208. “The actual bug I planted in the compiler would match code in the UNIX “login” command. ”

    — Ken Thompson in “Reflections on Trusting Trust”, his ACM Turing Award lecture. It’s about hacking, and hackers. Perhaps you (and others here) should read or re-read it. Thompson has a far more legitimate claim to leader of the hackers (if, indeed, there could ever be such a leader) than anyone here.

    1. >Thompson has a far more legitimate claim to leader of the hackers (if, indeed, there could ever be such a leader) than anyone here.

      Well, duh!.

      I’ve had friendly contact with ken (he helped with The Art Of Unix Programming). He’s got the claim all right, but neither any interest in leadership nor the personality wiring to do it.

      But, as I’ve written elsewhere, we don’t really need leaders in the way people normally think of that role. Trusted elders and spokespeople, yes, but not leaders. There was a time, back in the early days of my fame, when a lot of people seemed to want to make me a leader – patterning, perhaps, on the role RMS has tried to occupy. I dodged that; founding OSI was a way to get off the white horse.

      I suppose, as long as human beings are human beings, we’ll have an instinctive tendency to try to figure out who the Big Man is – even when we’re operating in social forms for which it doesn’t really make sense. Sigh.

  209. Patrick wrote “Since when is a deliberate backdoor a bug?”

    In any community where people are concerned about deliberate backdoors, places where such backdoors can lurk are at least borderline bugs. I’m not sure how practical it would be to try to avoid such bugs with today’s CS tech, but I expect it will become increasingly practical as techniques for providing machine-checkable proofs of correctness improve. Of course, no matter what you prove it’s fundamentally hard to avoid the lowest-level security problems — e.g., has the hardware been tampered with since the last time you inspected it? But many adversaries don’t have the technical capability to strike such low blows, so I think correctness proofs of a software toolchain would provide quite a lot of real security.

  210. @libreman

    You’re making my day, dude.

    Recommendation: try actually arguing with the points some of us are bringing up instead of throwing a fit when we do, exercise a little bit better critical reading skills, and try to do all that while addressing what we’re actually saying and not what you think we’re saying. Slow down, take a deep breath, and engage. Even the most hardheaded of the regulars here can often surprise you.

  211. @Cathy
    > I’d like to hear more about this! A&D is a great place, but it really needs killfiles.

    Right now it’s a very basic script that just does a regexp check. It replaces the whole comment block (username and all) with a “Comment Killed!” line. To add or remove someone you have to edit the .js file and reload the extension. I have plans to turn it into a more featureful thing with page-embedded controls, but I’m not sure when I’ll be able to tackle that. If you don’t mind the minor code twiddling I will put it up on my site, otherwise I’ll see about getting something a little nicer built.

  212. To escape that logic you either have to say that you do not care about backdoors, DRM and snooping as Jay Maynard has done

    I do care. I simply don’t think that a user must be forced to rule out the possibility that they are present at the cost of destroying his freedom to choose software that matches his needs better than the soi-disant “free software” world can provide.

    And I say this as an experienced Linux user, both on the desktop and on servers. I run OS X because it fits my actual needs – MY needs, not the needs you think I have – far better than Linux does.

    (which I suspect he actually does but doesn’t like the inevitable conclusion that realization leads to so just says he doesn’t to avoid conceding that)

    It’s a continuum, not black and white. Eric has destroyed your argument, but I’ll rephrase his refutation, just in case this version gets through your Stallmanite blindness better: In order for your argument to succeed, you must demonstrate there is no other alternative that will get rid of them at a cost to the clay of the earth…you know, users…lower than the backdoors themselves. The burden of proof lies with you, not Eric.

  213. liberman,

    > I could spend hours demonstrating

    Trying to demonstrate. What this usually comes down to is different priorities. Where I work avoiding copyleft lawsuits is best done by avoiding copyleft, and avoiding backdoors is best done by running good anti-threat software. Copyleft is nothing but a tyrannical pain in the ass that makes it harder to do my work. Truly free licenses, OTOH, make my life more free. I would never release under the GPL.

    Yours,
    Tom

  214. @kn:

    The point being that in the answer and reply Patrick Maupin quoted, rms doesn’t talk about anything but making money of proprietary software so his (Patrick Maupin’s) conclusions about anti-entrepreneurship and communism are a non sequitur.

    No, my conclusions are not a non-sequitur. I opined that enforcing lack of proprietary software requires control from the top. As William pointed out, this is the sort of control that is, in many minds, associated with communism. Nowhere did I say this was communism. There are many systems besides communism where the fruits of labor are stolen from the laborers.

    @Larry Yelnick:

    > Right, but back then Apple wasn’t taking over 50% of the available profit.

    This has been true for awhile, and has been discussed.

    > It takes profit (and cash flow) to run a business, Mr. Maupin.

    This has been discussed also. Do you have anything new to add? (For that, you’d have to go back and reread some of the discussions…)

  215. @Larry Yelnick:

    > Are you asserting that it isn’t? Ken thought it was.

    Are you sure he wasn’t just reaching for the closest reasonable word? FWIW, my brother (who works for the highway department) says derisively of the city putting in speed bumps “the state would never deliberately put defects in a roadbase.” I used to hate speedbumps with a passion, but now I’ve learned to accept at least one positive — if you’re on a road with speedbumps, you’re fairly well assured that it goes somewhere, otherwise they wouldn’t bother.

    > (Enjoying the cold snap in Austin today?)

    Yeah, I was starting to worry that global warming was here. 50 is more like it. 80 was a bit high…

  216. @William Newman:

    > so I think correctness proofs of a software toolchain would provide quite a lot of real security.

    Right. We can do that with lots of levels of chip logic, but a whole toolchain would be a harder proposition. OTOH, unlike libreman, I don’t believe that the odds of finding a CarrierIQ-like substance are really that small. The odds do get smaller as memory sizes and CPU speeds go up, but security research seems to attract some really bright people.

  217. >It’s a continuum, not black and white.

    Yes, it’s a continuum – exactly. And the threshold where the negatives of proprietary software outweigh its benefits compared to Free Software is subjective to every person. That’s a good point to bring up as that wasn’t mentioned so far. So when I say that “you do not care” I do not mean that in an absolute sense, I mean that in a sense that your personal threshold was not breached so you do not care *enough* to do something about it radical enough like switching to Free Software – so I get that. For you the benefits of closed source outweigh its negatives.

    >Eric has destroyed your argument, but I’ll rephrase his refutation, just in case this version gets through your Stallmanite blindness better:

    Unless I misunderstood him, this is not what he said at all … he was arguing that open-source is not strictly necessary to solve those problems if there are in principle any other solutions (no matter how impractical or unfeasible apparently as he said nothing about that, also nobody was arguing about absolute strict necessity of open-source in the first place so I do not know whose point he was refuting anyway – certainly not mine or FSFs)

    > In order for your argument to succeed, you must demonstrate there is no other alternative that will get rid of them at a cost to the clay of the earth…you know, users…lower than the backdoors themselves. The burden of proof lies with you, not Eric.

    So in essence you want me to prove a negative? Are you aware that this is a fallacy No. 1 in science and logic in general? That’s like saying that unless you can prove that there is no God then there is one by default and the burden of proof lies with you to show that there is none if you claim that. That’s clearly nonsense … if someone claims there is a better alternative than open source it’s them who need to demonstrate what this alternative is.

    By the way Eric does not claim that there is a better alternative apparently. He does believe open source is the solution, he just refrains from stating that proprietary software is therefore bad/immoral etc. which is however the inescapable implication if you believe putting backdoors in software and not giving users the option to disable them is immoral which he said he does. All of the prerequisites necessary for calling proprietary software immoral are there but he refrains from doing so maybe because he doesn’t like that conclusion. So he simply ignores the logic that leads to it.

    If you can explain to me where you see a mistake in the reasoning above I’ll be thankful. But keep in mind that it applies to Erics convictions, not yours – I know that for you the threshold is not enough to care about OSS and you prefer convenience before security, that is ok but Eric said he finds backdoors, DRM and such morally objectionable and speaks out against it so if you follow the above logic you need to do the same for proprietary software as a category as it enables said immoral practices and there is no other feasible solution to that than OSS.

  218. @libreman

    esr has already explained how your logic is sloppy. Please re-read his two posts going over it.

    Also, you’re failing to explain how proprietary necessitates back doors, and you’re arguing in a way that rather falsely equivocates the two. That is, sloppy.

    I won’t explain where your other mistakes in reasoning are, because you are constantly choosing your own meanings for what people say or cherry-picking phrases to suit your argument rather than arguing against what people are actually saying. Way too many of your premises are invalid or non-sequitur to make it worth going point-by-point.

  219. @libreman:

    All of the prerequisites necessary for calling proprietary software immoral are there but he refrains from doing so maybe because he doesn’t like that conclusion.

    No, they are not there. In fact, if you paid attention, you would find edge cases where esr agrees proprietary software may be better.

    Others have already demolished your “backdoor” argument, but you apparently haven’t noticed. There is nothing in the GPL that prevents back doors. A gazillion GPLed routers could all have “back doors” based on a proprietary root password, the TiVo could have a back door, etc.

    There is a tenuous moral argument to be made that copyright as practiced today goes too far, even absent the serious RIAA shenanigans. A slight possibility that if you hear a catchy tune and it invades your soul, you have some sort of right to it; at least to acquire a copy at reasonable cost. Maybe even a remote chance that if you spend a lot of time learning a piece of software and make it part of you, you ought to be able to use it on the next job without spending thousands of dollars.

    But just like you have no moral right to make the band that produced the song explain to you exactly how they achieved those dulcet tones, you also have no moral right to make the programmer explain to you exactly how he achieves his results.

    Is it smarter to use OSS? Almost certainly, for long-term strategy reasons. It can certainly be more convenient, tactical, to use non-OSS. Is the GPL required to use OSS? Not at all, as others have explained.

    Even if you truly believe that non-OSS software is evil, the GPL is your enemy. The GPL, like the FSF and its creator rms, brooks no compromise. As Tom DeGisi points out, it uses unforgiving copyright law to itself be unforgiving. It allows no baby steps towards open source. It’s the GPL way or the highway, baby. In some situations, you can’t gradually integrate it into your processes at all. Even worse, if someone drank some of the kool-aide but not all of it, and made some GPL2 stuff, now you can’t even link it with GPL3 stuff that someone else made. You have to rewrite something. And when you rewrite it, are you going to say “man, that Stallman is a genius! I should make sure to put the ‘or later clause here’ so this doesn’t bite me again” or are you going to say “that cocksucker Stallman fucked me over again and I’m going to share this replacement code under a permissive license so there’s one less piece of pain for others in the future!”

    I strongly suspect we can each determine which statement comes closest to the sentiment the other would have.

  220. @Tom DeGisi:

    I assume you mean the Virgin Mobile plan?

    We’re on Virgin Mobile now (voice only) and I’ve been looking at that, but I’m not sure which way we’ll jump. My preference would be to separate the device from the plan, but at $149 or $199, it’s not too bad.

  221. @tom

    Truly free licenses, OTOH, make my life more free. I would never release under the GPL.

    I would. GPL is a wonderful business tool in between proprietary and useful for everyone open source. As open source, it’s sucks balls in comparison to Apache. But just look at how much money MySQL AB made exploiting GPL fears. “Well sir, it is probably best if you had a corporate license for that purpose. It’s still a lot cheaper than Oracle and you’re safe from any IP violations…”

    It’s just really funny how rms has managed to redefine less free is actually more free. FSF newspeak sometimes irritates the hell out of me but mostly makes me laugh.

    But from a business perspective it’s a way of executing “open source” without actually giving your competitors anything. I bet Google wished they had a low cost proprietary license for Android and GPL V3 combo vs Apache after Amazon forked Android. The CM folks would still love them and Amazon would be SOL.

    GPL is no better than any of MS’ shared source licenses and Eric if OSI caved on anything it was caving on OSD 9. In my opinion, if you were going to draw the line anyway against non-commercial/academic licenses you should have drawn the line at weak-copyleft as being the most restrictive form that still qualified as open source. Meaning LGPL would have been OSI approved and GPL not.

    You know, if I ever won the lottery and had a couple millions to burn on legal fees for grins I’d aggregate a GPL work and a non-GPL work via linking and dare the SFLC to sue me. Then put to rest the notion that creation of a derivative work has zip all to do with linking, statically or otherwise.

  222. @jsk

    You’re entertaining, and a nice counterpoint to the Apple dudes we usually get. I especially like how you claim you aren’t a zealot, then proceed to fake or exhibit ignorance of anything beyond an FSF worldview, immediately try to re-frame the topic to match that worldview, and throw a minor hissyfit when people refuse to acknowledge it. Good times. Keep posting and giving me a chuckle.

    My $0.02, from my perspective your idea of ‘freedom’ is one that removes my own in order to combat some amorphous maybe-threat. If I could easily, fully expunge GNU’s overwrought crap from my linux boxen but keep the supports-everything kernel, I would. And please, don’t try the whole ‘if it weren’t for GNU there wouldn’t be Linux’ argument. It assumes too much.

    Well, thanks to Apple (oh noes!) you might sorta have that option GNU free option someday on one of the BSDs. I recall one of the sticking points of a totally free (heh…see what I did there?) BSD distro with GPL taint was pretty much stymied without including gcc. The userland and pretty much everything else had a free and open implementation. Non-gpl oss compilers tended toward suck. Until LLVM/Clang…

    Looks like they are pretty close:

    http://wiki.freebsd.org/GPLinBase

    But really, linux just works better than BSD on the desktop (that’s not saying much at all really). If you’re going to run BSD you might as well run the best of breed: OS X. It’s proprietary but verrra nice.

    Did I just reframe your request into the Apple worldview? :)

  223. Patrick Maupin,

    One was Virgin Mobile, the other was Boost Mobile. As far as I know the devices are no contract devices. Not sure what you mean, separate the device from the plan.

    libreman,

    Another way to put it is that copyleft, by being coercive, is the opposite of cooperative. I like OSS because I like cooperating with other developers, not forcing them to do things my ways. I find copyleft philosophically and ethically ugly.

    Yours,
    Tom

  224. Nigel:

    if I ever won the lottery and had a couple millions to burn on legal fees for grins I’d aggregate a GPL work and a non-GPL work via linking and dare the SFLC to sue me.

    This. A good starting step to write a useful piece of software that requires some GPLed software (preferably busybox, to attract the busybodies), ship it under a license that disallows it being placed under the GPL (but allows linking to GPLed code) and with a script that will automagically link it on the user’s computer, and watch the fireworks begin. Does your code incorporate busybox? Did you ship busybox? Contributory infringement, you say? But I thought the GPL allowed one who was in receipt of busybox to have their way with it — no direct infringement, no contributory infringement.

    Everybody marvels at what a great hack rms made with the gpl, but the problem is, as always, the edge cases. Copyright law won’t do what he really wants it to, but bluff and bluster, combined with the fact that none of the edge cases are valuable enough to challenge, makes the whole thing work.

  225. @NIgel:

    > Did I just reframe your request into the Apple worldview? :)

    In my case, no. Apple’s always been good at recognizing and nourishing goodness it can use, whether by buying a company, or simply buying a couple of developers. They don’t suffer from NIH and don’t usually suffer (rounded rectangles notwithstanding) from misapprehension about where their competitive advantages lie, and they realize that some developers (CUPS, LLVM, etc.) are motivated by more than just money.

  226. @Nigel:

    > I would. GPL is a wonderful business tool in between proprietary and useful…

    Yes, it’s a great advertising license, when you plan on making all your real money by selling proprietary licenses on the side. Of course, for that to work, you have to invest a lot of development time to keep a fork from occurring, or you lose the advertising edge, because all of a sudden your actual product is out of date.

  227. @Nigel
    > Did I just reframe your request into the Apple worldview? :)

    No, and it wasn’t really a request. Besides, Apple only makes OS X to support a fairly limited hardware set; I don’t see any of the more open BSDs gaining much from that.

    I like that so far Linux kernel has supported every piece of hardware I’ve thrown at it, with the exception of this one scanner a friend has. I actually happen to use OS X, too. as my art-making system, and not on Apple hardware. It took about two weeks of tinkering to get the right set of kexts. It works nice with my Cintiq. I would loathe having to actually develop on it. That’s what my linux boxen are for.

  228. They’d HUGE success, it’s quite possible Linux wouldn’t be even here without rms and GNU, the internet would be completely different place without the freedom it provided to its infrastructure.

    That’s so much revisionist bullshit. If Linux didn’t exist there’d be 100 different BSD distros running around today instead of linux ones.

    If the AT&T/bsdi lawsuit hadn’t happen then Linus wouldn’t have even needed to bother and contrary to popular belief the internet doesn’t just run on linux boxes. There’s a bunch of bsd servers out there too and they are actually free as opposed to GPL’d.

    This is the sort of FSF partisan BS that annoys me. Deliberate or intentional ignorance of hacker history so that open source began fully formed from RMS’ brow with GPL. It reminds me of the BS you see in North Korea and how Kim Il-Jong was an expert in everything from agriculture to zoology and invented the multi-camera shoot.

    If RMS and GPL didn’t exist we’d have a far saner open source community and exactly the same sort of internet we have today.

    1. >This is the sort of FSF partisan BS that annoys me. Deliberate or intentional ignorance of hacker history so that open source began fully formed from RMS’ brow with GPL.

      Indeed. Especially annoying to people like myself who can remember the hacker community before the FSF.

  229. @esr I actually came back to reply to your NC post. Yes, I agree that the NC clause is possibly a legal time bomb ticking away in the CC world.

    On the other hand I like the intent and honestly there are so many potential IP time bombs scattered across the legal landscape that one more probably doesn’t make any difference.

    From an intent perspective it strikes me as more healthy (or at least honest) than the GPL copyleft intent. It’s intent is “Hey, this is out there for friendly reuse…not for companies to make a buck off my stuff”.

    I use CC-BY-SA-NC on my soccer kids photos because it includes the images of other people’s kids as well as my own (kinda hard not to). I’d really hate to it to show up on some billboard or ad without some legal recourse if some other parent became irate with me.

    Me, I like inclusive communities. Proprietary, open should all be welcome. Apple, FreeBSD, Linux, Windows, iOS, Android, WP…it’s all good. There are hackers in all of those communities and they are more alike than different. I’m as happy jamming in .NET code camp as I am in a open source hackathon. Neither one in the last few years but still…

    RMS and the FSF promotes exclusive communities where what I do for a living is evil and immoral and he’d be happier if my kids were poor and I was flipping burgers instead of making a decent living as a dev. The FSF worldview is predicated on fear and hatred of others and it’s soooo not part of the hacker ethos in my opinion.

    I’m sure you disagree with the hatred thing but what is the normal emotion for evil things?

  230. Certainly, the MySQL example alone refutes everything libreman believes. The cognitive dissonancy alone should be causing local earthquakes.

    And it would be amusing to see a good case on the GPL get into a circuit that actually understands the law of derivative works and shreds the GPL language. Probably the Seventh Circuit with Posner would be the one. God forbid the Ninth Circuit as they’d make an incoherent hash of it.

  231. @SPQR fanaticism in anything inoculates you from cognitive dissonance.

    I’m sure you guys think I’m an apple fanatic but really, I just like it cause it’s shiny. Either the Firefly or cat distracting sense of shiny.

  232. @Tom DeGisi:

    As far as I know the devices are no contract devices.

    Correct, but they are locked, and if you wanted a different Android device they didn’t supply, you’re out of luck.

    Not sure what you mean, separate the device from the plan.

    There are technical reasons (frequencies, modulation) why some devices can’t work on some networks, but increasingly the reasons are purely business. For example, you can’t take your Virgin Mobile device and use it on a Sprint reseller such as platinumtel.

    Even though you “bought” the handset, you don’t “own” it, and the very act of “buying” it means you accept their contract:


    The Terms of Service become effective when you accept them by doing any of the following: (1) purchasing or activating a Virgin Mobile or payLo phone or device.

    Our mobile phones and devices may not be purchased in bulk or sold to third parties.

    It would be laughable except for the success tracfone has had putting people in jail for buying and reselling subsidized handsets.

  233. @jsk:

    I like that so far Linux kernel has supported every piece of hardware I’ve thrown at it, with the exception of this one scanner a friend has.

    Must be a Canon. I’ve got an old windows box around just to drive the damn thing. Not hooked to the network, though…

    @SPQR:

    Certainly, the MySQL example alone refutes everything libreman believes.

    Yeah, I’ve even seen it quoted that rms loves things like MySQL where proprietary people fund GPLed software development. Whenever I point it out that this is inconsistent with the worldview that proprietary software is evil, I’m told it’s the lesser of two evils.

    @Nigel:

    I’m sure we agree on lots of things, possibly even most things. Most of those aren’t usually up for discussion here :-)

  234. >> RMS and the FSF promotes exclusive communities where what I do for a living is evil and immoral and he’d be happier if my kids were poor and I was flipping burgers instead of making a decent living as a dev.

    We don’t have the right to make a living as computer programmers, artists, or musicians. Even if we were to grant that copyright provide incentive for the production of computer programs or musics, it does not follow that we should continue to fund the production of such goods.

    Even if we stop producing musics and artworks, humanity will have access to the largest library of work ever, unencumbered by the consideration of copyright law. We will also never enjoy them all, at least at our level of ability and lifespan. When we die, we will certainly never read most books in the world and we will never to remember most of faint details.

    If there is still commercial interest in producing computer programs, there will still be work even if everything is open source. If there’s not, we would need to find something productive to do, like curing aging and fund immortality research projects.

    In the end, RMS was never against “making a living”, but making a living in a way that contribute to evilness.

    I don’t necessarily agree with RMS, but I do not hold “making a decent living” as scarce as my long term interest. What we really should aim for is to never have to work at all just so we can simply exists. It’s a world where we can just idly program interesting things forever and make art forever, or enjoy them as audience, forever.

  235. @GPL & FSF
    This discussion does remind me of this argument that ripped apart the protestant churches in my country in the early 20th century. Villages and families were shattered by this fundamental theological rift: Did the snake speak in human language to Eva or not. Obviously those who did (not) believe this were in bed with the Devil!

    Those supporting the GPL are as much communists as were the abolitionists. Replace “Software” with “Slavery”, and you get the arguments used in the 19th century to fight or support slavery. As RMS has not tired of explaining, neither he nor the FSF extrapolates from Software to any other branch of the economy. Not even to music, books, or films. And certainly not to bakeries.

    And in the example of the “secret recipe” of bread, replace bread with chemotherapy, and then come back again.

    The FSF is targeting software because software is vital to modern life and freedom in general.

    (full disclosure, I support the GPL and our host and many regulars are convinced I am a communist)

  236. >No, they are not there. In fact, if you paid attention, you would find edge cases where esr agrees proprietary software may be better.

    It doesn’t suffice to procliam that “No, they are not there” without proper refutation of why you believe they are not there. Or are you saying that I should just take your word for it? Would you consider that a valid argumentation if I did that? So please, if you say they’re not there, take the time to demonstrate that and explain why.

    I agree that proprietary software may be better too, in individual instances. I think this is where the misunderstanding lies. I DO NOT claim that EVERY piece of proprietary software is bad as many seem to be interpreting it, not at all. I claim that IN AGGREGATE proprietary software is bad because things like backdoors, DRM, snooping manifest statistically much more often in it than in OSS (notice that I do not claim they do not appear at all in OSS either). So it’s not me who claims that, people just tend to caricature my position and then ascribe that extreme points of view to me instead of trying to understand what it is that I’m actually saying. They take every opportunuty to misunderstand me – that’s not a way to have a meaningful discussion. This is how we are wired, sometimes I do it too, I try not to and when I catch myself I try to rectify it so please, try it too.

    >Others have already demolished your “backdoor” argument, but you apparently haven’t noticed. There is nothing in the GPL that prevents back doors.

    Again, you just state things. Also, yes nothing in the GPL explicitly prevents backdoors, that’s quite obvious but it has the EFFECT of preventing them. You’re talking about what could have happened, could be, maybe, hypothetically … I’m not talking about some virtual hypothetical universe, I’m talking about the universe we actually live in and the data is clear, no matter what could, maybe, possibly .. it DOES prevent backdoors, DRM and snooping – at least in our universe. If you want to talk about some other one then I’m not interested.

  237. @ Tom DeGisi
    > No one who supports copyleft is supporting freedom. Copyleft is preemptively coercive.

    While I am favourably disposed to Eric’s open source approach (among other things because of its practical utility) statements like this bug me.

    If a developer sees a piece of GPL code s/he has a choice whether s/he wishes to use that code. If the choice is not to use the code, no issue of “pre-emptive” coercion can arise.

  238. > >This is the sort of FSF partisan BS that annoys me. Deliberate or intentional ignorance of hacker history so that open source began fully formed from RMS’ brow with GPL.

    Didn’t Thompson and Ritchie just get the Japan Prize last Summer, in recognition of having helped kick the whole thing off?

    Hey, Android owners: http://i.imgur.com/lFxdi.jpg

  239. Thompson and Ritchie didn’t kick the open source thing off. That’s as old as computing. To the best of my knowledge, the first open source, community-developed OS was released in 1959: the SHARE Operating System for the IBM 704, itself a development of the first OS. SHARE was four years old at that point.

    1. >To the best of my knowledge, the first open source, community-developed OS was released in 1959

      The CTSS/Multics people also have a claim. But I think theirs and yours are both dubious. I would say Multics and SOS approximated the traits of open source as closely as the technology of the day permitted, but for a variety of reasons that wasn’t very close and applying the label is more than a little misleading.

      For SOS’s claim, in particular, it’s a problem that all programs had to have wired-in knowledge of the physical configuration of the host computer. This meant that program portability across installations wasn’t possible. The Multics people pioneered device-independence abstractions that partially broke that link, but it took Unix to actually sever it.

  240. “But, as I’ve written elsewhere, we don’t really need leaders in the way people normally think of that role. Trusted elders and spokespeople, yes, but not leaders. ”

    This is true not just for open source but for freedom oriented movements in general. Free people do not need leaders.

  241. TomM,

    If a developer wants to use a piece of copylefted code on a proprietary project the developer is forced, before the fact, by copyleft not to use it.

    Forced before the fact = preemptively coercive.

    QED. The FSF should rename itself the Anti-proprietary Software Foundation, because it is not supporting freedom for what real people actually do.

    Here’s a disturbing fact. Most software written today doesn’t have sufficient utility or sufficient craftsmanship to be released as open source, nor would anyone benefit economically from the effort taken to do so. Most software is written for one company. It is used by tens, or hundreds, or thousands or employees of that company, not the general public. The hackers (or closest too it) in that company are already working on the software. Most of them don’t use it, and don’t want to use it. Furthermore, most such software contains a large number of temporary advantages which the company wants to keep private so their competitors can’t use them against them, as well as a possibly even larger number of much longer lasting disadvantages which the company really wants to keep private so their competitors can’t use them against them. Historical kruft is extensive, expansive and expensive to get rid of.

    Stallman’s neat legal hack really is neat, and beautiful. And it really is ugly, too.

    Yours,
    Tom

  242. That seems more like a problem with the definition of “operating system” and less like a problem with the definition of “open source”, to me.

    In any case, the point is that Thompson and Ritchie weren’t blazing a trail as far as being open source goes, and RMS most certainly was not. “SHARE. It’s not an acronym. It’s what we do.” Where was RMS in 1955?

    1. >Also, don’t forget that one major purpose for OS/360 JCL was to provide device-independent I/O…

      Well, except it didn’t. For example: how many blocking factors and other impedimenta did you have to specify for a disk operation, again?

      >In any case, the point is that Thompson and Ritchie weren’t blazing a trail as far as being open source goes

      No, I think they were. SOS and Multics pointed in the right direction – you can see the philosophy of open source being articulated pretty clearly in the writing on Multics – but not until Unix was software decoupled from hardware enough that (for example) cross-development was even possible. Without that, the intent of open source could exist, but not the social machine within which “many eyeballs” would be a significant phenomenon. The technology just wouldn’t support it.

  243. It is amusing to be lectured about the early days. While not quite as foundational as some here, I was around writing software in the ’80’s. Gave some donations to RMS in those days – sums that seem small today but felt larger back then. Was doing some jackleg admin on a VAX from tapes from Berkeley. Even contributed some bug patches back into thinks like bison.

  244. If you were doing card-image I/O, you could use the same program for reading from card, disk (any disk), or tape, using JCL to connect the data to the program. Yeah, it took a fair amount of specifying in JCL, but the program needed no alteration at all.

    No, it’s not device independence in the mold of Unix, but it was also designed 8 years earlier.

    As for the many eyeballs phenomenon, that the audience is limited by the cost of the environment – a 704 was a lot more expensive in its day than a PDP-11 was in its – is no more dispositive than, for example, a limited user domain is for many open source projects today. Is gpsd any less open source because a limited number of people have the understanding to hack on it?

    In any case, whether SHARE or Ken Thompson led the way, the key is that it was well underway long before RMS decided to start his jihad.

    1. >As for the many eyeballs phenomenon, that the audience is limited by the cost of the environment – a 704 was a lot more expensive in its day than a PDP-11 was in its – is no more dispositive than, for example, a limited user domain is for many open source projects today. Is gpsd any less open source because a limited number of people have the understanding to hack on it?

      You’re arguing causally, I’m arguing consequentially. I’m saying it’s misleading to call SOS or Multics “open source” because (to a person not quite familiar with the way computing was done back then) the “open source” label will imply a lot of behaviors that nobody engaged in. There weren’t any wide-area networks, there weren’t any archive sites, there weren’t any forges, and most importantly the expectations about development practice that those technologies support weren’t yet present. You didn’t have people checking distributions out of repositories to hack them and ship back patches. They couldn’t do that; the infrastructure didn’t exist.

      You say “the audience is limited by the cost of the environment”. Sure. So what? I’m not saying 704 developers were wrong or stupid for not doing the things we do today, just that they couldn’t do them. That’s why I think the label “open source” applied that retrospectively is misleading – it encourages people to back-project today’s behaviors and expectations to a time when they not only were not practised but would have been impossible.

  245. > Thompson and Ritchie didn’t kick the open source thing off.

    I said “>>helped<<", you took it out, Jay.

    In 1975, the sixth edition UNIX along with the source code was distributed to universities and research institutes. It was extensively studied as a practical example of an operating system and contributed greatly in the field of Information Technology as a whole. Coincidentally, it gave rise to a culture of sharing improved and enhanced source codes amongst UNIX users, and shaped the beginning of today’s Open Source culture.

    cite: http://www.japanprize.jp/en/prize_past_2011_prize01.html

    > Without that, the intent of open source could exist, but not the social machine within which “many eyeballs” would be a significant phenomenon. The technology just wouldn’t support it.

    We had tape. In particular, if you don’t remember the DECUS tapes, well, you weren’t there. The first Berkeley Software Distribution (1BSD) tape was released on March 9, 1978.

    There was also “Lions’ Commentary on UNIX 6th Edition, with Source Code” by John Lions, published in 1976 or 1977. When AT&T announced Unix Version 7 at USENIX in June 1979, the academic/research license no longer automatically permitted classroom use. Thus, licensees were no longer able to use the Lions notes for classes on operating systems.

    However, thousands around the world spread the book via photocopy. Many pioneers of Unix and open source had a treasured multiple-generation photocopy. I got mine in 1980, and still have it.

    1. >In particular, if you don’t remember the DECUS tapes, well, you weren’t there.

      I was there. It’s even chronologically possible that some of my very earliest code found its way onto one of the last DECUS tapes, though I don’t know that it ever happened. I was on the Unix side of things and of course the DECUS tapes were mostly aimed at VMS users.

      It’s precisely because I was there that I sometimes feel a need to debunk romanticism about the old days. “We had tape” is really a pretty weak answer when the implied question is actually “what was your communication-channel equivalent of pulling from a git/svn/CVS repo”.

  246. > In any case, whether SHARE or Ken Thompson led the way, the key is that it was well underway long before RMS decided to start his jihad.

    RMS never claimed differently, either.

    When I started working at the MIT Artificial Intelligence Lab in 1971, I became part of a software-sharing community that had existed for many years. Sharing of software was not limited to our particular community; it is as old as computers, just as sharing of recipes is as old as cooking. But we did it more than most.

    cite: http://www.gnu.org/gnu/thegnuproject.html

    Project GNU was the follow-on response to Symbolics’ destruction of the AI Lab. (The first was to re-implement all of the Symbolics changes on the ITS machines at MIT, and then give these to LMI.)

    But go ahead and re-create history as you see fit, so you can complain about it, Jay. It just proves over, and over again that you weren’t there, and you’re prone to barking.

  247. >If RMS and GPL didn’t exist we’d have a far saner open source community and exactly the same sort of internet we have today.

    except it wouldn’t be called “open source”.

  248. > It’s precisely because I was there that I sometimes feel a need to debunk romanticism about the old days. […] DECUS tapes were mostly aimed at VMS users.

    LOL. AYFKM? Your resume claims that you had access to a TOPS-10 machine, and yet you make this claim?

    > It’s even chronologically possible

    Dude, this is a weak-ass claim.

    “it’s even chronologically possible” that you sired several of the people who post here, but still improbable.

  249. When I have lectured about Open Source legal issues in the past, for a legal CLE audience, I usually date its roots with the ATT ver 6 & ver 7 / Berkeley collaborations but I admit that I’m oversimplifying just to anchor a timeline.

  250. Oh, yah, all of esr’s credibility disappears because he shared the common view that DECUS’ tape distributions were mostly aimed at VAX users …

    /sarcasm

  251. To tie back into the Non-Commercial clause my recollection was a lot of the open source in the day had some kind of academic license since a lot of the hacking was done in the university setting given the size and cost of machines. Distribution was via cards or mag tape…

    I still remember using punch cards in my FORTRAN class and DECUS tapes. Although we had bitnet, arpanet, etc. by the early/mid 80s.

  252. You’re arguing causally, I’m arguing consequentially. I’m saying it’s misleading to call SOS or Multics “open source” because (to a person not quite familiar with the way computing was done back then) the “open source” label will imply a lot of behaviors that nobody engaged in. There weren’t any wide-area networks, there weren’t any archive sites, there weren’t any forges, and most importantly the expectations about development practice that those technologies support weren’t yet present. You didn’t have people checking distributions out of repositories to hack them and ship back patches. They couldn’t do that; the infrastructure didn’t exist.

    Ah what? Open Source is not predicated on forges, WANs, development practices. Open source at the core is sharing code for the common good.

    Why does that even need to be stated here? Actual practitioners of open source existed back then as much as today. Not intent or theory of open source and it doesn’t matter if I pull from a git repo or get sent code as a print out or on tape via snail mail.

    This is like claiming that calling what they did programming would be misleading because they didn’t have IDEs, high level languages, keyboards/monitors, etc and all the stuff we associate with the programming label today.

    There weren’t many eyeballs back then because there wasn’t many eyeballs that did programming period. Even today, the number of eyeballs is greatly exaggerated in term of actual active commiters and the percentage of developers engaging in open source development back then was probably massively higher than today.

    Meh, the value of many eyeballs is IMHO vastly overstated for FOSS. The number of active devs for many proprietary code bases is much higher than for the average FOSS project and they have a significant eyeball advantage. For mission critical software a high degree of review is sometimes used and far more rigorous testing than for FOSS.

  253. @ Tom DeGisi
    > If a developer wants to use a piece of copylefted code on a proprietary project the developer is forced, before the fact, by copyleft not to use it.

    If the GPL code didn’t exist, the developer (in your usage) would also be “forced” not to use it.

    An interesting question is raised by the GPL requirement to provide source on distribution if the code is only used in house. I am not up to date on the FSF worldview to know whether they would argue that distirubtion includes intra-corporate use.

  254. @Nigel
    Yet somehow linux manages to patch serious security holes /much/ faster than closed source OSs.

  255. @TomM:

    An interesting question is raised by the GPL requirement to provide source on distribution if the code is only used in house. I am not up to date on the FSF worldview to know whether they would argue that distirubtion includes intra-corporate use.

    Internal usage is not counted as a distribution. OTOH it is only prudent to assume that any code stands a chance of being distributed later.

  256. @TomM:

    > If the GPL code didn’t exist, the developer (in your usage) would also be “forced” not to use it.

    Oftentimes, the existence of GPLed code forces a local minimum, by attracting developers who might otherwise be happy to work on permissively licensed code. Competition in code is good, but competition because of licensing wastes a lot of effort. Nobody rewrites a BSD or MIT licensed package because of licensing issues, but this happens with GPLed code all the time.

    See PySide vs. PyQT, or editline vs readline, or even the new call for developers to build a replacement busybox.

  257. Patrick Maupin Says:

    > Internal usage is not counted as a distribution.

    Citation here: http://www.gnu.org/licenses/gpl-faq.html#InternalDistribution

    IANAL, but I think I read that SPQR actually teaches lawyers, so maybe he will notice this.

    Patrick is correct, but to my knowledge, ‘internal usage’ is undefined within the GPL itself, but only interpreted by the FSF (which is why @TomM correctly needs to be up to date on the FSF worldview to know whether they would argue that distribution includes intra-corporate use). This makes the GPL ‘scary’ because many multi-national organisations are actually loosely organised. For example, the big accounting firms were actually a series of national firms individually chartered in each of the their territories. The national firms contributed some of their profits, staff, work-products, etc. to the the ‘world firm’, but the ‘world firm’ did not own the national firms. So Tom’s caution is prudent.

  258. @jsk:
    “Right now [the A&D killfile capability] is a very basic script that just does a regexp check. It replaces the whole comment block (username and all) with a “Comment Killed!” line. To add or remove someone you have to edit the .js file and reload the extension. I have plans to turn it into a more featureful thing with page-embedded controls, but I’m not sure when I’ll be able to tackle that. If you don’t mind the minor code twiddling I will put it up on my site, otherwise I’ll see about getting something a little nicer built.”

    I’d love to have it. Given the makeup of the typical poster to this board, I think it would see a lot of use. This is not a crowd afraid of hand-editing text files.

  259. >Where was RMS in 1955?

    According to wikipedia, about two years old.
    For all of RMS’ other real quirks, I don’t think we can really fault him for not being a leader in the programming comunity in 1955.

  260. TomM,

    > If the GPL code didn’t exist, the developer (in your usage) would also be “forced” not to use it.

    Exactly. You are reinforcing my point. Within the universe of proprietary code, copyleft negates the very existence of code, the very existence of those who wrote it, and the very existence of the cooperative impluse that they tried to exercise before copyleft killed it. Copyleft is death to software and to freedom. And since proprietary code is 90% of all code, copyleft kills alot of software and alot of freedom.

    esr,

    I just got through skimming The Magic Cauldron (rather than rereading it). I am afraid that it does not cover the economic case I gave above. Most software is not worth the effort of turning into open source. Where I work I am familiar with our code base and with the code base of one of our providers. We don’t throw anything away. After a point a significant fraction of the code is well running important code with very few people who know how it works. To do a good job of open sourcing it would be very expensive. To do a poor job of open sourcing it would reflect badly on the developers and the company. That doesn’t even take into account all the code developed by junior developers and maintained with good enough and inexpensive kludges that no developer with an ounce of self respect would ever let see the light of open source day. It really wasn’t worth making beautiful in the first place, and it still isn’t. From what I can see most code should not be open sourced, and this won’t change.

    Yours,
    Tom

  261. @Tom
    “Copyleft is death to software and to freedom.”

    You sound exactly like a protestant minister telling us about those heretics who do not believe the serpent talked to Eva in a human voice.

    How pathetic.

  262. @Tom DeGisi:

    The great is the enemy of the good, and most open source packages contain (at least starting out) a large quantity of embarrassing code. The best open source developers are those who can say “no it’s not great, but it’s a start.”

  263. >They’re solutions all right, just rather unpalatable ones. And I point them out to demonstrate that your reasoning is sloppy. This matters, because people who reason sloppily about important things often end up sabotaging themselves – and like RMS, you spend a lot of effort moralizing in a way that is seriously self-sabotaging.

    You’ve got to be kidding me, this reminds me exactly of my conversations with rms about global warming – no matter how many times I would show him the manipulations, just “hide the decline” for instance, explained in detail, he would still just repeat his “it’s out of context” nonsense and act as he was victorious even that the exact context was black on white explained right in the front of him … no, he had his strawman idea of what “hide the decline” means and just refused to replace that with reality.

    You’re doing exactly the same thing. You act like you’re so cleverly refuting my points but you’re not, it’s not even my points … you just attack some strawman idea of what you want those points to be so you can knock them down. I’ve told you several times now that it’s not my claim that open source is strictly necessary which is what your “demonstration” is attacking, you’re attacking your own fantasy not my actual position and then proclaim my reasoning sloppy based on that.

    And on top of that in your demonstration (that attacks a point I’ve never made) you’re using an example solution that you yourself believe is unacceptable (let alone feasible) – how do you hold that in your brain without exploding? You may just as well say that until there is an “possible solution” of implanting brain-chips into programmers heads that prevent them from writing harmful code we can never advocate for open-source over proprietary one on the basis of morals … that’s just total nonsense and non sequitur. How can it not matter on the acceptability or indeed of the feasibility of the proposed solution? It clearly does, but you choose to ignore that because your “demonstration” would fall apart …

    You may have different positions than rms but your mode of thinking is exactly the same, you’re unable just for a second to examine your own positions in an objective way, you replace other people’s positions with your strawman ideas of them so you can easily attack them and therefore convince yourself that your position is justified. It’s not your fault, that’s what we humans do – it’s called cognitive dissonance and is very well understood phenomenon.

    It is interesting that this manifests itself with intelligent people the most, I guess it’s because they’re so used to being right compared to others that do not think they could be wrong that much. Also one contributing reason of why smart people believe stupid things is becase they make very clever reasons to believe them few are able to properly refute – I do not want to claim that I’m smart but I observed that on myself, I’ve made witty arguments for my positions that people near me were not able to deconstruct so I automatically assumed I must therefore be right, then I found out about AGW and related issues – oh boy, there is no bigger mistake than that … I little humility doesn’t hurt, it’s then when I understood the full meaning of Socrates quote “I know one thing, that I know nothing”. You should try that too ;)

  264. Patrick Maupin,

    > The great is the enemy of the good, and most open source packages contain (at least starting out) a large quantity of embarrassing code. The best open source developers are those who can say “no it’s not great, but it’s a start.”

    Sure, but some of those embarrassing things are things our competition could take advantage of, right? When you are competing for best code it’s a little different than competing for most dollars.

    Yours,
    Tom

  265. @libreman

    Ok, you’ve gone from entertaining to outright surreal. Fess up, you’re really Mark Shaney, aren’t you?

  266. liberman,

    You are talking past each other. Everything you complain of are things you are doing. It’s not because you are a hypocrite. It’s because they are impossible to avoid. Cool your jets and stop feeling misunderstood. You are understood better than you think. Understanding does NOT produce agreement. Sorry. If your priorities are different (and based on what you emphasize, they are) you may never agree, unless you can convince esr to accept your priorities or vice versa. This is unlikely because you have wildly different life experience driving your priorities.

    Welcome to the monkey house.

    Yours,
    Tom

  267. esr,

    Please request liberman to stop ignoring everyone else. It’s rude, and it’s not helping him. He is not experienced with the blog culture and does not realize how easy it is to get good conversation going with people other than you. Patrick Maupin, Jessica Boxer and Cathy are especially tolerant and patient, although I have seen Patrick’s inner curmudgeon rise up. I am sometimes tolerant, although Patrick and I think Winter will confirm how intolerant I can become if you press the right buttons. Winter is pretty tolerant too, and jsk is offering some good arguments along with the generous helpings of amusing (to me, but never everyone) snark.

    Yours,
    Tom

  268. Winter,

    I know exactly what you mean and you are exactly right. That’s how a hell and brimstone sermon is preached and they can be preached on any subject and with absolutely no references to hell or brimstone. Chimpazees screech alot when they do it.

    I particularly like the sermons preached with this amount of fervor against slavery, mass murder and abortion. Don’t you?

    Welcome to the monkey house.

    Yours,
    Tom

  269. You’ve got to be kidding me, this reminds me exactly of my conversations with rms about global warming – no matter how many times I would show him the manipulations, just “hide the decline” for instance, explained in detail, he would still just repeat his “it’s out of context” nonsense and act as he was victorious even that the exact context was black on white explained right in the front of him … no, he had his strawman idea of what “hide the decline” means and just refused to replace that with reality.

    But it was taken out of context. Temperature data were being artificially depressed by bogus post-1960 tree-ring data; “[Michael Mann’s] trick to hide the decline” was a means to filter out the bogus tree-ring data and keep the observed instrumental temperature data — which clearly showed a warming trend.

    The necessity of “hiding the decline” is particularly damning because higher CO2 construction in the atmosphere is precisely the thing that makes tree-ring data unreliable: greater CO2 concentrations mean that trees grow thicker rings. Denialists thus have it exactly back-ass-wards: hiding the decline was necessary to correct for man’s very real impact on the environment, not as a fudge to synthesize evidence of an impact that isn’t really there.

  270. @Marco:

    > to my knowledge, ‘internal usage’ is undefined within the GPL itself,

    I think this may be true for v2, and that they tried to fix it in v3. Section 2 is quite torturous, but appears to cover giving modified copies to employees and contractors, and things like uploading them to run on Amazon EC2.

    Of course, the contractor thing might be problematic, since the normal mechanism for controlling those is NDA, and they have this to say about an NDA:

    http://www.gnu.org/licenses/gpl-faq.html#DoesTheGPLAllowNDA

    Bottom line, as we agree — the safest path is to “just say no” and even though the FSF seems to have tried to address some of these concerns and be more business friendly in some ways with v3, they may or may not have succeeded. The irony is that they now seem to be very supportive of secret software with language like “Those thus making or running the covered works for you must do so exclusively on your behalf, under your direction and control, on terms that prohibit them from making any copies of your copyrighted material outside their relationship with you.”

  271. @Tom DeGisi:

    > Sure, but some of those embarrassing things are things our competition could take advantage of, right?

    I’m certainly not arguing to open source everything, particularly not secret sauce. The best time to open source something is up front, when you realize there is nothing out there that does what you need, and that the code you will write won’t provide a major strategic advantage. Hell, you don’t even have to put the company’s name on it — let a developer release it.

  272. @Tom DeGisi

    This is not a matter of agreement or disagreement, disagreement I can and do respect as I do different priorities. What I do not respect is people who use bogus arguments to refute points you’ve never made and pretend they’ve somehow shown your logic to be “sloppy”.

    The only thing esr “demolished” is his own strawman as I have demonstrated several times already. If you believe otherwise please do explain this section of his where he supposedly “demolishes” my argument (that I maintain I’ve never made):

    esr:
    >Your logic is broken. Of course the best overall solution to these problems is open source, and that’s the one I work towards because it’s the best solution for a lot of other problems you didn’t mention, too. But those particular problems are only arguments for the necessity of open source if there are in principle no other solutions.

    >Take intrusive DRM, for example. It could be effectively prevented by a legal ban, because violations would be easy to detect and cause actionable harm. So, while open source is sufficient to solve this problem, it’s not strictly necessary. Which in turn means you cannot argue deontically from “DRM is evil” to “closed source is evil” – the relationship is underconstrained. Logic fail!

    There are at least 2 problems with this which I have already pointed out several times to no avail, everyone just ignored it including esr … so here they come AGAIN, please DO address them if you want to claim I’m wrong:

    esr states: “But those particular problems are only arguments for the necessity of open source if there are in principle no other solutions.”

    1. But I’ve never argued for necessity of open source in such an absolute terms. If you believe I did or such an absolute necessity is somehow implied in what I’ve said (or FSF or rms) then show me where. Otherwise this is bogus position to refute, nobody actually made that claim – he is “refuting” a strawman. Or possibly clarify what is meant by “necessity of open source” in case that’s where the misunderstanding lies.

    2. Even if somebody made that argument (which it didn’t) the example of why it is wrong is also bogus because it contains a “solution” that nobody actually considers a viable solution anyway, INCLUDING esr himself! A “solution” that is not considered a solution by any of the parties involved is not solution at all in any practical sense and can not be therefore used in such a way – that’s quite natural realization, is it not?

    I’m awaiting explanations of why the above is wrong if anybody believes so.

  273. @libreman
    > 1. But I’ve never argued for necessity of open source in such an absolute terms.

    From http://esr.ibiblio.org/?p=4118&cpage=6#comment-369476

    > Well, the reasoning of FSF advocate such as myself goes like this: backdoors are morally objectionable
    > (you agree) -> backdoors are readily found in proprietary software where they’re hard to detect and
    > remove (your position?) -> proprietary software is also morally objectionable because it facilitates former
    > morally objectionable practices (you disagree). The solution: Use Free Software/open-source where
    > detection and removal is a piece of cake (your solution: unknown)

  274. @Tom DeGisi:

    Actually, libreman is interacting with the rest of us a bit, and posted a question for me that I did not notice right away.

    @libreman:

    I agree that proprietary software may be better too, in individual instances. I think this is where the misunderstanding lies. I DO NOT claim that EVERY piece of proprietary software is bad as many seem to be interpreting it, not at all. I claim that IN AGGREGATE proprietary software is bad because things like backdoors, DRM, snooping manifest statistically much more often in it than in OSS (notice that I do not claim they do not appear at all in OSS either).

    But this is not a moral issue. Here, let me paraphrase for you:

    I agree that nails may be better too, in individual instances. I think this is where the misunderstanding lies. I DO NOT claim that EVERY nail is bad as many seem to be interpreting it, not at all. I claim that IN AGGREGATE nails are bad because things like split boards, tool marks, loosening up after use manifest statistically much more often with nails than with screws (notice that I do not claim they do not appear at all with screws either).

    Now, one difference between your copy and mine is that the problems you point out are caused by human behavior, and the problems I point out are caused by nature. But guess what? Human behavior is part of nature…

  275. @Jeff Read

    >But it was taken out of context.

    You see? You are doing the EXACT same thing, you do not ask for the context I’m talking about, you just jump right in with your idea what the context I’m talking about is and refuting that instead of the actual situation.

    rms thought it referred to global temperatures – some so called “right-wing” talking heads really did imply that or even said it outright and the so called “left-wing” media reinforced it by attacking that strawman instead of actual problem (you see how they work hand in hand?) So I sent him REAL context (which is what you’re talking about) but I suspect he never even read it as he repeated his strawman about global temperatures several times since then.

    I’m not going to engage in a discussion of why I think you’re incorrect (mainly because there is no reason the data before 1960 couldn’t be just as wrong as the data post-1960 :P) so let’s just agree to disagree on this one. The important point is the above which you have demonstrated for me nicely and for that I thank you …

  276. >But this is not a moral issue. Here, let me paraphrase for you:

    Yes you could paraphrase me like that but I do not get the point – your paraphrased paragraph is perfectly valid in the same way mine is. If the use of nails would cause significantly more injury than using screws I would make the same exact argument of using screws whenever possible and advocating for screws to prevent injury. (And if possible advice people to license screws under the GPL so nails could not be used on the same piece of wood as screws if they wish to prevent injury as much as possible :P)

    >Now, one difference between your copy and mine is that the problems you point out are caused by human behavior, and the problems I point out are caused by nature. But guess what? Human behavior is part of nature…

    So? You’re losing me … I’ve never claimed anything that contradicts this, in fact I do not see how it is relevant at all …

  277. @jsk

    The paragraph you’ve quoted doesn’t imply necessity in an absolute sense as esr is using it. It says that if you accept certain premises then it makes sense to license software under the GPL. It also does not imply there is no other solution in principle – that would be silly, of course there are other solutions but they’re either impractical, undesirable or outright impossible compared to going OSS. I thought that was obvious.

    Maybe you can explain how that made you think OSS is NECESSARY in an absolute sense that if there is any kind of another solution no matter how unfeasible that point is invalidated? Whicj is how esr is using it, I’d like to understand how you come to believe that so I can avoid it in the future, thanks.

  278. Marco, your observation is correct that the license provides no definition and even FSF’s “guidance” that you cited is problematic – see the bit about contractors working “off site”.

  279. @libreman

    You are, in your own words, presenting it as reasoning, and proceed to base your statements for several posts after it (or to frame your statements as if based on it). esr showed how your reasoning, thus your premise, is flawed, which invalidates much of your words.

    Now, you may have simply not explained yourself properly, but that’s your mistake, not the mistake of others who can only read what you write and not what you (opaquely) intend. All your attempts to correct that are done in a way that tries to make everyone else wrong. You also do the exact reverse and ascribe meaning and intent to other commenter’s posts that have only a tenuous relationship with the actual words written. This is present in every post, and I won’t individually quote them. Call it an exercise for you to find it.

  280. @libreman:

    > So? You’re losing me … I’ve never claimed anything that contradicts this, in fact I do not see how it is relevant at all …

    You claim it is a moral issue, and the rest of us claim it is a practical issue, or that, in any case, we don’t want to live in rubber rooms for our own good.

  281. Jeff Read writes: “The necessity of “hiding the decline” is particularly damning because higher CO2 construction in the atmosphere is precisely the thing that makes tree-ring data unreliable: greater CO2 concentrations mean that trees grow thicker rings. Denialists thus have it exactly back-ass-wards: hiding the decline was necessary to correct for man’s very real impact on the environment, not as a fudge to synthesize evidence of an impact that isn’t really there.”

    This is amusing because you pretend that Mann et al were “solving” a problem when in fact you touch on the original problem with tree ring data. Mann et al’s false assumption that all tree ring growth was temperature correlated. Actually, the problematic tree ring data was the bristlecone pines, where the literature shows that bristlecone pine tree ring dimensions are correlated to precipitation ( as one would expect given their environment ) rather than temperature.

    But nonetheless, the “hiding” of the departure of the tree ring dimensions from recorded temperature based on claims of a sudden CO2 concentration correlation was without any credible support. More of the AGW handwaving to cover their data manipulations e.g., the Yamal tree data.

    1. >Actually, the problematic tree ring data was the bristlecone pines, where the literature shows that bristlecone pine tree ring dimensions are correlated to precipitation ( as one would expect given their environment ) rather than temperature.

      It’s even trickier than that. When I read the original bristlecone-pine studies in the late 1970s, I learned that the ring dimensions are jointly correlated with precipitation and temperature – if you don’t have an independent record of one, you can’t deduce the other.

  282. Patrick Maupin – “software” – a French irregular verb meaning “to leave well enough alone”.

  283. liberman,

    > What I do not respect is people who use bogus arguments to refute points you’ve never made and pretend they’ve somehow shown your logic to be “sloppy”.

    You, yourself, are doing what you complain about right in that sentence! Language is so utterly imprecise that you, personally, are every bit as disrespectful and bogus as what you complain about. If you wish to avoid this, please use symbolic arguments only, or cut people a break. I don’t actually think you are disrespectful and bogus by my standards. I think you are *accidentally* disrespectful and bogus by your standards.

    > The only thing esr “demolished” is his own strawman as I have demonstrated several times already.

    No, you have not. Demonstration is much harder than you think. By the same token, esr demolished NOTHING. He just made a good argument. You are also making good arguments. None of the arguments in this thread are demonstrated. People talk like this because their own arguments and arguments they agree with are so completely persuasive to themselves. Still in the monkey house.

    > everyone just ignored it including esr

    There are many reasons people ignore arguments. Sometimes they are ignored because people think they have already been refuted. Does not matter whether they have or not. Sometimes people ignore arguments because they find them completely unpersuasive and don’t bother with refutation. Sometimes people ignore arguments because they have a sneaking suspicion they are correct and they don’t know how to refute them. In most cases, repeating the arguments won’t get anyone to change their behavior, since their reasons won’t have changed.

    In this case, because you chose to ignore alot of good people you may have been ignored in return.

    > But I’ve never argued for necessity of open source in such an absolute terms.

    OK, I can take this one. If it is not necessary, your back door argument loses a great deal of it’s force.

    > the example of why it is wrong is also bogus because it contains a “solution” that nobody actually considers a viable solution anyway, INCLUDING esr himself!

    Sorry. I consider legal remedies for back doors a useful solution, if well implemented. Not bogus and quite practical. In fact, legal remedies are probably preventing back doors right now. If I, as an employee, install a back door for my own benefit, I can be thrown in jail. I wouldn’t do that anyway, but others have and have been prosecuted for it. I am sure this has detterred some employees from doing this.

    I would bet a buck that esr would support penalties for an employee who installed a back door for his own benefit if the employee agreed to such penalties in her employment contract. I would also bet a buck that esr would support penalties for a software company which installed a back door for its own benefit if the company agreed to such penalties in its sales contract. And I also would bet a buck that esr would contend that both those contracts would deter back doors. Wanna bet?

    Know the people you are arguing with, if possible, and think how they might counter your points.

    In addition it has been pointed out – and either you appear to have ignored or I appear to have missed it – I do skim and I am old so my memory isn’t what it was – GPL does NOT prevent back doors. I suspect that good threat protection software which attempts via heuristics to catch back doors in the act is the best known technical method for finding back doors. No open source required. And I, personally would much rather trust behavioral analysis, since the actual behavior of actual code is my gold standard. I don’t trust people reading code. We are bad at it. It is too easy to obfuscate code to fool code analysis as well, but even then, I want my automatted code analysis to read the compiled code, not the source. Again, the actual compiled code is my gold standard. It’s too easy to compile in the backdoor without checking in the source that enables it. Open source is NOT the best way I can think of. Not even close.

    Yours,
    Tom

  284. Marco:

    >Patrick is correct, but to my knowledge, ‘internal usage’ is undefined within the GPL itself, but only interpreted by the FSF (which is why @TomM correctly needs to be up to date on the FSF worldview to know whether they would argue that distribution includes intra-corporate use). This makes the GPL ‘scary’ because many multi-national organisations are actually loosely organised. For example, the big accounting firms were actually a series of national firms individually chartered in each of the their territories. The national firms contributed some of their profits, staff, work-products, etc. to the the ‘world firm’, but the ‘world firm’ did not own the national firms. So Tom’s caution is prudent.

    You’re over-analyzing and trying to generate problems where there are none. If corporate internal usage is understood as non-problematic then it is trivial to solve it for multiple “chartered”, “franchised” or whatever other type of linked corporate entities by simply providing the source to every branch – what’s the big deal? The only ones who see a problem here are anti-GPL zealots who just want there to be a problem for ideological reasons rather than identifying true hurdles. It’s a typical FUD campaign.

  285. @SPQR:

    Marco, your observation is correct that the license provides no definition and even FSF’s “guidance” that you cited is problematic – see the bit about contractors working “off site”.

    The FSF’s twin goals of wanting to coerce authors into releasing new code under the GPL, and simultaneously wanting users to enjoy “maximum freedom” are problematic for them because authors are usually users as well. I wrote about a similar issue (on, e.g. distributing code that can dynamically link to readline) back in 2003:

    http://groups.google.com/group/comp.lang.python/msg/29c7588fbecd8289?pli=1

    Note that I also explained why IMO no court case has ever come up to test their goofy hypothesis.

  286. @ Doc

    >Yet somehow linux manages to patch serious security holes /much/ faster than closed source OSs.

    But such speed is useless if the patches aren’t being applied. Consider the differences in approach from running Ubuntu to running Debian Stable. In one, you will have consistently up to date software, at the risk of each update possibly breaking something. In the other, you accept slightly out of date software with the goal of stability and massive amounts of testing leading to mostly break free patches. I wonder if we did a survey of people using OSS software, especially in a server environment, how many have the absolute latest patched versions of the software they’re running. I would wager that it’s less than 50%

  287. @libreman:

    You’re over-analyzing and trying to generate problems where there are none.

    No, he’s not.

    If corporate internal usage is understood as non-problematic then it is trivial to solve it for multiple “chartered”, “franchised” or whatever other type of linked corporate entities by simply providing the source to every branch – what’s the big deal?

    The “big deal” is that, when the code is distributed from one entity to another, it is now all under the GPL, and the FAQ makes it clear that an employer cannot, for example, use an NDA to enjoin his employee from making a copy of GPLed code.

    But you probably don’t think it’s a problem that when someone gives a friend a Ubuntu CD, he’s violated GPL v3 copyright unless he accompanies it with a written offer, either.

  288. libreman,

    > You’re over-analyzing and trying to generate problems where there are none. If corporate internal usage is understood as non-problematic then it is trivial to solve it for multiple “chartered”, “franchised” or whatever other type of linked corporate entities by simply providing the source to every branch – what’s the big deal? The only ones who see a problem here are anti-GPL zealots who just want there to be a problem for ideological reasons rather than identifying true hurdles. It’s a typical FUD campaign.

    Work with many lawyers, have you? Not FUD. Sometimes interesting, but mostly pain. Do you have any idea what a complete and utter pain Sarbanes-Oxley has been for IT? And to NO benefit to people, but considerable cost to everyone. Copyleft is the same. NO benefit. High cost. You can avoid the cost of Sarbanes-Oxley completely by not going public in the U.S. Guess what is happening. You can avoid the cost of copyleft by not using copyleft software. Guess what is happening.

    In this case your priorities have been informed by your experience and mine have been informed by mine.

    Yours,
    Tom

  289. >If corporate internal usage is understood as non-problematic then it is trivial to solve it for multiple “chartered”, “franchised” or whatever other
    >type of linked corporate entities by simply providing the source to every branch – what’s the big deal?

    It’s not always so simple, and it may not be for tax or legal reasons. Consider the company I work for, in the medical field. My company is a pharmacy and distributes medications to various customers in different states from a central location. We do this via a series of company owned delivery hubs in each of the states. Recently, the company discovered laws in a few states has given us reason to believe that because we store these drugs overnight in these delivery hubs sometimes, that those hubs may be legally considered pharmacies, because they are owned by the head pharmacy. If they are considered pharmacies, we would have needed a multitude of additional licenses as well as different locations, different security and other legal hassles.

    Instead the company decided to split the delivery arm of the company into a separate delivery company, majority owned by the pharmacy. Because the delivery company is now an independent company, it’s hubs are as a FedEx or UPS hub, and not subject to pharmacy law. However, this also means that the delivery company must pay the pharmacy (or the pharmacy must pay the delivery company) for any good or services transferred. Incidentally, this is how you get such crazy crap as Wells Fargo suing Wells Fargo, because sometimes you have to jump over legal hurdles just to do business.

    Our software is developed entirely in house, and is a major asset to the company. Had we used any GPL software (and we didn’t in part because such a scenario was anticipated), it would require us to provide the source to our software to the delivery company if we “distributed” the software to our company. That would not happen, as it would be a financial mess and too large of a risk to the company. Had we used GPL software, the pharmacy would have rather paid the programmers to rip out the GPL modules and replace it with something else rather than have to give away an asset like that.

  290. >You claim it is a moral issue, and the rest of us claim it is a practical issue, or that, in any case, we don’t want to live in rubber rooms for our own good.

    I knew we would get to this projection of fear of every free-market ideologue eventually … it’s a fiction. There are no “rubber walls” in GPL, it’s exactly the other way around – it’s proprietary software that introduces “rubber walls” because it is made for the masses by a central authority with central planning which primary goal is to appeal to as many people as possible so it ends up implementing features “for the greater good” and doesn’t care about the fringes as they are not their target group. And there is nothing those fringes can do about it because the source is proprietary and they can not modify it for their individual needs. Proprietary software is in principle anti-individualistic … that never occurred to you? This is so symptomatic of free-market ideologues who take its prevalent dogmas as a religion rather than a set of true principles and never think about them.

    I’m free-market proponent too, I’m just not an ideologue and do not religiously follow dogmas imposed on me. I do believe that Free Software is much more in line with the principles of free-market than proprietary software and significantly so. Proprietary software produces monopolies (what I revelation, eh?), restricts healthy competition and stumps progress … this is why OSS is kicking its ass as esr would surely agree.

    But economic pressures while significant are not enough, they never were. There is also a human dimension, a social one – I can see the goosebumps setting in right now :) those dogmas and preconceived notions just popped up their ugly head again I suspect. As any talk of morality is strictly forbidden in the world of an free-market ideologue – it’s stylish not to care about the very real social consequences of censorship, oppression, persecution of inconvenient people and ideas such as wars, dictatorships, suppression – they’re out of the economic framework so they’re “unimportant”, I’ve read NOTHING of such issues here in connection with any analysis. It’s simply swept under the rug as if nobody cared … and my question is, do you really not care? Is a society as a whole not your concern?

    How is GPL less capitalistic than proprietary software? I believe it’s much more so … and also having the benefit of minimizing the use of software as a tool of oppression and control.

  291. @Tom DeGisi

    You’ve said exactly nothing of the actual problem I was commenting on … so representative of anti-GPL zealots (ok, Ill stop using that term, I do not like when someone uses it so I shouldn’t either, it detracts from the important parts … I just thought I’d give you a little of your own medicine – and it’s not unjustified either, is it?). So the important part … yeah … you said absolutely *nothing* relevant!

  292. libreman,

    > How is GPL less capitalistic than proprietary software? I believe it’s much more so … and also having the benefit of minimizing the use of software as a tool of oppression and control.

    First I will answer your question. Neither is more free market than the other under our current system. People who write code and release under GPL are making the same free choices as people who keep software proprietary.

    Now I will point out that you are knocking about a straw man yourself. People here have been arguing for open licenses other than the GPL, and haven’t made any reflexively capitalistic arguments that I have noticed. The ‘rubber rooms for our own good’ is not a capitalistic argument, per se. It’s a ‘leave me alone personally’ argument. There is some overlap, but if you know Patrick you know he is an employee. He gets paid whether GPL is used or not. I do believe that people have the right to keep software proprietary. I do not think people have the right to use the GPL as completely as it is used. That is because our copyright laws are overbroad. Note that you can always try to keep software proprietary by keeping it private, even in the absense of copyright.

    BTW, even if you narrow our copyright laws as I would like, the answer to your question does not change. Your question is a non-sequitor.

    Yours,
    Tom

  293. The prime example that it is the proprietary software that is locking you in a rubber room is, who else than – Apple. It’s called a walled garden for a reason, duh! “Oh, we can not allow this app on our market, it could hurt some of our users so now nobody can install it, it’s for your own protection”.

    Where that EVER happened in a OSS (or more specifically GPL)? The reality is staring you in the face and you still hold on to your dogmas, there is a name for that – religion! You’re so scared of the hypothetical immaterial concepts that you let infect your mind that you miss that the exact opposite is happening in the actual real world. Can’t you see that?

  294. libreman,

    > I just thought I’d give you a little of your own medicine – and it’s not unjustified either, is it?

    Sure it is unjustified, by my standards, which both you and I am following. By your standards all people are horrible sinners, including you, because we cannot follow the God like perfection of argumentation you desire. You think you follow it, because your own arguments are persuasive to you. But you don’t actually follow it, because you can’t.

    Try again, and start cutting people a break.

    Yours,
    Tom

  295. @libreman
    > Where that EVER happened in a OSS (or more specifically GPL)? The reality is staring you in the face
    > and you still hold on to your dogmas, there is a name for that – religion! You’re so scared of the
    > hypothetical immaterial concepts that you let infect your mind that you miss that the exact opposite
    > is happening in the actual real world. Can’t you see that?

    You’re doing the subject-shift thing again, as well as attributing intent where the words don’t support it. It’s getting old.

  296. libreman,

    > The prime example that it is the proprietary software that is locking you in a rubber room is, who else than – Apple. It’s called a walled garden for a reason, duh! “Oh, we can not allow this app on our market, it could hurt some of our users so now nobody can install it, it’s for your own protection”.

    > Where that EVER happened in a OSS (or more specifically GPL)? The reality is staring you in the face and you still hold on to your dogmas, there is a name for that – religion! You’re so scared of the hypothetical immaterial concepts that you let infect your mind that you miss that the exact opposite is happening in the actual real world. Can’t you see that?

    LSTM! (Laughing Silently To Myself, creator Jane DeGisi, placed in the public domain.)

    Dude! We know this! We have criticized Apple on this exact basis. However, there is more than one rubber room possible in the world. GPL has not (and will not) ever prevent the Apple’s of the world from building that sort of rubber room. The rubber room of the GPL is the known restraints it places on OUR personal use of it. Which you resolutely deny, of course, in your vast experience of what it is like to work where we work with the management and lawyers we work with. I can use code licensed under truly free (no copyleft, no commerical limitations) licenses all day long with NO effort at all. I have to get permission for GPLed stuff. This is not worth my time or effort.

    Yours,
    Tom

  297. >Now I will point out that you are knocking about a straw man yourself.

    No, the argument about “rubber rooms” is typical libertarian line and you don’t tell me it has nothing to do with dogmas of what capitalism should look like. It is *the only reason* Patrick said it, there is no rational basis for saying that therefore it has to be ideological. I’m almost sure of it and I hope Patric can confirm this (please be honest ;)

    >Neither is more free market than the other under our current system.

    Read a few posts back and you’ll find that this is absolutely not the impression commenters here leave. Many of them rail against the GPL as if it was anti-capitalistic devil and pondering about whether rms is a communist or has some messianic complexes. So no, this is not the overwhelming position of the people here, it’s a position I’ve formulated several times and was shouted down for it so do not shift argument as if it was obvious the whole time that this is the case.

  298. > No, the argument about “rubber rooms” is typical libertarian line and you don’t tell me it has nothing to
    > do with dogmas of what capitalism should look like. It is *the only reason* Patrick said it, there is no
    > rational basis for saying that therefore it has to be ideological. I’m almost sure of it and I hope
    > Patric can confirm this (please be honest ;)

    Ok, I’m bowing out right here, it has gone from surreal to distressing. I figure libreman has some kind of autism thing going, maybe something else but definitely has a terminally skewed world view. Nothing more to be gained here, even entertainment, moving along…

  299. libreman,

    > No, the argument about “rubber rooms” is typical libertarian line and you don’t tell me it has nothing to do with dogmas of what capitalism should look like. It is *the only reason* Patrick said it, there is no rational basis for saying that therefore it has to be ideological. I’m almost sure of it and I hope Patric can confirm this (please be honest ;)

    Stop reading people’s minds. Be careful at this point. You may be tempted to accuse me of reading your mind. Try to figure out how I will respond before you do. In addition, I *already* gave you a rational non-ideological basis for saying it.

    > Many of them rail against the GPL as if it was anti-capitalistic devil and pondering about whether rms is a communist or has some messianic complexes.

    I know they do. This is because of the way RMS behaves and the arguments he makes. You aren’t RMS, are you? I will do you the courtesy of responding to your arguments as you make them. I don’t have the somewhat stupid habit of calling someone a Stallmanite and mocking them for it. It’s tiresome, and not amuzing snark.

    Yours,
    Tom

  300. esr:

    >It’s even trickier than that. When I read the original bristlecone-pine studies in the late 1970s, I learned that the ring dimensions are jointly correlated with precipitation and temperature – if you don’t have an independent record of one, you can’t deduce the other.

    Exactly, the widths of the rings are dependent on many variables, precipitation is only one of the more obvious ones. It depends also on the amount of sunshine, wind, humidity … hell, if a bear takes a cra* near that tree that’s going to have an effect :)

    Also CO2 itself has an effect, except the exact opposite of what the Mann graph showed. CO2 promotes tree growth (as almost every other flora) so if anything it would show up as “warming”. The causes for the divergence and pure speculation, nobody ever committed to any serious explanation of the divergence in these terms and they know exactly why – because it’s bollocks.

  301. >GPL has not (and will not) ever prevent the Apple’s of the world from building that sort of rubber room.

    I don’t know how exactly you mean this, GPL is specifically designed to prevent exactly such rubber rooms, that’s it’s prime focus and reason for its existence! If you mean that in some other sense, like for example that it never gets such an adoption to dissuade Apple from running walled gardens then that depends on us. If we reject GPL then you’re right, if we embrace it then not so much. If significant portion of the good apps are GPLv3 then Apple will think twice about running a walled garden. In that situation it make much more sense (even commercial one) to run open market – a truly free market, and we all win.

    >The rubber room of the GPL is the known restraints it places on OUR personal use of it.

    Is this a joke? The proprietary software (the one that BSD-style code can be transformed into) places infinitely more restrictions on our personal use of the software – it forbids eeevrything! Sharing? Nop! Modifying? Nop! Studying (reverse-engineering)? Nop! and many many others. The GPL is *designed* to give users the ultimate freedom of choice – the only thing it “restricts” is in turn restricting others. It restricts restrictions, nothing else. It tries to make sure that the same freedoms that were granted to you are granted to everyone else down the chain so you can not take them away. But I guess you consider restricting others “your freedom”? That is the freedom you miss?

  302. William,

    Yes, that license does effectively keep anyone from using the software including the Allah and Christian Trinity, if they exist. People generate a suprising amount of self-refutation, don’t we?

    Yours,
    Tom

  303. me: Libreman’s argument about proprietary vs. GPL is same as screws vs. nails.

    Libreman: Yes, And (somewhat jokingly) if possible advice people to license screws under the GPL so nails could not be used on the same piece of wood as screws if they wish to prevent injury as much as possible.

    me: But people don’t want to live in rubber rooms.

    Libreman: I knew we would get to this projection of fear of every free-market ideologue eventually … it’s a fiction.

    Ummm, OKaaaay. I submit that the government letting somebody license screws and telling me I can’t use a nail because I might hurt myself is, more literally than figuratively, placing me in a rubber room. And you seemed to agree that this is, in some ways equivalent to proprietary vs. GPL.

    The rest of your posting on this doesn’t distinguish between people comfortably ensconcing themselves in rubber rooms (e.g. in Apple’s ecosystem) which I strongly discourage but can’t and don’t want to control, or the government forcing me into a rubber room, which, depending on circumstances and exactly how slowly they ramp up the heat, I might resist rather violently.

    This has nothing directly to do with capitalism, other than that capitalism will naturally flourish where people are free (but can flourish in other cases as well).

  304. This is an interesting question … what you you consider as “freedom”. Obviously you seem consider the ability to license the software with a proprietary license a “freedom”. But what does “proprietary license” really mean? It means putting *another restrictions* on the user of the software.

    So the question becomes: Is the ability to restrict others in itself a “freedom”?

    If so then this has very far-reaching consequences. If we call such a license that allows this “permissive” then similarly we could call a society where stealing is not forbidden by law a “permissive” society in the same exact sense. Apparently by this logic if you forbid the act of stealing then you’re putting additional restrictions on the people and this is supposedly less free than the previous situation where everything was allowed. Would anybody be willing to argue for this position or demonstrate where it differs from the logic of the “permissive” license?

  305. > Is this a joke?

    No.

    Did you read me explaining to you how it constrains me?

    > The proprietary software (the one that BSD-style code can be transformed into) places infinitely more restrictions on our personal use of the software – it forbids eeevrything! Sharing? Nop! Modifying? Nop! Studying (reverse-engineering)? Nop! and many many others.

    I know. It isn’t claiming to free me.

    > The GPL is *designed* to give users the ultimate freedom of choice

    But it doesn’t.

    > the only thing it “restricts” is in turn restricting others. It restricts restrictions, nothing else.

    Wrong. It places certain defined restrictions on me that I have to obey whether I want to restrict others or not. Those restrictions have costs it requires me to pay.

    > It tries to make sure that the same freedoms that were granted to you are granted to everyone else down the chain so you can not take them away.

    Yes it does, and it forces me to pay the costs of doing so.

    > But I guess you consider restricting others “your freedom”? That is the freedom you miss?

    No. In this case I miss the freedom of not having to ask permission to use the code, as I mentioned. However, I also miss the freedom of taking code proprietary. In most cases this harms no one. The employees where I work, who are the only people who use the software I work on, don’t benefit from it being open source. Quite the opposite, since the extra expense of opening the source would probably come out of our paychecks. This is true for 90% or more of the software written on this planet. Why should we bear the expense of scratching your itch? If you really want to cooperate and be open and free with people, like me, who write software for fellow employees, and who are 90% of the developers in the world you will avoid the GPL.

    BTW, I know you hate it when people ignore your points. When are you going to stop ignoring mine?

    Yours,
    Tom

  306. @William B. Swift:

    That license was obviously written by a piss-poor coder who got tired of people criticizing his code. It’s probably the best thing he’s ever written:

    “The software may not be used in any context for the purpose of any communication that disparages or discriminates [sic] a person or a group on the basis of some characteristic … or other characteristic.”

    You can’t criticize me for writing that!!! I’m going to sue!

  307. @libreman:

    > So the question becomes: Is the ability to restrict others in itself a “freedom”?

    Yes. This is fundamental. Are you free to own property?

  308. @libreman:

    To expand a bit on Tom DeGisi’s post about freedom.

    Let’s say I write some software for someone else and give it to him. I have a few choices:

    1) I could tell him “this is a work for hire, I own the copyright and you can’t give this to anybody else.”

    2) I could tell him “here it is. Do what you wish. Keep it or give it to your friends.”

    3) I could tell him the same as (2), except “If you give it to your friends you better be sure to give them the source code on this other disk, or at least an offer to give them the source code on this disk. Actually, better give them the code to be sure, because if they give it away they have to make sure they do the same thing.”

    Now, in which case have I offered him the most freedom?

  309. >I submit that the government letting somebody license screws and telling me I can’t use a nail because I might hurt myself is, more literally than figuratively, placing me in a rubber room.

    Whoa! Where did the government come from? I told you this was clear libertarian bias and projection LOL

    Since when does the government force anybody to license anything under the GPL? It’s just as a voluntary contract as anything else. You want to use nails? Feel free to do so but not on the wood I sold you under a voluntary and mutually agreed contract that you would not. Make your own piece of wood if you do not like that contract.

    You see how you inserted government in there even there is no reason to? And coincidentally it’s on the side of the GPL! Which is perfectly equivalent of any proprietary license in legal terms. I told you this was like a religion, this is your proof …

  310. >Yes. This is fundamental. Are you free to own property?

    Are you proposing that abstract immaterial ideas/information like code is “property” without government decree of declaring it as such by fiat and then intervening in exclusive and mutually agreed transactions between two or more people exchanging those ideas/information?

    I anxious to hear an answer to this from a libertarian!

  311. @spqr> Oh, yah, all of esr’s credibility disappears because he shared the common view that DECUS’ tape distributions were mostly aimed at VAX users …

    Well, no, but it did take some shine off.

  312. > Whoa! Where did the government come from?

    Who do you think enforces the license on the screws vs. nails?

    > You see how you inserted government in there even there is no reason to?

    You said that open source is a moral issue. Whenever people do that, I grab my wallet, do a weapons check, and head to the voting booth.

  313. >>> So the question becomes: Is the ability to restrict others in itself a “freedom”?

    >>Yes. This is fundamental. Are you free to own property?

    > Are you proposing that abstract immaterial ideas/information like code is “property”…

    Stop right there and try again. I made no such proposal.

  314. @Patric

    3 of course! The fact that he distributes the sources along with the program does not put any restrictions on him whatsoever, it is trivial to do and he is free to do with the software what he wants for any purpose he wants (while not restricting others, not taking away the same freedom from others) as opposed to other two cases where in 1. he can’t do anything whatsoever with the code except give it to you or 2. where he is free to do whatever he wants but he also can take that freedom away from others therefore restricting them in a way similar to that in case 1 – that is not freedom that is power over others, you’ve made him into a micro government dictating to others down the chain what they can and can not do – just like governments do.

  315. >Stop right there and try again. I made no such proposal.

    Are we talking about software still?

  316. > Are we talking about software still?

    You asked if the ability to restrict freedom of others is a freedom. That seemed like a more general question than software.

    But if you want to restrict it to software, that’s fine.

    Getting back to my example of giving software to my friend (wrong answer, btw — he has to copy an additional CD)…

    IF I license the software under the GPL, then I have, in fact, restricted my friend. He cannot give away the object without giving away the source. You claim that this restriction provides a freedom, so you have answered your own question.

  317. >You said that open source is a moral issue. Whenever people do that, I grab my wallet, do a weapons check, and head to the voting booth.

    EXACTLY!!! This is exactly the mindset of a so called libertarian or free-market ideologue haha lol, you so nailed it! They run like hell at the mere mention of morals, I’ve said that earlier in the thread – you can search for it, I said that they wouldn’t mind even baking Jews if it made money lol it seems to be true. They despise any moral judgment by default and they’re proud of it! I’m not sure these creatures are even human! The more amoral and inhumane a system the more they like it.

    This is really sad that there are “people” like that but someone said that I shouldn’t attempt to read minds, what do you say now!? I read it perfectly, I know this mindset .. it’s extremely predictable. This is hilarious. I couldn’t satirize this better. Well done, well done my friend …

  318. >wrong answer, btw — he has to copy an additional CD

    What? So the sources don’t fit on the CD and the Internet doesn’t exist … okaaay, you’re beyond help – you’re generating nonsensical “problems” as you go. Have a nice day for now …

  319. libreman,

    > you can search for it, I said that they wouldn’t mind even baking Jews if it made money lol it seems to be true. They despise any moral judgment by default and they’re proud of it! I’m not sure these creatures are even human! The more amoral and inhumane a system the more they like it.

    What a completely repulsive, bigoted and ignorant statement. Patrick Maupin is an ethical and moral person. You, by contrast, don’t measure up to his demonstrated standard. When you reach his standard I will let you know.

    Yours,
    Tom

  320. >IF I license the software under the GPL, then I have, in fact, restricted my friend. He cannot give away the object without giving away the source. You claim that this restriction provides a freedom, so you have answered your own question.

    So now providing sources with the code is a “restriction”, I’m wondering how far are you willing to go to torture reason and English language, here is the definition for you:

    Definition of RESTRICT
    transitive verb
    1: to confine within bounds : restrain
    2: to place under restrictions as to use or distribution

    Examples of RESTRICT

    Her eye problem restricts her reading.
    She was told to restrict the amount of salt she uses.
    The new law restricts smoking in public places.
    They have accused the government of trying to restrict free speech.
    They say the government is trying to restrict them from speaking out.

    Do you see there the word “restrict” used in the twisted nonsensical way you used it? No? Well, that’s because it’s nonsense! If anything the requirement of including sources is an obligation not restriction, it doesn’t restrict you from doing anything … you can do whatever you want, it doesn’t ‘restrict’ your freedom in any way. And if you want to be so clever and say that it “restricts you from distributing the program without sources” that’s another mindfuck of using double negative that doesn’t actually mean anything. As far as the uses of the program go, GPL doesn’t restrict you any any way as long as you do not plan to restrict others.

  321. @libreman:

    > you’re generating nonsensical “problems” as you go. Have a nice day for now …

    A made up problem, is it? I’ve already mentioned twice that if you burn a Ubuntu CD for a friend and neglect to give him a written offer of source, you’ve violated GPL v3, and you’ve completely ignored that. Have you ever burned a CD for someone? Did you include all the source or a written offer?

    > I said that they wouldn’t mind even baking Jews if it made money lol it seems to be true.

    You’re a fucking idiot.

  322. >Patrick Maupin is an ethical and moral person.

    Psst, don’t say that! You’ve used the m-word – he is going to grab his guns and run to the nearest voting booth!

    But jokes aside, thanks for being so frank … I knew this was the mindset, I just never heard it said so directly :)

  323. >You’re a fucking idiot.

    The idiot is the one who despises morals enough to run away at the sound of the word and enables such atrocities by such a mindset. It’s you who does not mind people getting tracked and killed as a result because of using proprietary software with backdoors which they’re powerless to remove – do you think to moral issues associated with Free Software are not serious enough? Think again! It’s you who prouds himself of not caring about morals and shouts down people who do. So who is the idiot?

  324. @ libreman

    I have to assume at this point you are a troll, there is no way you can miss this:

    From your quote :

    Definition of RESTRICT
    transitive verb
    1: to confine within bounds : restrain
    2: to place under restrictions as to use or distribution

    From the GPL

    You may not propagate or modify a covered work except as expressly provided under this License. Any attempt otherwise to propagate or modify it is void, and will automatically terminate your rights under this License (including any patent licenses granted under the third paragraph of section 11).

    Sounds like placing a restriction on distribution to me.

  325. ‘GPL – it’s For the Children(tm)’.

    Make your moral judgements all you want, but don’t you dare use YOUR so-called morality as an excuse to take away MY freedom.

    It’s not often you see someone completely go off the rails (or unmask, if you’re being more cynical). Like walking into a room and seeing someone you think you know eating feces with a spoon. Jarring, it is.

  326. I may be the one who makes a “joke” of poor taste on an Internet forum (I originally used it as a extreme example because I was pissed and admitted so in the post right there) but you’re the one who actually enables similar atrocities by not caring about wider social issues. So don’t act like you’re all of a sudden outraged while just a moment ago you were Mr. “I grab my guns at the sound of morals” – you expose yourself as a freaking hypocrite of the first grade …

  327. @libreman:

    It’s you who does not mind people getting tracked and killed as a result because of using proprietary software with backdoors which they’re powerless to remove – do you think to moral issues associated with Free Software are not serious enough?

    Let me get this straight. The problem is not bad enough to mandate government intervention (no, no, no, you would never suggest that), but yet bad enough that when I suggest your proposed remedy is no good, I am somehow complicit in the killing of millions of innocents (for money no less), and a coward to boot. We certainly do inhabit different planets.

    In any case, I see you still haven’t answered the question I posed.

    Did you give a Linux CD to your friend and neglect to give an written offer of source, thus violating the GPL v3?

    Or do you simply not have any friends to whom you might give such a CD?

    Because I could believe either one.

  328. @tmoney

    This is exactly the kind of smartassery that I noted in the same post. It is a “restriction” if you really want to fit a square block into a round hole – feel free to do that and live in your virtual nonsense world. That doesn’t change the fact that distributing sources is not restriction in any practical sense in the real world occupied by real people who are not ideological zealots – it’s so extremely easy to do that any claim otherwise is just being a smartass and nit-picking for the sake of it.

    I’m done here, it’s pointless – you just make up total idiotic nonsense to justify your religious anti-GPL beliefs. It’s not reason, it’s dogma. esr resigned on defending his “refutation” probably because he realized that it’s bogus … well, that’s a shame. He doesn’t wade into the forbidden area of morals either … I suspect that’s exactly what is making him to advocate BSD-style licenses and oppose GPL – those damn morals, we need to get rid of them, it’s unfortunate that people think like that and it’s going to do a lot of harm to society.

    Certainly much more than bad references to the WW2 era.

  329. >Did you give a Linux CD to your friend and neglect to give an written offer of source

    It’s written on the CD you moron … there is no other word for this level of douchebaggery, I appologize, I’m pissed so I’ll swear more than healthy instead of actual arguments so I disconnect myself from this discussion right now (for your own protection, nonetheless, heh)

  330. liberman,

    > The idiot is the one who despises morals enough to run away at the sound of the word and enables such atrocities by such a mindset.

    You are, apparently, incapable of reading or reasoning.

    Yours,
    Tom

  331. Section 6 of the GPLv3 (https://www.gnu.org/copyleft/gpl.html):

    “You may convey a covered work in object code form under the terms of sections 4 and 5, provided that you also convey the machine-readable Corresponding Source under the terms of this License, in one of these ways:”

    ** in one of these ways ** !!!

    There are 5 of them of which ONE (1) is written offer (and it doesn’t say it has to be on paper). ANY of them will suffice you punk!

    I’m pissed, pissed, pissed at stupid ignorant people! GTFO DIAF you !#%&

    It’ this kind of ignorant people that fuck it up for everyone else, it takes just few of such idiots! And the best thing is that I’m going to be a loser in the eyes of everyone here even that I’m right because they’re apparently bigoted religious anti-GPL freaks – you can’t reason with that kind … aah, and I’m not helping by typing while I’m pissed as ever!

  332. > It’s written on the CD you moron …

    What do you mean “the” CD? Do you mean that you downloaded an ISO and burned a CD and wrote the offer in teensy, tiny letters?

    Or is it entirely possible that you’re just not very imaginative, and that’s why you can’t see why any of our scenarios could be real?

    Whether you have violated the GPL or not, can’t you just think for one minute that there are possibly thousands of well meaning people out there right now who helped their friends get started with free software and who are now in technical violation of the license?

    > I disconnect myself from this discussion right now (for your own protection, nonetheless, heh)

    Well, obviously if I killed a zillion people I deserve to die, so sure, be a martyr and save the next zillion. Unlike some brave people, this coward uses his real name and is actually pretty easy to find in the real world.

  333. @Patrick — Ubuntu CDs are covered under 6.c, or at least that’s the general consensus over in Ubuntu-land. I personally find words like “occasionally” to be troublesome when found in legal documents, but IANAL.

    I have no particular dog in this fight otherwise; I just wanted to clear up that particular gotcha. On the other hand, the point about large multinationals is a very good one.

  334. >> The idiot is the one who despises morals enough to run away at the sound of the word and enables such atrocities by such a mindset.
    >You are, apparently, incapable of reading or reasoning.

    Oh yeah?

    Patrick Maupin:
    >You said that open source is a moral issue. Whenever people do that, I grab my wallet, do a weapons check, and head to the voting booth.

    Does this stupidity fest ever end? esr was apparently right, 93% are incapable of grasping abstract concepts … they see words which they can find “repulsive” and act outraged but they do not see the concepts behind them at all. You’re part of those 93% apparently, congratulations!

  335. libreman,

    > And the best thing is that I’m going to be a loser in the eyes of everyone here

    Because you are behaving so badly.

    > I’m not helping by typing while I’m pissed as ever!

    True. Cool off and come back, and apologize for the extremely poor behavior, especially to Patrick. Don’t apologize for your arguments. As I’ve said, your arguments are good. They are worth making. They have flaws, like nearly every argument. They are worth discussion. They will probably not convince us. Our priorities are different.

    Yours,
    Tom

  336. liberman,

    I’m an ENTP. I’m in the 7%. Patrick, given his profession and clear competence at it is also in the 7%. Stop flailing around and cool off.

    Yours,
    Tom

  337. @Bryant:

    > Ubuntu CDs are covered under 6.c…

    I absolutely believe this. However, I am sure there are a lot of well intentioned people who have burned CD copies and never passed along the requisite copy of the written offer. Practically, this is no big deal. Technically, it’s a license violation.

  338. >That doesn’t change the fact that distributing sources is not restriction in any practical sense in the real world occupied by real people who
    >are not ideological zealots – it’s so extremely easy to do that any claim otherwise is just being a smartass and nit-picking for the sake of it.

    So in addition to ignoring the very real world example I provided you above as to how it is not always “extremely easy to [distribute source code to all down stream recipients]” you are seriously arguing that a license which explicitly states that you may not distribute the code or a modification of the code without meeting certain conditions isn’t applying a restriction? Perhaps I should give you a better definition than the one you presented.

    restriction |ri?strikSH?n|
    noun (often restrictions)
    a limiting condition or measure, esp. a legal one: planning restrictions on commercial development.
    the limitation or control of someone or something, or the state of being limited or restricted: the restriction of local government power.

    Are you suggesting the GPL does not place limitations on what someone may do with the given code?

  339. @Tom DeGisi:

    > You are, apparently, incapable of reading or reasoning.

    Thanks for the spirited defense. I’m sure my attitude is partly responsible for libreman’s attitude, but I also believe that he is capable of reading and reasoning, although his reasoning is unfocused while his reading is — well, let’s just say that although he expresses himself fairly well, I don’t think English is his first language, and I think he misses some of the more subtle points we are trying to express. I share your sentiment that if he slowed down a bit he might consider that perhaps he missed a few things (like that not every downloaded ISO burned onto a CD will have a written offer on it) and might come up with more reasoned arguments.

    BTW, I took one of those Myers-Briggs tests around 15 years ago or so, and they told me I was an ENTJ, but I never realized until esr mentioned it that so few people were XNTX. I’d like to see a bit more about what that means — all the standard info seems to be oh so politically correct.

  340. Patrick Maupin,

    > I also believe that he is capable of reading and reasoning

    So do I. He just isn’t showing it very well, most recently because he is pissed.

    Yours,
    Tom

  341. @libreman:

    That doesn’t change the fact that distributing sources is not restriction in any practical sense in the real world occupied by real people who are not ideological zealots

    Yes, the problem is the ideological zealots. They are unpredictable. And if their license says that someone could be in trouble for burning a CD and giving it to a friend without writing a little note (on the paper, on the CD, in email, whatever) that says “you can get source code here”, then — well, that someone probably isn’t in trouble.

    Unless, maybe, that someone is a big corporation. Or someone/something else the FSF could make an example out of. Which is why Tom DeGisi slowly and carefully explained to you that the GPL is a license that any sane corporation will handle like nitroglycerin.

    Bear in mind that someone who has violated the GPL in this manner hasn’t just done it for one program, or for just a handful of copyright holders. No, he’s violated the licenses of literally thousands of copyright holders, any one of which could theoretically sue him for statutory damages running into the hundreds of thousands of dollars for each such CD he hands out.

    Is it likely that a GPL copyright holder will sue and bankrupt someone? Nah. The odds are very low, about the same that my outspokenness about the evils of the GPL will cause someone to use Microsoft Windows instead of Linux or OpenBSD and get tortured by the Stasi because of that mistake. So I’ll work to educate people not to use Windows (oh, wait, I already do that! my part’s done) if you work to educate people about how they have to be really careful to dot the i’s and cross the t’s whenever they distribute software, whether it’s free or not.

  342. @Cathy
    > Oh, this so totally rocks…thank you!

    Quite welcome. It is a sanity-saver, especially on smartphone wars posts.

  343. @Patrick
    “No, he’s violated the licenses of literally thousands of copyright holders, any one of which could theoretically sue him for statutory damages running into the hundreds of thousands of dollars for each such CD he hands out.”

    In the USA, anyone can sue anyone for anything. And the suit can run for years without the need for neither wrongdoing nor evidence.

    That said, unchanged transmission of GPLed binaries where source is available elsewhere has never been prosecuted (or do you know some?). On the whole, there have been very few suits over breach of the GPL license. The whole “written offer” is archaic and ignored by everyone, including the courts. There have been two kinds of suits for breach of the GPL:

    1) Company A sells (embeded) binaries of a changed GPL program but refuses to make the source available, even after many requests. These were willful breaches of the GPL, either by A or by a supplier. There have never been rewarded statutory damages. This is the type of suits the FSF cs are involved with.

    2) Company B has contributed to a GPLed program and tries to extort company A who uses the program (MySQL comes to mind). In these cases, statutory damages have never been rewarded, only a retraction of the license (and small damages?). I do not even know whether there really has been a suit like this, or only settlements. The FSF has never participated in such suits.

    So, just FUD.

    @Patrick
    “Which is why Tom DeGisi slowly and carefully explained to you that the GPL is a license that any sane corporation will handle like nitroglycerin.”

    Most companies use Linux, perl, and/or Samba somewhere. Many companies contribute to Linux. So it does not seem it is the license is that toxic.

    I think the real reason corporations do not like the GPL is that they want to be able to rip off the users of their software at some point in the future. The first sample of $DRUG from your local dealer is often free.

  344. It may be interesting to look at a small, real life case of GPL and open source software, from someone who has no invested interest in the proprietary wars. Sorry if this has already been covered, since I’m late to the 400+ comment party. I originally came here to see if anyone has commented on the article http://venturebeat.com/2012/02/06/the-hacker-way-and-facebook/ that esr mysteriously alluded to in his post. I also apologize for the length but this is the condensed version. Feel free to read to the end for the punch line.

    Recently I released my very first bit of open source software. Nothing too big and it may not even sure it qualifies as software. It’s just a theme I made for the Cinnamon desktop environment. Like most themes out there, it’s basically a remix with some modifications to make everything fit together. Shouldn’t be too hard. Even though I would love a less restrictive license than GPL, most Gnome themes are released under the GPL. I thought all I had to do was slap the GPL license from the themes I modified, give attribution to the Copyright holders and I’m good to go.

    Nope. After spending hours on google, I’m still not sure I go the copyrights…, well right. had to create an image for the hot corner, so I scoured google for an image I could use. Turns out there was a good one I could use from Wikepedia but it’s offered under GFDL and under CC BY-SA 3.0, neither of which are compatible with GPL. Ok, so now I have two licenses. Not too bad. It’s just a bunch of separate files anyway. I can list the licenses separately.

    Oh, but what about the preview image? It includes a photograph I use for my wallpaper. Not only that but you can’t have a theme without screenshots to show how great it is. Would the screenshot be considered a derivative work? Well more googles and I find that whole flap about wordpress and themes and what I thought were just simple files to theme a desktop could be a whole lot more. But what if someone wants to use my theme to include in a commercial product? And what is commercial anyway? If I get a job because I develop into a mad themer, is the work commercial? What if I create a website, which showcases the theme, and I decide to put ads on the website to make some extra money? Sure that’s way outside the realm of possibility, but don’t laugh. Talk to music publishers and you’ll find out that they are in negotiations to make money off of ads on websites. And you know, once there’s money involved, that’s when lawyers come knocking. Who wants the hassle? The license isn’t listed anywhere on the website I took it from and the person who owns it hasn’t replied to my inquiry.

    So I went to search for wallpaper with an appropriate license. Ok, got an image that’s just CC-BY. Now I’m up to three licenses and four copyrights and I’ve spent more time on copyright then figuring out which rgba code to pass to which css property. So after all this, finding out that I can only guess at who wrote the Cinnamon port of one of the themes I based my own theme off of is just a bump in the road.

    Then, because I’m using this as an opportunity to teach myself how what it’s like to have something out there in public I need to figure out how to package it and distribute it on launchpad. Cool! Well, it looks like Debian understands that it’s easier to have machines figure out which copyrights apply to which software. That’s a good idea. Except that there’s no clear explanation for how I should include the text for the CC licenses.

    Now in terms of the theming and modding communities I probably went above and beyond what I needed to do. As far as I know, nobody is making huge money off of incidentals from their themes. In other words, I doubt it could be used the same way programmers could make money by using open source software experience as a marketing tool for their talents. But because I want my theme to be used as widely as possible, including the possible use in commercial products, I wanted to make sure I had everything covered. In fact I may have a feeling that I made things worse by being so diligent. I probably should have just included everything under GPL and gone about my day, but that would have possibly opened up problems for someone else.

    The point is that the GPL and open source is great. Without it I would never have the wonderful tools to make my system my own system. But it appears to me that the ambiguous copyright situation still pose barriers of entry for new participants. I wouldn’t be surprised if most modders and themers were very casual about noting copyrights and including attributions. Now there is the notion of “Who’s going to prosecute?” under which a lot of micro businesses use to get around regulations, i.e. they don’t follow the regulations because they could never hope to comply and make money and the risk of prosecution is minute. That’s a poor way to live. Surely there’s a more optimal system, especially in a software ecosystem that is based on many people making small contributions.

  345. @Patrick
    It’s an interesting exercise to extend the analogy to show how some of the GPL gene mutations (from v2 to v3) attempted to increase the fitness of the GPL license gene by allowing it to mate with even *more* licenses (and still always come out on top), yet one side effect of the v2 -> v3 mutation is that the mutated license can no longer interbreed with any software with the original v2 license, unless that other software also contains the “or later version” clause gene.

    This is the sort of thing that makes this poor modding hobbyist crazy. I’ appreciate that I have access to all of that wonderful GPL software, but I wonder how well people follow it. Without a bunch of googling, I don’t think people realize that GPL v2 is not compatible with GPL v3 unless it’s GPL v2+. And given how much people don’t read EULAs, I doubt most hobbyist coders read the details of the copyrights. Instead they think of it as a signaling mechanism and just slap that license on.

  346. There are 5 of them of which ONE (1) is written offer (and it doesn’t say it has to be on paper). ANY of them will suffice

    Just for due dilligence…. the 5 methods in Section 6 of the GPLv3 (https://www.gnu.org/copyleft/gpl.html) are :

    a) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by the Corresponding Source fixed on a durable physical medium customarily used for software interchange.

    Does the Ubuntu CD in question have sources on it? From memory the Debian sources came on a separate disk last time i looked at Debian so in my mind it’s certainly theoretically possible to hand a Ubuntu disk over without sources on it thus failing this method.

    b) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge.

    This is pretty much Patrick’s method. Stipulated as not done.

    c) Convey individual copies of the object code with a copy of the written offer to provide the Corresponding Source. This alternative is allowed only occasionally and noncommercially, and only if you received the object code with such an offer, in accord with subsection 6b.

    I’m not entirely sure what this is meant to mean. Considering Patrick’s stipulation was “no written offer” i think this one is out as well.

    d) Convey the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party) that supports equivalent copying facilities, provided you maintain clear directions next to the object code saying where to find the Corresponding Source. Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.

    Ok so how many people host the Ubuntu sources? Hands anyone? Yeah that’s what i thought. This one is out.

    “But Canonical hosts the sources so I don’t have to!” Sure but if canonical goes down then you’re contractually obliged to “ensure that it is available”. I don’t know what kind of idiot lawyer would suggest a company agree to that contract but I hope to hell they never advise me.

    e) Convey the object code using peer-to-peer transmission, provided you inform other peers where the object code and Corresponding Source of the work are being offered to the general public at no charge under subsection 6d.

    This doesn’t match the way we’re sending the object code (Handing a CD to someone doesn’t pattern match to “Peer-to-peer transmission” in my lexicon… though i bet a lawyer would argue it) and even so, the previous point applies (because this method specifically references 6d).

    From my reading of section 6 (IANAL) if you’re handing a CD over you’re pretty much limited to a) a written offer(I imagine any FSF lawyer would accept an electronic offer… but the license does say “written”) or b) putting the sources on the CD (or on an accompanying CD). TECHNICALLY anyone shipping a physical product (like say… a phone) should contain a written offer of source or provide said source in a physical medium with the product.

    Of course as mentioned(ad infinitum) the chances that anyone is going to sue over such a violation is slim to none. But it IS a technical violation and as Patrick says… if you’re talking about the CEO of a company then any decent lawyer is going to at least advise caution.

  347. Oh and lest I be charged with “moving the goalposts” regarding the handing over the CDs and CEOs of Companies…

    Technically speaking if you hand a CD over you don’t qualify for method D. More specifically you can’t hand someone a CD and offer the source on a web server and satisfy Method D. This is because for Method D to apply the source and object code must be received “in the same way through the same place” which if physical mediums are involved would qualify for A anyway.

    So there’s actually two arguments in my post.

    In terms of the original theoretical from Patrick, your options are A or B(maybe C). Given that B(and C) have been stipulated as not happening (no written notice) then you’d better hope there’s source code on that there CD.

    As a separate argument, A CEO of a Company could very easily receive cautionary advice if they are planning on using methods d and e as there are obligations there that might come back to bite them.

  348. @JonCB
    “Ok so how many people host the Ubuntu sources?”

    Just for clarity and precision.

    Passing on Ubuntu CD’s, that is, giving away CD’s you yourself were given or sold, does not fall under copyright. Unless you actually copy the bits in some form, you are not bound by the GPL.

    Think of a book you bought and pass on.

    Note that USING a program on a CD is covered by copyright laws (special cases etc.).

    Obviously, you got access to the sources by way of Ubuntu. If someone would sue you for the sources, you could in turn sue Ubuntu if they do not have the sources anymore. Or else, you just ask on the Internet. Someone will send you a link to the sources. On Groklaw, they tried to resurrect the sources of old Caldera distributions. And invariably, someone would pass over the sources. Then if you are sued, the judge would ask why the plaintiff would insist to force the accused to deliver the source, when she could get it from someone else. If the accused did not change the sources in any way, it will be difficult to get a lot of slack out of the suit.

    But that is all US legal intricacies, and no sane person expects the US courts to behave in a sane way.

  349. Another way to describe the GPL is as a purity law. GPL is designed to keep copylefted code from being sullied by contact with proprietary code. Lots of religions have purity laws. Puritans are named after purity. “Stallmanite” is humorous with this in mind. Religion is not the only reason for purity laws. Germany has a famous one for beer. All beer is restricted to having four ingredients only: water, malted barley, yeast and hops. So in Germany “free as in beer” is never “free as in speech”.

    Yours,
    Tom

  350. Winter,

    I know that companies use Linux and MySQL, etc, etc, a lot. We use Linux a lot where I work. But there are some open source freedoms we are too cautious to exercise with it. We don’t modify it. We don’t combine it’s source with our source. And we don’t distribute it outside the company. Inside the company we do our best to outsource the distribution to RedHat. We have legions of people who work on license compliance, and they are as careful to get RedHat licenses as they are Microsoft licenses. If you had read about the specific problems several of us mentioned you would know that we are complaining about lack of developer freedom, not user freedom.

    Yours,
    Tom

  351. I am rather baffled. Why wouldn’t a text file on the CD containing the text of Canonical’s offer of source code count as a written offer?

  352. @Tom DeGisi
    “If you had read about the specific problems several of us mentioned you would know that we are complaining about lack of developer freedom, not user freedom.”

    Sweeping statements were made (GPL is toxic). And I do not subscribe to this whole Mighty Developers vs lusers nonsense.

    What I develop, I develop exclusively under the GPL. That is easy. Everything I, and my colleagues, do is made available under a single license. The problems in a mixed environment are not due to the GPL, but due to the toxic nature of copyright law and other copyright holders.

    You can never ever mix code under two different licenses. You have to relicense all the code integrated in the work under a single license. What you want is to obey the proprietary licenses to the last iota and adapt the FLOSS code to that license.

    It is true that the GPL forces you to relicense the proprietary stuff under the GPL. And I am fine with that. Tit-for-tat is a very successful strategy. It is also very morally satisfactory. If you do not want to “work” for me, I see no point in working for you. This is exactly the position of Torvalds, who is no fan of RMS. The GPL still gives you every freedom to do with the work whatever you want, except to relicense it under a different license.

    And I look at discussions about Freer-than-Free like I look at “washes Whither-than-White” and holier-than-thou ads. This “Freedom to rip off my customers” discussion as a holier type of freedom is pointless. I saw the same polemics in earlier discussions on this forum against a “Duty to rescue”, where people feverishly defend the right (freedom) to not reach down to save a drowning baby. I consider such “freedoms” more as an evil fetishism.

  353. Then if you are sued, the judge would ask why the plaintiff would insist to force the accused to deliver the source, when she could get it from someone else. If the accused did not change the sources in any way, it will be difficult to get a lot of slack out of the suit.

    I wouldn’t think too difficult. The license is pretty explicit that you the distributor are responsible for providing the source and providing access to the source. You are allowed to point to a third party location, but if that location is down or removed, you are still responsible for locating and making the source available.

    Of course, the overall point is that this sort of uncertainty is exactly why large companies try to avoid the GPL as much as they can. In many cases it’s simply easier to use software that is really free.

  354. @Bryant
    “Why wouldn’t a text file on the CD containing the text of Canonical’s offer of source code count as a written offer?”

    It has to be signed and legally binding. Just a piece of binary text on a CD is not legally binding.

  355. @tmoney
    “The license is pretty explicit that you the distributor are responsible for providing the source and providing access to the source.”

    Indeed. But in any recompensation or punitive action, the fact whether the source is really unavailable would be taken into account. Ubuntu sources will be available for a long time from everywhere. If the plaintiff knows where he can download the source for free, the judge would consider his time wasted.

    Given the legal landscape in the USA, you can hardly move a finger without needing a lawyer. And the lawyer will tell you not to move the finger unless you absolutely have to.

  356. @ Winter

    >Given the legal landscape in the USA, you can hardly move a finger without needing a lawyer.
    >And the lawyer will tell you not to move the finger unless you absolutely have to.

    A good lawyer might, but there are plenty of trolls under that bridge looking to make money off scaring the common man (see the entire method of the RIAA prosecutions) , and honestly, as far as most people are concerned, it would be much simpler to avoid the GPL entirely, since a lawsuit, even if you win is damaging, expensive and can put you out of business and home. And to be honest, I don’t see that changing unless US law becomes a loser pays system, and even then I have my doubts.

    As I’ve mentioned before, I strongly suspect this is why Apple disallows GPL software in the App store, but has no problems in general with OSS software in store. They don’t want to be responsible for ensuring source code availability.

  357. @tmoney
    “I strongly suspect this is why Apple disallows GPL software in the App store, but has no problems in general with OSS software in store.”

    Without any real evidence, the story goes that in the early days, Steve Jobs while at Next tried to get around (an early version of) the GPL on GCC with a trick. They used the compiler, but did not want to publish the improvements they made. Either the GPL or GCC were amended to force Next to publish their improvements (I do not really remember which). Steve rather abandoned the GCC than to do that.

    Steve Jobs was so mad about RMS/FSF/GPL that from that day he tried to cleanse every company he lead from all GPL.

    As I wrote, that is the story, without any substantiation.

  358. Winter writes: “Passing on Ubuntu CD’s, that is, giving away CD’s you yourself were given or sold, does not fall under copyright. Unless you actually copy the bits in some form, you are not bound by the GPL.

    Think of a book you bought and pass on.

    Under US law, this is not actually correct. First of all, the book example does not apply because the first sale doctrine, which applies to books, does not apply to software distributed under a license. Secondly, distribution is itself an exclusive right in copyright.

  359. @Winter

    As I wrote, that is the story, without any substantiation.

    Yeah, and complete bullshit as far as I can tell.

    Apple has made substantial contributions to a number of open-source projects that are distributed under FSF-approved licenses, including WebKit, Darwin, and the LLVM compiler.

    The only reason you can’t put GPL stuff in the app store is that the *GPL* forbids it.

  360. @SPQR
    “Under US law, this is not actually correct. First of all, the book example does not apply because the first sale doctrine, which applies to books, does not apply to software distributed under a license. Secondly, distribution is itself an exclusive right in copyright.”

    IANAL etc. did not study law and I am not even in the USA.

    However, the GPL does not limit the user in using the software. So any limitations on anyone obtaining the CD under copyright law would not apply under the GPL. And I always understood that the first sale doctrine would apply to the mere passing on of CDs. All the other limitations normally placed into EULA’s etc are not in the GPL.

  361. @Joe Presley:

    Thanks for the interesting real-world example of what happens when you really try to play by the book.

    @Winter:

    Passing on Ubuntu CD’s… Think of a book you bought and pass on.

    You weren’t paying attention. I was discussing downloading an ISO and burning a CD.

    Most companies use Linux, perl, and/or Samba somewhere.

    Sure, but they just use binaries and never mess with or distribute the source.

    Many companies contribute to Linux. So it does not seem it is the license is that toxic.

    The benefits completely outweigh the risk, for many companies who really need to mod the OS. They just have to be careful.

    “So, just FUD.” “But that is all US legal intricacies, and no sane person expects the US courts to behave in a sane way.”

    Which is it? Seriously, it can’t be both, and you can’t seriously claim that zero zealots who would sue on a whim have ever contributed to GPLed software that winds up on a Ubuntu CD.

    I think the real reason corporations do not like the GPL is that they want to be able to rip off the users of their software at some point in the future.

    That’s because you are not paying attention and/or are sometimes apparently incapable of attributing positive motives to those you view as the enemy.

    This “Freedom to rip off my customers” discussion as a holier type of freedom is pointless

    That makes zero sense, unless you’re talking about the freedom to write some software for someone, take their money, and then tell them “oh, by the way, it was a work for hire, and I’ve GPLed it and I’m giving it to the world, and there’s nothing you can do about it. Oh, and by the way, don’t you ever think of giving anybody a copy without making sure you give them the source!” Yeah, if I were a naive customer who had neve had software written before, I’d be seriously pissed off at the way you ripped me off by doing that.

    But in any recompensation or punitive action, the fact whether the source is really unavailable would be taken into account.

    Have you been following the **AA saga? It can cost you tens of thousands of dollars in legal fees to defend against a copyright suit even when you’ve done nothing wrong at all, and the plaintiff has the wrong party. Suits can get a lot farther and cost a lot more when you’ve actually technically done something wrong.

    @Marco:

    Thanks for the link. I have to say that esr comes across much better than rms in that article :-)

    @Bryant:

    I am rather baffled. Why wouldn’t a text file on the CD containing the text of Canonical’s offer of source code count as a written offer?

    I would think it could. But (and it’s big but) it would have to be clearly visible to the user. Not sure you do that when you have no idea who the user is or what his competencies are.

  362. @SPQR:

    First of all, the book example does not apply because the first sale doctrine, which applies to books, does not apply to software distributed under a license.

    Is it really that simple? The GPL claims that, unlike most licenses, it doesn’t try to limit your rights, but rather adds extra rights that you wouldn’t have under copyright law. One of those rights is to make a copy if you follow all the rules.

    So if I make a copy and follow the rules and fix the copy in the physical medium of the CD, ISTM it’s a legally made copy, just like when a publisher who doesn’t own a copyright makes a book. Since the GPL purports not to take away any rights, if I give that legally made copy to someone, it’s theirs to do with as they please, including re-gifting it, selling it, etc. It’s only the making of additional copies that triggers the GPL.

    Now I have no idea whether this is right or not, but let’s have a thought experiment from the other side as well. If you carefully follow Apple’s EULA, then in theory you’re in violation if you buy a Macbook, open it up to set it up and install a few packages, and then wrap it back up and give it to your kid as a Christmas gift…

  363. Winter, most respectfully my friend, I would disagree. Within the US, the GPL does use the law of copyright and is a “license” to the software. As such, the first sale doctrine does not apply.

  364. Patrick Maupin, basically GPL licensed software is not public domain. Therefore, one does not have a “right” to use, distribute or anything else without the license ( the same principle that arises when someone violates the license and is deemed to no longer have a license ). And that the GPL itself enforces restrictions of any kind outside those of the statutory copyright law also take it out of being a “sale” and into “license”. It is therefore clearly licensed and not sold. No “first sale” ever occurred.

  365. All this talk re: GPL has been very centered on using a specific program and code base. As in, you get the source of a program that is GPLed and compile it, or else you get a binary with source. If you modify, rebuild, then distribute, you trigger GPL.

    We haven’t much talked about what I think is the really insidious aspect of GPL, and the one that raises my hackles in the biggest way: library linking.

    Whenever I’m working in C on my own software, and pulling in other libs, I have to remind myself constantly to double- and triple-check the licenses of those libraries. Because all it will take is one single slip-up and a link to a GPL lib at release time and I’m screwed; I’ve just lost control of my own code.

    Even LGPL is hairy, since I can’t static link if I want, which means my software has to live at the whim of other user’s systems and may run into all kinds of versioning problems.

    GPL/LGPL is more than just a legal PITA, it’s a very practical one too.

  366. @Patrick
    ““So, just FUD.” “But that is all US legal intricacies, and no sane person expects the US courts to behave in a sane way.”

    Which is it? Seriously, it can’t be both, and you can’t seriously claim that zero zealots who would sue on a whim have ever contributed to GPLed software that winds up on a Ubuntu CD.”

    You can be sued by the BSA for using Windows, and pay up again for all your installations even though all your copies were legally bought and paid for. You can be sued and lose to the **AA even though you never downloaded a single file. You can be sued and win, and still be bankrupted.

    You can use code under an MIT license, and be sued for copyright or patent infringement, and lose. The person who wrote the code might not be the “owner” of the IP in it (I am deliberately using the ridiculous term IP),

    So, getting sued and lose while you are innocent is simply a fact of life in the USA. There is nothing that makes the GPL special in this matter.

    In the USA, you can always be sued, by anyone, for anything. But I have yet to see an example that comes even close to what you present here as a risk. One of the contributers trying to sue you over the fact that you cannot supply stock standard Ubuntu sources is less likely than being sued by both MS and Oracle at the same time over the distribution of said Ubuntu code.

    So, unless you can show some real examples that tell us companies got sued for redistributing Ubuntu binaries, or any other distribution, I consider this FUD.

    The best counter example are CentOS, Scientific Linux, and Oracle, who all happily distribute Red Hat code. They just have to ensure that they also put up the sources. Which is easy. And if you cannot put up the sources, the worst that happened was you were asked to stop. Only when a company refused to make sources available and refused to stop distributing were they sued in the first place.

    Another reason companies are weary of distributing FLOSS has nothing to do with the license, but with the fact that they will get sued for patent infringement if they show the source. Because we all know that all non-trivial code is patented in the USA, often patented several times over.

  367. @Patrick Maupin

    If you carefully follow Apple’s EULA, then in theory you’re in violation if you buy a Macbook, open it up to set it up and install a few packages, and then wrap it back up and give it to your kid as a Christmas gift…

    Is this really true? I’m not saying you’re wrong (God knows it wouldn’t surprise me to find such a provision in a EULA), but would you mind quoting from the relevant section, or providing a link?

  368. Because we all know that all trivial code is patented in the USA, often patented several times over.

    Fixed that for ya, my friend.

  369. @patrick

    You’re a fucking idiot.

    Wow, we agree again. I was going to respond to some of his rants but like the samsung commercial that comment was a bit over the top.

    @winter

    That said, unchanged transmission of GPLed binaries where source is available elsewhere has never been prosecuted (or do you know some?).

    Yes, Verizon over BusyBox. They didn’t change the binaries provided by Actiontec given it was in the router.

    I really doubt Actiontec changed the BusyBox software in any way though. Either way, VZW got sued as a publicity stunt for the SFLC and the FSF.

    End result? BSD replacement projects for Busybox and GPL zealots are now whining about this.

    http://mjg59.dreamwidth.org/10437.html

    Rob Landley wrote this in the response thread which sums up the difference between coders and zealots. Rob is also the lead dev of ToyBox and former maintainer of BusyBox. There are lots of nice posts in the thread from Rob and Tim along with the usual anons saying dumb stuff.

    I’m the one who connected busybox with SFLC, because when I became BusyBox maintainer I inherited Erik Anderson’s “hall of shame” page, and couldn’t maintain that part of the project. So I went “if this is really an issue, let’s deal with it properly”, and asked around about pro-bono legal representation, and Pamela Jones of Groklaw gave me contact info for the SFLC.

    I started this process, and Erik went along with _me_. And the objective was to shake these guys down for THEIR BUSYBOX CODE. Not for other projects’ code, and not for money (although money is how you get their attention). I continued for a year, and at the end of that year we had not gotten a single line of busybox code added to the repository because of all this.

    I note that gpl-violations.org already existed at that point, with Harald Welte suing over kernel stuff before the SFLC was founded. We felt no need to handle that because somebody else was already doing it.

    The lawyers had a larger agenda, but the engineers didn’t. But their main agenda was to get paid: the SFLC made sure that everybody they contacted paid them $20k for legal fees, so they could afford to keep doing it. Whatever the outcome code-wise, the SFLC got paid, so of course they kept filing suits. The first I _heard_ of most of these companies was when the SFLC informed me we were suing them. (Presumably Erik had lots of complaints he’d never acted upon, and then there was the gpl@busybox address.)

    The SFLC never asked “what code do we get out of this”. They got their $20k. Whatever happened, the lawyers got paid, bring on the next lawsuit!

    This self-perpetuating lawsuit machine is what the FSF hijacked to stop all internal Linux development at Cisco. that had the potential of producing some great code, they were ALREADY planning to release it, and the FSF got those engineers reassigned and the division dissolved. (I was there when it happened, as was mirell.org.)

    From an engineering perspective, destroying Linksys was a BAD THING.

    You’re complaining about the loss of side effects you like. I’m pissed about A) the complete failure of the original purpose, B) massively counterproductive side effects that would have made the original purpose not worth it anyway.

    (Emphasis Mine)

    So from my perspective, I agree with Patrick. When someone tells me that software licensing is a moral issue I check my wallet because someone is selling something and it aint code and it aint in my interest as a dev.

    Then you have that whole MEPIS fiasco.

    http://lwn.net/Articles/193852/

    So the end result is, happily from my perspective, a decline in the use of GPL software in favor of more permissive open source:

    A new analysis of licensing data shows that not only is use of the GPL and other copyleft licenses continuing to decline, but the rate of disuse is actually accelerating.

    That was the conclusion of Matthew Aslett’s analysis of recent data from Black Duck Software, which shows that while use of the GPL, LGPL, and AGPL set of copyleft licenses dominates free and open source projects, that use is still on the decline.

    “The GPL family now accounts for about 57% of all open source software, compared to 61% in June,” Aslett wrote. More troubling for copyleft advocates, though, could be the projection Aslett and the 451 Group make based on the data.

    “…[I]f the current rate of decline continues, we project that the GPL family of licenses will account for only 50% of all open source software by September 2012.”

    http://www.itworld.com/it-managementstrategy/233753/gpl-copyleft-use-declining-faster-ever

    Cue Nelson’s HA HA from the Simpsons…

    1. @Nigel
      “Yes, Verizon over BusyBox. They didn’t change the binaries provided by Actiontec given it was in the router.”

      In this case, Verison bought BusyBox from Actiontec, and did not tell anyone it was GPLed, and the sources from Actiontec had never been made public. So Verison was just a front for Actiontec.

      “I really doubt Actiontec changed the BusyBox software in any way though. Either way, VZW got sued as a publicity stunt for the SFLC and the FSF. ”

      If Actiontec did not change the BusyBox software, why did Verison not simply dish them up?

      I suspect this was a case where Verison simply bought firmware from Actiontec, and Actiontec sold them BusyBox without telling them it was FLOSS, and never giving them the sources. Then, when the FSF came calling, Verison or Actiontec (or both) refused to answer and were humiliated in court. Verison could simply have sued Actiontec for the sources.

      So, this case is completely different. We are talking about a binary for which there was never any source made public. And it was never intended to be made public. The whole point of Actiontec was to keep the source hidden to undercut the prices of competitors.

  370. @SPQR

    Within the US, the GPL does use the law of copyright and is a “license” to the software

    Absolutely. Without copyright the GPL would be meaningless.

  371. Winter,

    > And I look at discussions about Freer-than-Free like I look at “washes Whither-than-White” and holier-than-thou ads. This “Freedom to rip off my customers” discussion as a holier type of freedom is pointless. I saw the same polemics in earlier discussions on this forum against a “Duty to rescue”, where people feverishly defend the right (freedom) to not reach down to save a drowning baby. I consider such “freedoms” more as an evil fetishism.

    I suspected you also enjoyed preaching hell and brimstone sermons! Welcome to the monkey house.

    Yours,
    Tom

  372. Winter, with respect to the Autodesk decision, I can see where you thought your argument made sense. However, the real holding in that decision is about a contract doctrine and its applicability to EULA. Its a decision about the lack of notice and consent to the contents of the EULA that I don’t think applicable in a GPL situation (where for instance no money changes hands) and the decision itself does not seem to be influencing other courts much. I would not place in any weight on it outside of its narrow facts.

  373. @Winter:

    One of the contributers trying to sue you over the fact that you cannot supply stock standard Ubuntu sources is less likely than being sued by both MS and Oracle at the same time over the distribution of said Ubuntu code.

    Yet another good reason to avoid the GPL, or shipping software with source in general :-)

    Seriously, I have showed that it is easy to technically violate a license. We both know that it is easy to sue for anything, but the part you are missing is that in most cases it is much easier to win, or at least prolong the lawsuit, on a technical violation, than on no reason.

    The best counter example are CentOS, Scientific Linux, and Oracle, who all happily distribute Red Hat code. They just have to ensure that they also put up the sources. Which is easy.

    And again you’re missing that somebody who receives a free CD and likes it and is told it’s OK to share won’t know all the intricacies.

    Another reason companies are weary of distributing FLOSS has nothing to do with the license, but with the fact that they will get sued for patent infringement if they show the source.

    At least this we agree on.

    1. @Patrick (not needed anymore?)
      “Yet another good reason to avoid the GPL, or shipping software with source in general :-)”
      So you prefer Open Source without Source? Rather pointless in my opinion.

      “And again you’re missing that somebody who receives a free CD and likes it and is told it’s OK to share won’t know all the intricacies.”
      I still think you are FUDing.

      Let us make your scenario explicit.

      G, copies an Ubuntu CD and distributes a few dozen of them in class. A parent of one of the kids, P, is a contributer and decides to make life hard for G. He demands source (the FSF can only sue when they are either a copyright holder or a receiver). Meanwhile Ubuntu has folded and their servers are off line. G has to deliver source on pain of a suit. Note that G has stopped distributing long ago. So this is only about a single, past, distribution.

      Where can G go?
      0 https://launchpad.net/ubuntu/+archivemirrors There are hundreds of mirrors for Ubuntu, they won’t all fold
      (because they will honor the GPL)
      1 http://distrowatch.com/
      2 http://www.linuxmint.com/ or any of the other derivatives (just email them and ask for the relevant sources)
      3 http://ubuntuforums.org/ Explain and ask for anyone to make a source CD image available
      (the forums are independent of Canonical)
      4 Any other forum where you might find Ubuntu fans, ask for the sources

      So, exactly, what where the risks again? That you can be sued frivolously? And you cannot for any other stupid reason?

      1. You finally admit that there is a technical license violation that numerous parties are committing daily, yet you refuse to admit that there are other equally bad things lurking in there for well-meaning people with deeper pockets, despite the evidence Nigel gave.

        Amazing.

        1. Patrick
          As my comment below indicates, passing on a legitimate CD is within 6C of the GPL. Furthermore, as an American you should be aware that you will technically break USA law almost by the act of breathing. The point in the USA seems to be not so much whether you break a law, but whether you can be prosecuted for it. Technically, every user of one of the BSD’s is violating some USA patent because it is inconceivable that there is no existing patent that covers the BSD’s. Also, it is inconceivable that any BSD user will be prosecuted for such a patent violation.

          The example Nigel gave (see my answer there) was about Verison selling a binary of BusyBox without license notice and source. They might not have even been aware of the fact that it was BusyBox they sold. Source was never published anywhere (I do not buy his idea that the sources were used unchanged). So this situation was in breech of the GPL from the very beginning.

          In contrast, your Ubuntu CD example is about a legitimate copy with ample opportunity to get hold of the source where the “person” just passes on this CD which should have the offer of source burned on it.

          So, I could be completely wrong on the legal side, and this could indeed be a (serious) technical violation of the GPL (IANAL etc.). But I have yet to see a credible reason why this person could (even technically) be prosecuted given GPL 6C, and the ample availability of the source.

    2. Also, I asked around and got the following response:

      The GPL says you may do this:

      c) Convey individual copies of the object code with a copy
      of the written offer to provide the Corresponding Source.
      This alternative is allowed only occasionally and
      noncommercially, and only if you received the object code
      with such an offer, in accord with subsection 6b.

      Properly released GPL programs all include licensing terms
      and offers to provide source in the documentation. So you
      are completely covered if you pass on an Ubuntu CD.

      As only the upstream developers have standing (in the USA), the offer should be on the CD, or the developer itself has been sloppy.

      1. I never at all disagreed with this. Look. You won’t find it.

        The Ubuntu CD has the written offer silk-screened on the top.

        I explicitly specified that someone downloaded an ISO and burned the CD.

        I even had to repeat this 27 times for libreman. So here we go once more:

        IF YOU BURN AND HAND SOMEBODY A UBUNTU CD, YOU’RE VIOLATING THE GPL UNLESS YOU GIVE THEM A WRITTEN OFFER OF SOURCE CODE.

        1. @Patrick
          With 400+ comments, I missed that part (Libreman’s posting were too long to read in my short life). My error.

          Burning is indeed worse.

          Not much of a problem in reality. If you get a request for Ubuntu source, it will be very easy to fulfill it even if Ubuntu folds. And I need a real lawyer tell me before I believe that you will be slammed with statutory damages just for not passing on an offer for source while you are able to deliver the sources on request.

          Btw, anyone trying to extract statutory damages must have their work and copyright registered. Good luck finding a GPL coder who has registered the copyrights to all his code.

  374. Nigel,

    You and Joe Presley are doing pretty well with the cautionary tales. Wow.

    Yours,
    Tom

  375. @Winter>

    Without any real evidence, the story goes that in the early days, Steve Jobs while at Next tried to get around (an early version of) the GPL on GCC with a trick. They used the compiler, but did not want to publish the improvements they made. Either the GPL or GCC were amended to force Next to publish their improvements (I do not really remember which). Steve rather abandoned the GCC than to do that.

    Steve Jobs was so mad about RMS/FSF/GPL that from that day he tried to cleanse every company he lead from all GPL.

    As I wrote, that is the story, without any substantiation.

    And it’s wrong. The GPL was not amended, Apple voluntarily complied with the GPL, and ever since RMS has been paranoid about any attempt to rationalize the intermediate form(s) used in gcc. On the whole, this has been to gcc’s detriment.

    In Stallman’s own words:

    Consider GNU Objective C. NeXT initially wanted to make this front end proprietary; they proposed to release it as .o files, and let users link them with the rest of GCC, thinking this might be a way around the GPL’s requirements. But our lawyer said that this would not evade the requirements, that it was not allowed. And so they made the Objective C front end free software.

    cite: http://www.gnu.org/philosophy/pragmatic.html

    So, again your frothy-mouthed story is hollow and false.

    But while we’re on the subject:

    Apple did start funding and using LLVM once GCC went GPLv3, but that is much more recent ‘history’ that you’ve created out of thin cloth above. In-return, the FSF/SFLC has been vicious about LLVM, and Apple’s choice to fund it.

    Consider this: http://www.groklaw.net/article.php?story=20100806143457345

    That’s Eben Moglen saying that Steve Jobs is “a man whose selfishness surpasses any recorded selfishness”, simply because Jobs opted to support a modular compiler started at the University of Illinois instead of the GCC.

    The LLVM’s license is free (and open), and even classified as such by the FSF. To smear participants in a fellow free software project like this or those that fund it is beyond low.

    In fact, the FSF says the license is GPL-compatible, which means they are not only free to fork LLVM if they think Apple’s stewardship of that project is risky, they can make their fork GPL.

    Another thing to consider: Apple employs the majority of LLVM developers, including the lead developer. If Moglen is correct, why is the new code these people are contributing, as Apple employees, still under the old license? Wouldn’t Apple be putting their new code under a less open license? And wouldn’t they be getting the authors to relicensing the existing code under a less open license, or rewriting it?

    The fact that Apple continues to put a lot of work into LLVM under a license that is very open (essentially a BSD license) is inconsistent with the motives Moglen ascribes to them.

    LLVM provides numerous purely technical benefits to Apple, including allowing for better graphics performance (which is an area that is important to one of Apple’s key customer demographics), allowing better integration with debuggers, profilers, and IDEs, and allowing easier evolution of Objective-C. Moglen is making it sound as if these technical benefits weren’t a factor, that rather: Apple is supporting LLVM just to undermine free software.

    First: he can’t give any example of them doing anything with LLVM that actually undermines free software. He can only speculate that they could do so in the future.

    Second: there is no rational basis for claiming that this hypothetical future ability to undermine free software is the sole (or even a major) reason Apple supports LLVM.

    Moglen is just spreading FUD, pure and simple.

    I’m sure the FSF feels threatened by the potential elbowing of GCC out of its dominant position in the market with a non-GPL’d component. One that can easily be “taken private” with enhanced, patent-encumbered, closed-source variants is not a win for the FSF. This could potentially deflate the giant, gas-filled balloon that is gcc. No doubt the FSF thinks its important that the most widely-used C compiler is Free Software.

    GCC is the primary “monopoly” GNU has, so it must be scary to see that go away.

    I’m pretty sure the FreeBSD folks welcome the choice.

    1. @Larry Yelnick
      “So, again your frothy-mouthed story is hollow and false.”

      Btw, I can see a “selfish” reason for Apple to fund LLVM as OSS: You do not get the best compiler without the most and best developers. To hire them all is awfully expensive (if not downright impossible). With OSS, you can recruit those who you cannot or will not hire. Eg, all those who are disappointed by GCC.

      Apple’s great GPL purge
      http://meta.ath0.com/2012/02/05/apples-great-gpl-purge/

      There are two big changes in GPL v3. The first is that it explicitly prohibits patent lawsuits against people for actually using the GPL-licensed software you ship. The second is that it carefully prevents TiVoization, locking down hardware so that people can’t actually run the software they want.

      So, which of those things are they planning for OS X, eh?

      I never assume Apple acts out of the best motives, never. And I have never been disappointed.

      PS: I never froth at the mouth, except sometimes when I brush my teeth :-)

    2. @Larry Yelnick
      “So, again your frothy-mouthed story is hollow and false.”

      Btw, I can see a “selfish” reason for Apple to fund LLVM as OSS: You do not get the best compiler without the most and best developers. To hire them all is awfully expensive (if not downright impossible). With OSS, you can recruit those who you cannot or will not hire. Eg, all those who are disappointed by GCC.

      Apple’s great GPL purge
      http://meta.ath0.com/2012/02/05/apples-great-gpl-purge/

      There are two big changes in GPL v3. The first is that it explicitly prohibits patent lawsuits against people for actually using the GPL-licensed software you ship. The second is that it carefully prevents TiVoization, locking down hardware so that people can’t actually run the software they want.

      So, which of those things are they planning for OS X, eh?

      I never assume Apple acts out of the best motives, never. And I have never been disappointed.

      PS: I never froth at the mouth, except sometimes when I brush my teeth :-)
      (incorrect tags in my previous version, sorry)

  376. @Tom:

    > Is this really true?

    I don’t have the info handy, but back when groklaw was covering the whole Apple vs. Psystar fiasco and I was reading Apple court documents and the source EULA, that was the conclusion I came to after a lot of thought.

    @SPQR:

    I certainly agree with you that anybody relying on first sale for software is an idiot, just like anybody relying on fair use. But even though the result from Davidson & Associates was “The first sale doctrine is only triggered by an actual sale. Accordingly, a copyright owner does not forfeit his right of distribution by entering into a licensing agreement.” I am not sure the law is fully settled.

    Look at UMG vs. Augusto, where transfer of a CD for zero money, with a big scary “not for resale” on it was nonetheless deemed a first sale. Although it relates to music promo CDs, some of the parallels with software CD distribution seem uncanny, and, I might be wrong about this, but I don’t remember Davidson or similar cases arguing anything about software that makes it significantly different than music…

    Although this entire discussion will be mooted when all software is delivered through the cloud and nothing is fixed in a physical medium ever again :-)

  377. @tom

    Yeah, and complete bullshit as far as I can tell.

    Apple has made substantial contributions to a number of open-source projects that are distributed under FSF-approved licenses, including WebKit, Darwin, and the LLVM compiler.

    The only reason you can’t put GPL stuff in the app store is that the *GPL* forbids it.

    Not to mention gcc itself. With much whining from some folks about Apple not assigning copyright to the FSF. Pretty much all the objC support in gcc is contributed by Apple…for obvious reasons given the Apple ecosystem is the primary (if not only major) user of ObjC.

    http://lwn.net/Articles/405417/

    @esr

    Curious about your thoughts about Matthew’s comments. I know you and Rob wrote the world domination paper together.

    It feels weird calling people I’ve never met by their first names but this is the internet…

  378. @Larry Yelnick:

    The fact that Apple continues to put a lot of work into LLVM under a license that is very open (essentially a BSD license) is inconsistent with the motives Moglen ascribes to them.

    I actually think it is completely consistent with motives that Moglen would find offensive. Namely the ability to use LLVM inside software where Apple doesn’t have to show all their source code. Apple is pragmatic about LLVM and CUPS and other packages, both about keeping their developers happy, and about where their value add actually lies, and thus, where secret sauce is needed. Moglen would prefer to deny them the ability to have any secret sauce or value add.

  379. @Patrick — “I would think it could. But (and it’s big but) it would have to be clearly visible to the user. Not sure you do that when you have no idea who the user is or what his competencies are.”

    I am pretty sure there’s the legal concept of reasonable. I.e., you don’t have to worry about the case of the illiterate user or the user who isn’t capable of understanding the location of the source code even if the document is right in front of his eyes. But IANAL, so I am on shaky ground and probably won’t pursue the discussion further. Particularly since I don’t much like the GPL and wouldn’t use it for my own stuff unless I had to.

  380. @Larry Yelnick:

    I’m sure the FSF feels threatened by the potential elbowing of GCC out of its dominant position in the market with a non-GPL’d component. One that can easily be “taken private” with enhanced, patent-encumbered, closed-source variants is not a win for the FSF. This could potentially deflate the giant, gas-filled balloon that is gcc. No doubt the FSF thinks its important that the most widely-used C compiler is Free Software.

    GCC is the primary “monopoly” GNU has, so it must be scary to see that go away.

    This. Remember the whole Tivo-ization clusterfuck? Avoiding that in the future seemed to be the primary rationale (well, there was additional patent protection, too) behind the painful transition to GPL v3. And the end result is that Tivo can’t put a compiler on their box. Poor Tivo. The ability to do that is so important to them that they are heavily funding LLVM. Oh, wait — I get them confused with Apple sometimes.

    Anyway, not only will the FreeBSD folks welcome the choice, as you point out, but the Linux pragmatists who stuck with v2 will probably welcome the choice as well.

    As Nigel points out, GPL use is declining. I’m not sure that would be happening absent the revision to v3.

  381. Although this entire discussion will be mooted when all software is delivered through the cloud and nothing is fixed in a physical medium ever again :-)

    It has been my assertion that even the FSF was afraid of Google and hence the removal of the Affero parts in GPL V3.

    If everything is in the cloud then the FSF will have no choice but to release GPL V4 with the Affero parts as mandatory.

    That will be a massive popcorn event and I expect Google to crush them in ways that would make Microsoft hugely envious.

    And then you Googlites will understand the true power of the dark side and whom the Sith really are (ie not Apple).

  382. My apologies for dropping the smiley. “You Googlites” is intended as playful as I would hope the Star Wars reference might indicate.

  383. @Nigel:

    > And then you Googlites will understand the true power of the dark side and whom the Sith really are (ie not Apple).

    I don’t think that an Affero GPL v4 will be The Revelation.

    Oh, sure, google will reveal its power if it needs to.

    But after the last GPL license rift, people will approach v4 more cautiously, watching to see if others are taking it up. Google won’t have to do anything unless it starts to get major traction on packages that google uses, which means (IMO) that google won’t have to do anything.

  384. AUGH! Total site look change! And now the HTML tagging is different and my killfile extension is rendered useless! (Of COURSE it happens within 24 hours of a ‘release.’)

  385. What the blazes?!

    Site looks all different!

    Where’s that comforting picture of Eric gone from the top right?

    This feels all wrong ;)

  386. For what it’s worth, LLVM is a massive win. One of the things I work on is a 900KLOC C++ package. Building it from a clean start with Xcode 3.2.5 (gcc 4.2) takes about 40 minutes on my 8-core desktop. Building it from a clean start takes 19 minutes on my 4-core laptop with half the RAM, in Xcode 4 with LLVM. The resulting binary is 10% smaller and runs 20% faster.

    gcc can go take a flying leap.

  387. @Jay Maynard:

    I knew they were getting there, but that’s still impressive. And LLVM is much easier to work on than GCC as well, so between the ease of modification and the higher performance, it can expect a lot of developer attention.

    This will be a good test — AFAIK there’s nothing preventing the GCC people from stealing whatever parts of LLVM they want to use to beef up GCC. It will be interesting to see how this shakes out and which codebase becomes dominant. (Obviously it’s a closer call whether someone creates Fortran or ADA frontends for LLVM, so GCC’s not going to die a quick painless death in any case.)

    FWIW, I think LLVM would have gotten to this point without Apple (with some small risk that some of the LLVM principals would have abandoned the project when they got employment doing something else), but that Apple’s money and involvement greatly sped up the process.

    1. This will be a good test — AFAIK there’s nothing preventing the GCC people from stealing whatever parts of LLVM they want to use to beef up GCC.

      I don’t know whether the development of GCC is as tightly controlled as it used to be in the late 90s. If it is then I suspect they might look down on the LLVM code and not incorporate it for religious purity reasons. Which would be stupid.

      Like you say, this will be a good test.

  388. WRT the original article… DAMN that journalist has just sprung to pretty high on the heap in my opinion.

    I expected some chopped liver of a massacred quote since it didn’t seem like ESR’s response was very “quotable”. Instead I got the (happy) surprise that it is faithful in both the quote and the editorialisation surrounding it.

    Full marks to Jolie O’Dell.

    1. >Instead I got the (happy) surprise that it is faithful in both the quote and the editorialisation surrounding it.

      I suspect Ms. O’Dell came to me with a pretty clear notion of what sort of critique I would write, and wanted exactly that.

  389. @Tom:

    Is this really true? I’m not saying you’re wrong (God knows it wouldn’t surprise me to find such a provision in a EULA), but would you mind quoting from the relevant section, or providing a link?

    It looks like you might be in the clear if you sit down with whoever you give the computer to and say “honey, you have to read this entire license and accept it.”

    See section 1, which says among other things, that you don’t even own a copy of the boot ROM, and section 3 gives the conditions of transfer, with 3(c) explaining that the recipient has to read and agree to the license:

    http://images.apple.com/legal/sla/docs/macosx106.pdf

    Pretty bogus if you ask me, but just like the FSF isn’t going to sue an end user who shares a CD with another end user, Apple’s certainly not going to sue anybody for using an Apple computer without a license, although if you look on the bulletin boards, you will find lots of examples of people who had trouble transferring stuff (especially OS-upgraded computers and additional software packages), because apparently a lot of Apple software phones home these days and they know who owns which computer and software package.

    1. See section 1, which says among other things, that you don’t even own a copy of the boot ROM

      It says you do not own outright, but are licensed to use, the Apple software (including the boot ROM and all other system software and packages). I think this is standard for all software. You are not going to find any proprietary software that is sold and not licensed.

      section 3 gives the conditions of transfer, with 3(c) explaining that the recipient has to read and agree to the license

      Section 3 is a natural consequence of section 1. If you don’t own the software you can’t sell it. But it says you can transfer your rights as a licensee to somebody else provided they accept the EULA. They accept the EULA just by using the software, so that is no problem.

      This stuff is completely standard.

      you will find lots of examples of people who had trouble transferring stuff (especially OS-upgraded computers and additional software packages), because apparently a lot of Apple software phones home these days and they know who owns which computer and software package.

      Before selling your computer you should erase the hard drive and restore it to its original condition. Common sense.

      1. > They accept the EULA just by using the software, so that is no problem.

        Section 3 explicitly says you need to get the party you are transferring the computer to to read the license.

        1. Yeah, and every software license in the world ever written says that *anybody* who uses the software is supposed to read the license. This is not new.

    2. because apparently a lot of Apple software phones home these days and they know who owns which computer and software package.

      Interesting, as far as I’m aware, no Apple software phones home since no Apple software (save old versions of FCP and Logic) have any license keys or dongles associated.

      Admittedly, I haven’t looked into what happens if you resell something with software bought from the App Store, but I assume that the software will continue to work and exist on the computer for as long as the computer remains unerased. At which point obviously anything that the new owner doesn’t have purchased themselves would be unavailable to them (but equally, anything they have purchased would be, including OS upgrades)

    1. My thinking is, a killfile keeps me sane by not showing me the posts of the worst offenders; if others reply to those posts, then I anticipate they will reply to the most useful or interesting parts.

      In other words, by killfiling I am letting others filter the discussion for me. Works pretty well.

  390. Does anyone know whether a Ubuntu .iso does in fact include a text file with an offer of source?

    My memory is a little hazy (been a coupla years since I used Ubuntu) but did the “About” menu items at least used to include statements about obtaining source?

Leave a comment

Your email address will not be published. Required fields are marked *