Hackers and anonymity: some evidence

When I have to explain how real hackers differ from various ignorant media stereotypes about us, I’ve found that one of the easiest differences to explain is transparency vs. anonymity. Non-techies readily grasp the difference between showing pride in your work by attaching your real name to it versus hiding behind a concealing handle. They get what this implies about the surrounding subcultures – honesty vs. furtiveness, accountability vs. shadiness.

One of my regular commenters is in the small minority of hackers who regularly uses a concealing handle. Because he pushed back against my assertion that this is unusual, counter-normative behavior, I set a bit that I should keep an eye out for evidence that would support a frequency estimate. And I’ve found some.

Recently I’ve been doing reconstructive archeology on the history of Emacs, the goal being to produce a clean git repository for browsing of the entire history (yes, this will become the official repo after 24.4 ships). This is a near-unique resource in a lot of ways.

One of the ways is the sheer length of time the project has been active. I do not know of any other open-source project with a continuous revision history back to 1985! The size of the contributor base is also exceptionally large, though not uniquely so – no fewer than 574 distinct committers. And, while it is not clear how to measure centrality, there is little doubt that Emacs remains one of the hacker community’s flagship projects.

This morning I was doing some minor polishing of the Emacs metadata – fixing up minor crud like encoding errors in committer names – and I made a list of names that didn’t appear to map to an identifiable human being. I found eight, of which two are role-based aliases – one for a dev group account, one for a build engine. That left six unidentified individual contributors (I actually shipped 8 to the emacs-devel list, but two more turned out to be readily identifiable within a few minutes after that).

I’m looking at this list of names, and I thought “Aha! Handle frequency estimation!”

That’s a frequency of just about exactly 1% for IDs that could plausibly be described as concealing handles in commit logs. That’s pretty low, and a robust difference from the cracker underground in which 99% use concealing handles. And it’s especially impressive considering the size and time depth of the sample.

And at that, this may be an overestimate. As many as three of those IDs look like they might actually be display handles – habitual nicknames that aren’t intended as disguise. That is a relatively common behavior with a very different meaning.

122 thoughts on “Hackers and anonymity: some evidence

  1. And it’s amazing how so many people consider my ham radio call sign, issued by the FCC and readily linked to my real name on the FCC database, a “dodgy” hacker handle…

  2. I don’t think that actually tells you much. You are attempting to categorize people into subcultures, while only looking at data for a single (possibly very small) part of various people.

    It seems very plausible to me that someone could use their real name to work on Emacs while using a concealing handle in many other aspects of their online existence, and you have no evidence either way.

  3. Indeed. I’ve been using my 3 initials since I had to come up with a login name for my first time sharing experience, and V6 UNIX(TM) in the summer of 1978 (sic). I’ve gotten into at least one tiff with a small startup that insisted my handle be a mechanical function of my first and last name, and the attitudes behind that also caused them to fail miserably.

  4. Allow me to present a counterexample: the community that develops the various third-party viewers for Second Life. Out of that entire community, I know of exactly one person, the leader of the Firestorm project, who does not use her Second Life name exclusively to identify herself. (And she’s not a programmer; you’ve commented before on how unusual this is.)

    TPV developers use their SL names both because that’s how they’re identified in the world of SL, and because many of them do not wish their SL identities to become linked with their RL identities. (I know three people who do not keep them strictly separate, and they are very much the exception.)

    Are these folks hiding behind handles, or…?

  5. I’m wondering if there is any correlation between code quality and anonymous commits?

    Me? I always use my real, verifiable, identity. I’ve been using my full real name for EVERYTHING on the Internet since 1997/1998.

    But my original reason for making that change was my wish to find things I did or wrote because searching for different handles is a pain in the ass. Especially when some of the handles are also used by other people. To this day I think I am the only ‘Jack William Bell’ getting indexed by the major search engines; all the counter examples are things like “Steven Jack, William Bell, Roger Grayson, . . .”

    I also figure you shouldn’t say anything on the Internet you don’t want your boss or your mother to read, because even with a handle it could get traced back to you and because maybe you shouldn’t say those things anyway.

  6. This stuff makes me think about the magical concept of the “true name”. In contexts where it is commonly used, true names are sources of power and control. Knowing your true name can give others power over you, so you are circumspect about just who knows it, and the name you give out to the public is an alias. (And this is precisely the reason the “hackers” the media goes on about use aliases – the alias is known in their community and can accrue status, but the true name is a carefully guarded secret from *everyone* because the society in which that hacker lives disapproves of what they do and will take action against them if their true names are known.)

    There are any number of cases of “display handles” in Internet usage. For practical purposes, they can *become* the user’s name – the identifier by which they are known to the world. Indeed, one of the early gripes about Google+ was the insistence on “real names” as opposed to aliases, as many folks had established display aliases they used everywhere else which was how they were known to the online community, and simply wanted to use them on G+ as well. (It took a bit, but Google finally got the hint and permitted aliases.)

    Display handles are certainly common in open source projects. A difference is how easy it is to discover the “true name” of the person using the handle, with the corresponding question of why you might *need* to. In software development, you need to know what committer made what changes when and for what reasons. How often do you need to know the committer’s true name or location?

    I’d be curious about the time distribution of the IDs, and when they were first established. I’d expect usage of an alias rather than a real name to increase over time Back in 1985, the notion of using an alias and reasons for needing to do so were less common. I’d also be curious to see a graph of when IDs were established. I’d expect a curve with a period of maximum committer ID generation that gradually fell off.

    You commented elsewhere that the project lead was behind your conversion to a git repo to make it easier for new people to get involved and become committers. The automatic question that raises is “Yes, it will, but how many new people *want* to, and have been scared off by the friction built into doing so?”

  7. @Frank Ney, in my anecdotal experience, the is a significant overlap between the hacker culture and ham radio operators, though this may be generational. Many of them regularly use their callsigns as display handles, and at least one named their open source project after it (Phil Karn).

  8. >Are these folks hiding behind handles, or…?

    Partly. But it’s also, from your account, a display behavior by which they connect their identities as developers with their identities in Second Life – which, for this purpose, may be more important and more worth advertising than their real names. This behavior is revealing from one angle, potentially concealing from another.

    In fact what this is telling us is that to these developers their rep in Second Life is more important to them than their rep among hackers. Thus, this group is functionally on the periphery of the hacker culture at best, and arguably outside it.

  9. >You commented elsewhere that the project lead was behind your conversion to a git repo to make it easier for new people to get involved and become committers. The automatic question that raises is “Yes, it will, but how many new people *want* to, and have been scared off by the friction built into doing so?”

    That’s something we can’t know in advance. But not knowing this in advance shouldn’t stop us from running the experiment, because your number doesn’t have to be large to have a significant benefit for the project.

  10. > (It took a bit, but Google finally got the hint and permitted aliases.)

    That’s news to me. Do you have a source?

  11. I think in the modern day of google-able identities, there is some call for ‘handles’ or stable pseudonyms. My legal name is moderately uncommon, and as such, googlable. I use my proper name for technical topics, linked-in, and facebook (where I have restricted the privacy controls).

    For blog comments or other social media, I use a pseudonym to separate my political and religious opinions from anything that appears on a resume. Other people have personal history or other reasons to not be google-able, while still maintaining an online identity.

    I think the key is stability. I picked my pseudonym in college, and I have never changed it.

  12. >I use my proper name for technical topics, linked-in, and facebook (where I have restricted the privacy controls). [...] For blog comments or other social media, I use a pseudonym to separate my political and religious opinions from anything that appears on a resume.

    The key point, it seems to me, is that you remain accountable under your own name for the code that you write and its consequences. The fact that you have a concealing handle that you use elsewhere is less relevant.

  13. @random832: It was widely discussed on Google+, and should be easy enough to Google for. I use my real name there, didn’t care about the issue, and haven’t saved links. (I understood why those complaining were unhappy, but the requirement didn’t bite *me*.)

  14. Currently, contributing (non-trivial) code to Emacs requires a copyright assignment to the FSF (https://www.gnu.org/software/emacs/CONTRIBUTE). I’m not sure how long they’ve had that policy, but I would imagine that the policy would filter out contributors who might like to have an anonymous handle. I could also imagine folks who habitually use an anonymous handle using their name for Emacs development, seeing as they have already identified themselves in the copyright assignment step. Either way, it may result in a lower than average usage of anonymous handles.

  15. The only exception to this that I can think of would be if you were a hacker bound within the confines of a suppressive sensorship driven regime like say, Iran. Especially if you were coding for a project like TOR in which case revealing your real name could not only kill your tube time, but your very life as well.

  16. Part of the problem here is that you are using “hacker” in what is probably its original meaning, where a “hack” is an ingenious and unconventional solution to a problem—the elephant’s trunk being a prime natural example. People who expect hackers to want anonymity are likely to be using the term with its currently popular meaning of people who break into other people’s computers, presumably by a false back etymology to “hack into.” One would expect very different behavior in the two groups.

  17. @esr: Agreed, we can’t know in advance. But the distribution of *when* committer IDs have been created would have bearing.

    In larger terms, it becomes a question of market share. I’m not sure it’s possible to get meaningful stats of who uses what editor. There are *many* editors which have been developed (I’m aware of 1,780 – total http://texteditors.org/cgi-bin/wiki.pl?EditorIndex – but there are doubtless some I’ve missed.)

    Emacs has been around a *long* time, and it and vi are two of the oldest still in use editors in existence. But new editors appear all the time, become popular, and build audiences at the expense of older established editors. At a guess, the two most popular open source editors these days are Bram Moolenaar’s Vim and Don Ho’s Notepad++ for Windows, and they are likely to be the most often recommended when someone inquires “I’m not happy with editor X. What should I use instead?”

    (And there is a new NeoVim project which hopes to refactor and simplify Vim’s code base and standardize on Lua as the extension language to make it easier foir new folks to become involved in the product’s development.)

    The question “How many new folks will want commit access to Emacs sources?”, will be affected by “How many new people have shifted *to* Emacs from something else?” My suspiscion is the answer to the latter question is “Not many.” My feeling is Emacs has an existing market share that is stable but highly unlikely to grow significantly.

  18. >People who expect hackers to want anonymity are likely to be using the term with its currently popular meaning of people who break into other people’s computers

    Of course you’re right. But remember the premise of the OP – I want to use the distinction between transparency and concealment to educate them about who hackers really are. Thus, pointing out that they’re proceeding from the wrong definition simply restates the problem rather than assisting with a solution.

  19. >Either way, it may result in a lower than average usage of anonymous handles.

    Possibly. On the other hand, the Emacs 1% rate is consistent with my rather broad experience of other open-source projects without copyright assignment, which is that concealing handles are very very rare. This fact is obscured by the higher incidence of display handles.

    Eric thinks…hm, IIRC the Subversion project has a large contrib list. I’ll go look…..looks like about 170 committers, nothing that looks like a handle on any of them. That’s a lower incidence than Emacs. What other large projects can you think of that we could audit?

    I have a general impression that we might find a higher incidence of both kinds of handles in game projects.

  20. >The only exception to this that I can think of would be if you were a hacker bound within the confines of a suppressive sensorship driven regime like say, Iran. Especially if you were coding for a project like TOR in which case revealing your real name could not only kill your tube time, but your very life as well.

    Yes. I think most hackers have a firm grasp on this category of exception.

  21. >I have a general impression that we might find a higher incidence of both kinds of handles in game projects.

    What’s your experience regarding Wesnoth?

  22. >What’s your experience regarding Wesnoth?

    Fair number of display handles, few or no concealment handles. I’d audit, but the only equivalent of an AUTHORS file is in a format that’s a pain in the ass to parse.

  23. I guess as a “display” handle, I’ve got a login I still use a couple places that’s been around since the very early 90’s – one of two I used in college systems.

    Otherwise, I generally use my name. First initial and last name is identifying enough since I believe I personally know everyone with my surname in the US.

  24. >First initial and last name is identifying enough since I believe I personally know everyone with my surname in the US.

    I wouldn’t be a bit surprised. What language does it come from?

  25. Lithuanian. Grandad came over post WWII one step ahead of the Soviets. Interestingly, his name “Vladas” got the Ellis island treatment, and became “Walter”. Settled in Mass and among other things worked on some apollo-era camera optics.

    So we have his three kids (including my dad) and their various kids and grandkids.

    One of which, my son, is up your way at Upenn…

  26. esr: ” What other large projects can you think of that we could audit?”

    I think Linux would be a good example. Maybe Apache as well.

  27. >Lithuanian.

    Guess confirmed – I suspected it would be something Baltic! Helps that I’ve actually been in Lithuania.

  28. Sometimes a handle isn’t an attempt to hide anything, it’s just what people call a “brand” nowadays. They were common in the British magazine industry right into the late 1960s. though apparently never very popular in the US. Mostly, there was no secret about who the writers in “Model Engineering” or “Popular Cycling” were, but noms de plume were what were on their articles.

  29. Hope it was a fun trip. I haven’t – though my dad spent a few years consulting after retiring from the Marines (I was in the Navy at the time), and so my younger siblings spent some time there.

    Lithuanian surnames tend to stand out when you know what to look for – though they can at times be confused for Greek (or vice versa).

    Probably the strangest “small world” moment I’d ever had was going to the touristy cheesesteak corner when we were looking at colleges with my son, and a govt. van pulls up. A bunch of people hop out – some having that not-quite-american look to their clothes that europeans have – and one or two in army uniforms. After overhearing the Lithuanian I introduced myself, only to discover that a couple knew my other grandad (who moved back and has since passed on), and a couple had worked for my dad while he was there.

  30. @dgarsys: I use first initial, last name as an online handle too, with little fear of confusion. There *aren’t* all that many McCunneys. Great grandad came over from Ireland and set up as a parish undertaker in Phila. Grandad married out of the faith and was promptly disinherited, so I’m from the black sheep branch of the tiny clan and there are McCunneys back in Phila. I don’t know at all. There’s even another Dennis McCunney, but he’s not active in any area I am. I got a PM on a website where I’m a mod asking if I was the Dennis McCunney that lived in Phila, and was involved in science fiction. I replied that if there was *another Dennis McCunney that lived in Phila. and was involved in science fiction, it was news to me, so yes. It turned out the asker was an ex-girlfriend from way back I was quite happy to *not* be in contact with for both our sakes. She was going to be in NYC for a visit and proposed lunch. Didn’t actually see the PM till too late, and probably just as well.

    Your small world moment wasn’t a huge surprise, as Lithuania isn’t a large country, and I could easily see that sort of connectivity. I had one back in Phila. when a receptionist at a place I worked asked if I had a relative who worked there. Not as far as I knew… Wandering around on an upper floor one day for reasons I’ve forgotten, I saw a door with a nameplate that said “Systems Science. W. D. McCunney” When I knocked, I was greeted by a chap about my height, build, and coloring. I said “W. D. McCunney?” “Yes.” “Pleased to meet you. I’m D. W. McCunney!” We chatted a bit, and figured out that we had to be cousins from the sundered sides of the family, but never got to trace it farther than that.

  31. >Sometimes a handle isn’t an attempt to hide anything, it’s just what people call a “brand” nowadays.

    Right, that’s what I mean by a display handle.

  32. You may be able to distinguish between obvious pseudonyms like “The Monster” and names that appear to be “real”, but it is impossible for you to know whether any of the latter is in fact the person’s name as shown on government-issued documents without having some external source of such knowledge.

  33. Frank and Morgan, I never used my callsign as a display handle, to use Eric’s term, simply because by the time I would have needed it as such, I had already gotten into the habit of using the Unix account name that I’d been assigned at my first job that needed one.

    I doubt anyone in the hacker community would recognize K5ZC; the number that recognize jmaynard is at least somewhat higher.

  34. But it’s worth noting that my license plates have my callsign on them, and will for as long as I am able to do so. It wouldn’t surprise me to see a car Eric owns have ESR on the plates, if he cares about such.

  35. >it is impossible for you to know whether any of the latter is in fact the person’s name as shown on government-issued documents without having some external source of such knowledge.

    That is true. The net archetype of this is probably the Elizabeth Bimmler hoax on net.suicide.

    On G+ I just coined the common noun “bimmler” for handles of this kind. My reasons for believing these are uncommon are (a) I have only personally discovered one bimmler in near 40 years of being active online, and (b) most people simply are not very good at managing lies. If this sort of thing were really common, awareness of that fact would have seeped into folklore – individual instances like the Bimmler and Shaney hoaxes would not be remarkable.

  36. At a guess, the two most popular open source editors these days are Bram Moolenaar’s Vim and Don Ho’s Notepad++ for Windows,

    I would guess Eclipse instead. Between all the corporate java, web app and android developers that’s a huge number of developers. There were 6M downloads for Indigo in 2011. More for Juno in 2012.

    That’s a large chunk of the estimated 18M devs out there.

    http://www.infoq.com/news/2014/01/IDC-software-developers

    I doubt any other open source editors come close in comparison.

  37. > For blog comments or other social media, I use a pseudonym to separate my political and religious opinions from anything that appears on a resume.

    I’m in this camp, too. I use a “concealing” handle, but it really does need the quotes; I don’t hesitate to give the name printed on my driver’s license to anyone who asks. I’m much more concerned about HR departments or bureaucrats finding my personal business than I am about aquaintances or Internet kooks finding my legal name. The handle I use here is intentionally ungoogle-able (mostly because of the political content of this blog and the fact that people are retarded about politics), but ESR could trivially get my legal name just from the email address I post with, and that doesn’t bother me.

    On the single occasion I made an open-source contribution, I used my legal name; but I wouldn’t hesitate to use a handle if I were contributing to something that was, say, associated with a political or social group that could potentially be disapproved of by an HR droid in a position to spike a job offer. The concealment is in just the opposite direction from that engaged in by a script kiddie. It’s not about anonymity; it’s about keeping personal and professional life safely separated, in a world that makes that quite difficult.

    (The previous is just for comparison’s sake; I don’t claim it as a counterexample because I don’t claim to be a hacker. On a completely different, non-anonymity-related note, I don’t really like equating one’s legal name with a real or True Name. A name is an identifier. If I must have an identifier, then one I choose for myself ought to carry as much or more weight than one I had no part in)

  38. How about people who grew up on BBSes? Using handles was standard, expected. I find it completely normal… people who started their hacking in a more university setting would, of course, be used to using their real name.

    Maybe this is a generational thing. ;)

  39. Willingness to openly stand behind your work is a character and integrity reveal; and ultimately a positive attribute for any social organization. That said, there are a lot of adverse forces pushing people into online concealment as a self-defense strategy. Sadly, this is not a good long term trend.

  40. >Lithuanian.
    And here I thought I was the only one on this blog. Great-grandad came over in 1918 and his surname before they butchered it was, “Andruskevicius”. He got here just before Statlin started his genocide. He assembled bombs, was a violinist and orchestra conductor in Vienna, and designed and engineered conveyeor belts for the Huffy bicycle company.

  41. @nht: Yes, I should have mentioned Eclipse. I have it here, though I seldom use it, but if you develop in Java you probably use Eclipse, and you may well use it even if you don’t. Eclipse pretty much single-handedly killed off the market for commercial third-party IDEs like Borland’s tools: why pay for one when Eclipse had the same capabilities and was free and open source? If you use an IDE, chances are excellent you either use Eclipse or MS Visual Studio. If you develop for Windows, Visual Studio gets the nod, but it’s not an option on OS/X or Linux, and Eclipse is.

    But the additional qualifier will be why you use what you use. Many developers who do it for a living are doing so in a corporate environment, and their employer may well dictate what tools they use to have a common development environment. For that matter, the platform you develop for might do so. There are a lot of development cases where the whole toolchain needed to *do* development uses Eclipse with custom plugins and a version of GCC set up to cross-compile to the target architecture. The extensibility of Eclipse to do that sort of thing is one reason for its dominance.

    If you get to pick what editor you want to use, the equation changes, and the fundamental question becomes “What do you need your editor to do?” For an awful lot of folks, things like Eclipse, Vim, and Emacs *don’t* get the nod because the user simply doesn’t *need* all of what they can do. For instance, I have Vim installed under Windows and Linux, but for the use I make of it, it might as well be the original vi I learned on AT&T Unix System V. I simply don’t need all of the other stuff Vim brings to the table. On Unix, I had Emacs but generally used vi, because again, I didn’t *need* all the stuff Emacs could do, or the overhead that running Emacs brought with it. I needed to hop into the editor, do some edits, and leave, and do so fast. I didn’t need the editor to be my interface to the system.

    On Windows, I normally use an open source product called Notepad2, based on the Scintilla edit control. A registry hack lets it replace Windows Notepad as the default text editor. It doesn’t have a fraction of the power of Vim or Emacs, but it invokes nearly instantly and does everything I actually need to do.

    The barrier to using Emacs at all *is* the size and complexity of the product. To make really effective use of it, you need to customize it, and to be *able* to customize it, you need to learn a fair bit about it. The learning curve will be a substantial barrier to adoption, and many folks simply won’t see a benefit sufficient to justify the effort. It’s no surprise things like Sublime Text are getting popular. They may have 20% of Emacs’ features, but it will be the 20% developers are most likely to use, and they will be a lot more pleasant to deal with while learning to use them.

  42. @dmccunney (And there is aLuau w NeoVim project which hopes to refactor and simplify Vim’s code base and standardize on Lua as the extension language to make it easier foir new folks to become involved in the product’s development.)

    This is quite possibly the stupidest thing I’ve ever heard.

    Who’s going to want to learn Lua to customize their editor?

  43. @Katie: “How about people who grew up on BBSes? Using handles was standard, expected.”

    I didn’t quite grow up on BBSes. But once upon a time, I moderated ten conferences for RIME (Relaynet International Message Exchange), which at the time was the second largest BBS network in the world after Fidonet.

    RIME’s policy was “real names only”, which was a challenge. Most BBS networks assumed a single sort of BBS software, like RBBS or PCBoard. Fido was based on a technical standard, and any BBS that could run a standards-compliant front end mailer could join. RIME had a somewhat different approach, and a BBS on RIME needed to run a specialized program to communicate with the network and pass traffic. Some BBSes, like MajorBBS, didn’t *use* real names. Your ID on a Major system was a handle. There was a real names table on a Major system, as you likely provided one when you joined the BBS, and the RIME client for MajorBBS had to dig the real name corresponding to a handle out of that table and substitute it for the user’s handle when sending message traffic to the network, and perform the reverse operation when importing network traffic to the BBS.

    I used First Initial, Last Name back then, too. And I was “in the book”, and not circumspect about where I lived, so if anyone wanted to track me down in person, they could have done so easily. No one ever did that I was aware of.

    RIME’s policy was in part based on liability concerns on the part of the Sysops. Mine, for instance, tried to audit as much as he could of the message traffic posted to his BBS, because he felt he could be held responsible as the Sysop if someone posted traffic that could be considered criminal or offensive in nature. Another part of the policy had roots in disdain for the sort of behavior seen on some BBSes where users used handles precisely to conceal their real identies, and permit them to behave in ways they wouldn’t do in real life for fear of the consequences.

    Some things don’t really change.

  44. @Morgan
    > This is quite possibly the stupidest thing I’ve ever heard.
    > Who’s going to want to learn Lua to customize their editor?

    Learning Lua takes only a few minutes if you’re already a programmer of any skill, and it’s designed to be quickly and easily embedded in other software, with ittybitty overhead. Lua is rapidly gaining ground in the program-customization realm. Expect to see more and more of this.

    I, for one, am a fan.

  45. >Is Satoshi Nakamoto a “hacker”?

    Yes. And that’s his real name, too – everybody thought it was a concealing pseudo, but it turns out not.

  46. @Morgan Greywolf: “This is quite possibly the stupidest thing I’ve ever heard.”

    If it really is the stupidest thing you’ve ever heard, I consider you lucky.

    “Who’s going to want to learn Lua to customize their editor?”

    As opposed to Lisp, or VimScript, or REXX, or JavaScript, or Visual Basic, or…?

    A fair number of folks, as Lua is *intended* to be an embedded scripting language, and a fair number of editors (many based on the Scintilla edit control) already embed it.

    If you are implementing an editor, and want it to have an embedded script language in which it can be extended, Lua is a decent choice to use for the language, since it already exists, is intended for embedding, has been ported to an assortment of platforms (I have a version that runs on Palm OS), and is widely adopted.

    What would you choose instead?

  47. Yes. And that’s his real name, too – everybody thought it was a concealing pseudo, but it turns out not.

    Or it is. The much-hyped Newsweek “outing” was soundly and plausibly denied by the man they identified.

  48. For another data-point I had a look at the list of Kudo Rank 10 users on Ohloh. In a quick browse through, I only found three names that weren’t obviously a variant of a “normal” name, and two of them I managed to track back to their real names through their home pages and/or project pages pretty quickly. It’s likely that the other one, “nd” from Apache, is also easily tracked back to his real name as well, but I couldn’t find it in thirty seconds of searching, so I moved on.

    There was a Chinese name which I assume is also a real name, and not a handle, but even if I include that we’re limited to 2/64, or about 3%.

  49. > And here I thought I was the only one on this blog. Great-grandad came over in 1918 and his surname before they butchered it was, “Andruskevicius”.

    Cool. Sounds like a pretty impressive and accomplished guy. I know some of the projects my grands on both sides worked on and damn if that doesn’t make me go “wish I’d worked on a piece of that” – though mostly their lives were pretty quiet and unassuming. Sadly, I’ve run into a fair number of Lithuanians my sons age who have no idea what our parents and grands ran away from because yes, their lives depended on it.

    fwiw – I’m at gmail.

  50. I don’t know if you can safely count Eclipse users as editor users.

    I use Eclipse, so I’m one of the statistic. It’s part of the corporate standard software toolchain (or perhaps the toolchain is part of Eclipse?) where I work. I use it about once a year, usually to run some program analysis tool that exists only as an Eclipse plugin. This seems to be typical usage. Outside of official corporate training, I’ve seen only a handful of developers use Eclipse to edit code. That includes employees and contractors of partners and customers too. People I’ve met will use anything–proprietary, open source, shareware, beta software, alpha software, pre-alpha software, free software, bundled-with-the-OS-ware…in some cases even literally write their own editor–instead of using Eclipse to edit text files.

    A few employers ago I did meet one guy who used Eclipse as an editor in preference to readily available alternatives. We could tell when he touched a file because Eclipse would silently screw up the character encoding of some string constants in our project. I’m sure there’s a way to fix that, but I couldn’t care less what it is.

    A few more employers and a different IDE and toolchain ago, a GUI designer asked the developers what helpful features they could use in the IDE. I think they were expecting lists of checkbuttons to put on dialog boxes. The response mostly consisted of ways to make the IDE look and feel more like a good Unix command line, but with better autocompletion. The GUI designers stopped talking to us after that…I’m not sure who they thought their target market was, but it was clearly not us.

  51. Speaking as someone who has gone by both pseudonyms and “WesleyAC” (Because I don’t like to type out my 20-character last name), I see the benefit in both, just in different contexts. I would guess that the reason that you see more people in the FOSS community going by their real names than “hackers”/security researchers is because people who do “hacking” (in the *incorrect* use of the word) can have it held against them. I have found some interesting exploits and such that I’m sure are unique, but not published them under my real name because people have preconceived notions of what a “hacker” is, and I don’t want people (especially employers) to think that I’m untrustworthy because I have an interest in security. On the other hand, if I mention contributing to an open source project, people think that that’s a good thing.

    My point is that “various ignorant media stereotypes” are often just portraying a different community, thus from the perspective of the media, they are correct, only because they don’t know the proper word to use.

  52. Morgan, I would expect more people to learn Lua to customize an editor than I would expect to learn LISP to customize an editor. Lua has uses outside that one program; to many people, the only purpose for LISP is to customize EMACS.

  53. >Morgan, I would expect more people to learn Lua to customize an editor than I would expect to learn LISP to customize an editor. Lua has uses outside that one program; to many people, the only purpose for LISP is to customize EMACS.

    and, returning IT, is it (still) true that most of EMACS is written in elisp itself?

  54. >and, returning IT, is it (still) true that most of EMACS is written in elisp itself?

    whoops, IT for the previous thread, sorry

  55. @ESR

    >The key point, it seems to me, is that you remain accountable under your own name for the code that you write and its consequences.

    This sounds as if contributing code to OS projects could be a potential liability. In fact, linking your code to your resume _even when it is mediocre_ is a huge bonus in getting a job (or a consulting gig), because it immediately sets you apart from the large number of job-only programmers who never do any work for fun or any other reason, just as a 9-17 job (and usually they cannot share that).

    About 1% of the people I know who work in ERP as developers or consultants has anything public – a tips and trick blog, some utilities developed and shared, or helping people on forums, and so on. Seeing something like this in a resume as opposed to just a listing of employers and responsibilities immediately sets it apart. Even when the quality is raher “meh” it shows a good kind of attitude.

  56. That’s a frequency of just about exactly 1% for IDs that could plausibly be described as concealing handles in commit logs. That’s pretty low, and a robust difference from the cracker underground in which 99% use concealing handles.

    I see this as simply the difference between legal and illegal work.

    Few people are willing to acknowledge illegal work under their legal name. Most people are willing to acknowledge legal work under their legal name. Exceptions like (legal) commercial SX workers (fill in moderation hook) prove the point: Coding in general is honorable.

    Nevertheless, this distinction might be a good one to present to clueless outsiders among politicians and journalists.

  57. >This sounds as if contributing code to OS projects could be a potential liability.

    So far it has not been a legal liability in any case I know of. Hackers don’t worry about this much, except in some cases for concerns about being sued over some junk patent.

    More relevantly, by shipping under your own name you accept the reputational risk if you screw up. Your willingness to do this is considered, as another commenter pointed out, an honest signal of fitness.

  58. @Federico: “and, returning IT, is it (still) true that most of EMACS is written in elisp itself?”

    Yes. Gnu Emacs is essentially a Lisp interpreter, and most of the editor is written in the dialect of Lisp it implements.

  59. >And it’s amazing how so many people consider my ham radio call sign, issued by the FCC and readily linked to my real name on the FCC database, a “dodgy” hacker handle…

    I have an FCC-issued call sign. No I’m not a ham, though I have been studying for the Technician exam in my spare time- I actually bought a GMRS license. I think I’m about half the people who’ve actually done that.

    http://wireless2.fcc.gov/UlsApp/UlsSearch/searchLicense.jsp (try it, it’s scary)

    That database is open, easily online searchable, and gives away WAY too much and too personal information. The FCC is not your friend in that regard. I’d be careful who you leak your callsign to- the ham community may be small and relatively tight and trustworthy, but the world is also full of identity thieves and meth heads, and hams tend to have portable valuables….

  60. @esr: “More relevantly, by shipping under your own name you accept the reputational risk if you screw up. Your willingness to do this is considered, as another commenter pointed out, an honest signal of fitness.”

    Agreed, but the concept of “your own name” is slippery. If you habitually contribute code under a display name or alias that is not your legal name on government documents, you can still be greatly concerned about your reputation, as that reputation exists in the context in which you contribute code, and attaches to your alias. Arguably, “your own name” is the name by which everyone knows you, and that may not be the one on your birth certificate.

  61. @zygo: “I don’t know if you can safely count Eclipse users as editor users.”

    It’s a good question. Where do you draw the line between an editor and an IDE? Emacs is thought of as an editor, but has IDE features and can reasonably be called an IDE.

    Eclipse includes an editor among many other tools, but IIRC, if you really want to you can replace it with something else like Vim. I have no idea how many people do.

  62. > More relevantly, by shipping under your own name you accept the reputational risk if you screw up. Your willingness to do this is considered, as another commenter pointed out, an honest signal of fitness.

    If you have a reputation, you have reputational risk. Even if you can abandon your name and start over, that only gets you to zero.

  63. Eclipse pretty much single-handedly killed off the market for commercial third-party IDEs like Borland’s tools: why pay for one when Eclipse had the same capabilities and was free and open source?

    Borland killed itself really. Sure Eclipse helped a bit in the IDE department but folks (maybe not linux folks) will pay for good tools if they are indeed good. On Windows Visual Studio is top notch. They abandoned Kylix despite the fact it was essentially just Delphi running inside Wine.

    Eclipse became dominant because it had decent refactoring tools early on. Whenever folks talked about Netbeans vs Eclipse most of the time IME refactoring was the deciding factor.

    If you use an IDE, chances are excellent you either use Eclipse or MS Visual Studio. If you develop for Windows, Visual Studio gets the nod, but it’s not an option on OS/X or Linux, and Eclipse is.

    For OSX XCode is dominant for obvious reasons.

    If you get to pick what editor you want to use, the equation changes, and the fundamental question becomes “What do you need your editor to do?” For an awful lot of folks, things like Eclipse, Vim, and Emacs *don’t* get the nod because the user simply doesn’t *need* all of what they can do.

    For larger projects an IDE greatly improves productivity. What I need my editor to do is write code. The most efficient in terms of coding support is an editor integrated into an IDE that doesn’t suck.

    I’m not sure that Eclipse qualifies for “doesn’t suck” but it’s better than the other free alternatives…at least for Java. For C/C++ I’d move to QT Creator and not look back. CDT = yuck. One of these days I will try MonoDevelop for kicks.

    Emacs is what I used 25 years ago. I suppose some new folks still use it but I’d guess most folks that do could meet AARP’s minimum age requirement.

    As far as “hackers” and anonymity goes, a lot of the payout for open source is egoboo. So yes, it strikes me as obvious that most folks used their own names. For fresh outs it’s usually the only body of work in their portfolio that isn’t coursework. For older hands too since a lot of stuff we do for work can’t use used in your next interview.

    But that’s not the interesting question (hacker vs cracker). The interesting question is whether the hacker sub-culture is really relevant anymore. I would assert it has been overtaken by the Maker subculture and its ethos. There’s a lot of overlap of course but the focus has shifted a bit. More of an inclusive CC than exclusive FOSS mindset. Maybe that’s just my biases talking.

  64. May I suggest looking a the contributors for WINE as a prime candidate to test the “more obvious pseudonyms among gamers” logic? I haven’t really examined it (just a cursory scan) but that first glance didn’t see any handles, just legal names.

  65. ESR: You’re not saying Satoshi Nakamoto’s identity is Dorian S. Nakamoto, are you? There’s convincing evidence otherwise, and very little supporting evidence. Among the evidence that Dorian Nakamoto, outed as Satoshi Nakamoto by NewsWeek, is his public denial (link goes to TechCrunch coverage of said denial.

  66. Some follow up digging around led me to this: the original user posting about BitCoin denied being Dorian Nakamoto. The original post, made in 2009, has no reason not to be Nakamoto’s, as there was no notoriety to bitcoin yet. And the comment was made by the same user that was discussing BTC back in 2009. So, denial on both ends.

  67. >Arguably, “your own name” is the name by which everyone knows you, and that may not be the one on your birth certificate.

    Agreed. But so what? If (a) that’s what you put in commit logs, and (b) it’s a stable pointer back to you (through, say, a web search) the fact that it isn’t your legal name is not relevant to the discussion. The essential property is that it’s a stable identity for purposes of reputation and trust, one which is public and difficult to repudiate.

  68. >Among the evidence that Dorian Nakamoto, outed as Satoshi Nakamoto by NewsWeek, is his public denial

    Thanks, I didn’t know of this.

    Now that I do…there are honorable reasons he could be lying.

  69. “The interesting question is whether the hacker sub-culture is really relevant anymore. I would assert it has been overtaken by the Maker subculture and its ethos.”

    ‘Hacker’ always has covered a lot of things, and what esr considers to be the definition of the word is only one small part. I am definately not a hacker as he would define it, but I certainly consider myself one, as I so considered all the guys I used to meet down at the Trenton Computer Fest while looking for stuff to enhance my homebrew microcomputer.

  70. >I so considered all the guys I used to meet down at the Trenton Computer Fest while looking for stuff to enhance my homebrew microcomputer.

    Huh? It is pretty likely I would have considered those guys hackers too. (I used to go to TCF, so I know exactly the sort of people you’re speaking of.)

  71. > Now that I do…there are honorable reasons he could be lying.

    It doesn’t matter if he’s lying or not – if he’s telling the truth, it means the real Satoshi is still out there, and may not be his real name. If he’s lying, then honorable or not it’s equivalent to using a pseudonym in that it means choosing not to connect one’s contributions to one’s real-world identity.

  72. >It doesn’t matter if he’s lying or not – if he’s telling the truth, it means the real Satoshi is still out there, and may not be his real name. If he’s lying, then honorable or not it’s equivalent to using a pseudonym in that it means choosing not to connect one’s contributions to one’s real-world identity.

    This is true, but a reasonable person could consider it to fall under the generally-recognized exception for well-founded fears of persecution.

  73. Use of handles vs real names among committer names depends IMHO on the version control system used. Centralized VCS, such as CVS and Subversion required account on server to commit, so it used login name… which might or might not be connected to real name, depending on account creation policy, login names being available etc. Distributed VCS use user-configurable name; Git for example asks for email (user name + user email). On one hand side best practices lead to real names, on the other hand you are not beholden to any “initial + full surname” rules of creating accounts.

    BTW. I have encountered strange issue wrt. real names. When working on some “open source” scientific software (you have to ask via web form / email to get even read access to repository), I was using git-svn as my client. I wanted to have git-like names in git-svn, so I have started to create authors file. At beginning it was easy, as all contributors also wrote academic papers referenced on sofware homepage, and was quite easy to map logins to names. Later there were some new committers, so I asked for real names… and got answer that it is impossible because of Europe privacy regulations ;-(

  74. >The interesting question is whether the hacker sub-culture is really relevant anymore. I

    If you got my mail you wouldn’t wonder about this. I still hear from lots of people who want to join us.

    I think we can actually get a pretty good estimate of size and relevance from the committer lists of open-source projects, and contributor statistics on forges and places like Ohloh. These indicate to me a hacker population worldwide that is well into multiple millions and rising. And, just on my own projects, if the hacker culture were fading I’d expect the fequency of names unknown to me on incoming patches to be stable or falling. It isn’t.

    More can be said. When I look at those new names, it seems to me the population growth is tracking availability of Internet access pretty closely. I think it could be modeled fairly accurately on the assumption that populations have a small fixed frequency of potential hacker types who join the culture pretty reliably once they have reliable Internet.

  75. >Centralized VCS, such as CVS and Subversion required account on server to commit, so it used login name…

    On those you look at things like the Subversion committer list, which does list committer names as well as IDs.

  76. @esr:

    This is true, but a reasonable person could consider it to fall under the generally-recognized exception for well-founded fears of persecution.

    No kidding. If there’s one person with coding skills in the United States who has well-founded fears of persecution, it’s guy who invented Bitcoin. Does anyone not remember what happened to Phil Zimmerman in 1996, who merely wrote a public-key encryption program?

  77. @esr:

    Arguably, “your own name” is the name by which everyone knows you, and that may not be the one on your birth certificate.

    Agreed. But so what? If (a) that’s what you put in commit logs, and (b) it’s a stable pointer back to you (through, say, a web search) the fact that it isn’t your legal name is not relevant to the discussion. The essential property is that it’s a stable identity for purposes of reputation and trust, one which is public and difficult to repudiate.

    So far you have specified Concealing Handles, Display Handles and Real Names; where do you feel that a persistent identifier that is unconnected to a legal (or physical) identity falls in this continuum?

    Also how exactly would you define a Concealing Handle?

  78. >So far you have specified Concealing Handles, Display Handles and Real Names; where do you feel that a persistent identifier that is unconnected to a legal (or physical) identity falls in this continuum?

    Well, if you’re thinking of ‘dak180′, it is relevant that your real name appears associated with it in commit logs I have access to. So it is connected to a legal/physical entity after all.

    >Also how exactly would you define a Concealing Handle?

    One that is designed to be easily repudiated so that the user can avoid the legal or reputational consequences of behaviors performed under that identity.

  79. @esr

    Reconstructive archaeology… I like that term :)

    Your work on Emacs changelogs brings up a few more questions too though. Could it be that contributors to Emacs were more inclined to use their real name due to Emacs target audience? Were Emacs contributors encouraged to provide their real name with submitting patches?

    I did something similar about 4-5 years ago with Eggdrop’s changelog files. It had no detailed version history, no ‘AUTHORS’ file, and prior developer groups had tended to delete old the changelog files (yes, really!) from the source tree when they made a new major release.

    Ultimately, what I ended up doing was taking the changelog files from every single old release tarball I could find (or had in my personal archives) and built the most complete set of changelog files I could. [There is a gap due to missing versions between 0.7d and 0.9 (~Apr 11, 1994 to Jul 23, 1994) but maybe those will eventually turn up somewhere.] I had to do a lot of this by hand (with scripts and macros to assist where possible) since the changelogs had been maintained by hand and were in different formats. There were also errors I had to correct (old email list archives helped a lot), people who had changed handles and/or the person adding the entry to the changelog had made a typo, etc. Unfortunately, a lot of the old email list archives (eggdrop @ sodre.net, etc) had long since been lost, too. In the end though, I had a set of changelog files I could parse with other tools I’d written that could auto-generate the ‘AUTHORS’ and ‘THANKS’ files.

    While doing this work, I found the majority of Eggdrop’s contributors did not contribute under their real name. I was able to find some people’s names with Google by searching for their handle and/or email address, but not everyone. I also never completed the task due to a huge mess elsewhere and the implosion of Eggdrop’s development community after someone basically took it over and locked others (including myself) out. The guy who locked me out thought I was wasting my time in compiling this stuff too, and thought I should spend my time writing code instead. Go figure.

    Here are the results of the changelog stuff I did with Eggdrop:

    THANKS
    AUTHORS
    Versions
    doc/ [see the individual 'Changes*' files]

  80. >While doing this work, I found the majority of Eggdrop’s contributors did not contribute under their real name.

    Looking at the file, it seems pretty clear that the reason for this is like Jay’s case in Second Life – they may have contributed under a stable identity more important to them in this context, namely an IRC handle. But from what you tell me we don’t actually know they did that, because we don’t have the actual commit logs. What we have is one remove from that and may have been (in fact, probably was) abstracted through someone’s knowledge of how committer identities corresponded to IRC handles.

    Whether an IRC handle qualifies as a concealing handle depends on policy details at individual networks. Broadly speaking, freenode is designed to support persistent identies and non-repudiation; other networks, not so much.

    Further note: because your list doesn’t have a real-name field, I can’t use it to gather the statistics I want.

  81. >Were Emacs contributors encouraged to provide their real name with submitting patches?

    Not particularly. But under bzr your name is going to show up in commit logs, and most people configure bzr with something that is recognizably a real name as a matter of course.

  82. because your list doesn’t have a real-name field, I can’t use it to gather the statistics I want.

    The best way you could obtain that would be the individual Changes* files using the ‘Found by:’ and ‘Patch by:’ lines. One thing to note though, for reliable statistics, you can’t simply use a space to determine if a name is a handle or a real name, but for simplicity (and my sanity) I wrote my tools to do just that. Some contributors also only gave their first name, or even an initial and last name.

    The tools I wrote to build the THANKS and AUTHORS files output the text in 3 columns as “handle name email” so the columns will line up. I have a datafile that those tools use to map other information not in the Changes* files (handle/name/email) to build the full output. That file has the format (tab delimited):

    [handle or name] email foo@example.com
    [handle or name] handle Handle
    [handle or name] name Name

    If you’d like, I can email you some of the files I used.

    There were no commit logs for Eggdrop either, since Robey didn’t use a revision control system when he was created and maintained it. The CVS tree the next development group used was deleted when they abandoned Eggdrop, and several more CVS histories were lost when Eggdrop’s was moved a few more times to different development machines (I was -not- happy about this once I discovered what had happened).

  83. >The best way you could obtain that would be the individual Changes* files using the ‘Found by:’ and ‘Patch by:’ lines.

    Seems like a lot of trouble to go to for a project with a history that is known to be corrupted and unreliable. But I’ll do it if I don’t get a large enough sample of lists.

  84. > I use first initial, last name as an online handle too, with little fear of confusion

    An interesting tale. Heh.

    I get why a bunch of Lithuanian officials would know my dad. It’s not a big place at all these days (though in its heyday….)

    Running into them in Philly was a bit unexpected though.

  85. >I have an FCC-issued call sign … I actually bought a GMRS license.

    I raise you: one (each) FCC Citizens Band license, from their brief foray into licensing CB radios in the mid-1970s. A brief look at the FCC web site doesn’t mention anything about it, and searching for my call sign yields nothing there or on popular search engines, but the officia paperl document is still in a file folder somewhere…

  86. @LS:
    >‘Hacker’ always has covered a lot of things, and what esr considers to be the definition of the word is only one small part.

    What amuses me greatly are the people who think that doing what esr’s kind of hacker would call “baggy pantsing” to somebody on Facebook qualifies as “hacking” even in the modern pop-culture sense of the word.

  87. TRX, the FCC actually had licensed CB radios since they were introduced back in the late 1950s. (The band used to be allocated to ham radio.) It’s just that few paid attention during the CB craze.

    I got a CB license so I wouldn’t jeopardize my newly-obtained Amateur Extra license by operating a CB without one. That’s long since expired, of course.

    For several years, the FCC didn’t do away with CB licensing, per se. (They were still required to license anything that’s not an incidental radiator or of very low power.) Everyone in the US had a CB license, with a callsign of K, their first and last initial, and their ZIP code. I thought this was still the case, but I just looked at the CB rules (47 CFR 95.401 et seq.), and that’s all been done away with. I guess the FCC got Congressional approval to do away with individual licensing in the CB service.

  88. What I find interesting about the comments is that as far as I can tell, people in the Western world are more afraid of HR departments than they are of their government. That’s scary.

    Is this worth changing, and how would one go about doing that?

  89. how would one go about doing that?

    The NSA and CIA are hard at work on the problem.

  90. One other area where people who might be generally considered to be hackers tend to use pseudonyms is in the security/malware research space. There’s a good reason for that because criminals and nation state actors get upset when their plans are thwarted. Thus public announcements etc. often come from pseudonyms. Having said that in a number of cases inside the kimono where the actual research is discussed (and where leaks are unlikely) pseudonyms are deprecated because we need to be able to trust the others in the group.

    I’d be curious if you get name collisions in hacker space. I found a recent example somewhere else where two people with identical first and last names were due to speak at a conference on different but related topics. There was considerable confusion that might have been avoided if the two were known by their initials, handles or similar.

  91. @Garrett:

    > Is this worth changing, and how would one go about doing that?

    I disagree with the premise. Anybody who isn’t deeply scared of the government’s powers is not paying attention, and anybody who is scared of HR obviously has employability issues.

    But for the sake of argument, I’ll agree and play along: Why on earth would you want to give the government more power? Conversely, why would you make it difficult for companies to fire people?

    @Christopher Smith:

    > The NSA and CIA are hard at work on the problem.

    Exactly.

  92. I think we can actually get a pretty good estimate of size and relevance from the committer lists of open-source projects, and contributor statistics on forges and places like Ohloh. These indicate to me a hacker population worldwide that is well into multiple millions and rising.

    I disagree that this is evidence of a growing hacker population. I have thousands of commits on Ohloh and a good chunk of that was on the clock as a corporate developer. Even if I self identified as a hacker (and you accepted that claim) that wouldn’t be while wearing my hacker hat as much as my pro developer hat.

    My belief is that a large percentage of FOSS commits are made by corporate devs. Some will self identify as part of the hacker sub-culture. Many would not.

    While I meet the jargon file criteria for hacker I’m not really…in the same way that I’m not really fen. Or more accurately for me FIJAGDH as opposed to FIAWOL. In a room of fen I’m the one that looks (and sometimes feels) like a ‘dane. Lots of folks avid read and watch SF. Far fewer go to cons. Even fewer believe FIAWOL.

    And you can replace Fandom with FOSS in FIJAGDH.

    A lot of OSS commiters I’d put in that category…not a “hacker” in the sub-cultural sense.

    And, just on my own projects, if the hacker culture were fading I’d expect the fequency of names unknown to me on incoming patches to be stable or falling. It isn’t.

    Well, that’s kinda like RMS observing that he sees no shortage of new FSF converts right? And yet many projects on GitHub didn’t have any sort of license a year or so ago and of those that do MIT and BSD outnumber the GPL ones.

    http://www.theregister.co.uk/2013/04/18/github_licensing_study/

    I think the general dev community is different now from the hacker dev community one whereas in the 80s I would claim that the hackers were in the mainstream. Maybe because I was a unix weenie back then.

    That ESR doesn’t see a reduction of hackers on his projects isn’t an indication that the pool isn’t getting shallower. You’re swimming in the deepest part of the pool and you’re going to attract what hackers exist in greater number than average anyway.

  93. @Nigel
    “That ESR doesn’t see a reduction of hackers on his projects isn’t an indication that the pool isn’t getting shallower.”

    I want to step aside from all attempts to redefine words and suggestions that anyone who gets paid to contribute to OSS should be discounted as a “pure” hacker. It is rather odd to claim that many people go into programming purely for financial reasons and not for at least a minimal love of (computer) technology.

    There is no reason to even suggest the number of people qualifying for the title “hacker” is declining.

    Until 2007/2008 the growth of OSS was exponential in every metric:
    http://www.riehle.org/wp-content/uploads/2008/03/oss-2008-total-growth-final-web.pdf

    The size of Debian grew by almost 50% between 2007 and 2012:
    http://blog.james.rcpt.to/2012/02/13/debian-wheezy-us19-billion-your-price-free/

    The number of people contributing to the Linux kernel is still growing in 2013:
    http://www.linuxfoundation.org/news-media/announcements/2013/09/linux-foundation-releases-annual-linux-development-report

  94. >My belief is that a large percentage of FOSS commits are made by corporate devs.

    I can think of some important projects for which this is probably true. The Linux kernel might be like that, or Apache. But there are thousands upon thousands of projects of open-source packages that thrive without the kinds of incentives that draw in those people.

    Games are the clearest category where nobody is going to show up wearing a corporate-dev hat, and projects like Battle for Wesnoth are in rude good health with growing developer populations.

  95. There is no reason to even suggest the number of people qualifying for the title “hacker” is declining.

    I qualify but I don’t self identify as hacker. I love coding and I didn’t start just because it was lucrative…although I’ve met some who did.

    What does that mean then? At what point do you stop being a hacker and become merely a good coder? I say it’s not just the skills and ability to do a good hack any more than it’s just the love of SF to be a fan. Some are fen and part of that sub-culture while others, despite the same affection for the same material, aren’t and just fans.

    And looking back at some prior threads regarding shareware devs in the 80s I don’t think that ESR classifies many of those devs as hackers even if they meet the criteria in the jargon file. Maybe that’s an incorrect impression on my part but it seemed to me that to be a considered a “real” hacker you typically were a systems dev (a unix one at that) and not an app dev.

    If all it takes to be a hacker is the ability to sling some mean code I think that the label has low value as an identification of a sub-culture.

    And I don’t think that contribution to an OSS project automatically qualifies you either. That’s just a thing you did, not necessarily what you are.

    I’m officially an OSI member (paid my dues and everything) but not really part of the FOSS crowd because my beliefs are different than the norm for that sub-culture. Too much dominated by the FSF segment. I’d much rather be grouped with the CC folks and maker folks that are less politically strident about open and more about cool builds than how evil IP is and how Microsoft and Apple is bad.

  96. @esr looking at the ohloh stats for wesnoth the activity seem to have peaked around 2010. Could be it’s just mature now.

    How much did the apple App Store/GPL controversy impact contributors? Seemed pretty minimal from the outside looking in but that was around the same time frame.

  97. >How much did the apple App Store/GPL controversy impact contributors? Seemed pretty minimal from the outside looking in but that was around the same time frame.

    That’s right. I don’t think it had any lasting effect.

  98. >And looking back at some prior threads regarding shareware devs in the 80s I don’t think that ESR classifies many of those devs as hackers even if they meet the criteria in the jargon file. Maybe that’s an incorrect impression on my part but it seemed to me that to be a considered a “real” hacker you typically were a systems dev (a unix one at that) and not an app dev.

    No, that would be far too narrow. What excludes most of the shareware devs isn’t that they were writing apps and games rather than system stuff – today you can earn hacker cred on projects like Wesnoth, after all. It’s that they didn’t have the ethos that later developed into “open source”, of taking pains to make your code available for re-use and improvement by others.

  99. >And I don’t think that contribution to an OSS project automatically qualifies you either. That’s just a thing you did, not necessarily what you are.

    One open-source contribution doth not a hacker make. But show me someone who exhibits a preference for that over writing closed source, and gets why a picture of a Volkswagen Bug with a license plate reading “FEATURE” is funny, and knows who RMS and Linus and (er…) ESR are, and I’ll show you a hacker.

    By your tenth open-source project or so, denying that you’re a hacker is about as futile as collecting Star Wars figurines and denying that you’re a geek…

  100. ESR, what do you think of the landscape around modding Minecraft?

    The game itself is closed-source, but written in java and has been thoroughly decompiled [with the company that makes it largely turning a blind eye to it, as long as there's no interference with their revenue collection mechanism, which there's generally a cultural aversion to anyway].

    Some of the mods are open-source [though the extent to which this is encumbered by being a derivative work of the game is unknown due to the company's lax enforcement], including a couple of major frameworks that are designed to make it easier to build other mods and have them work together, but I don’t think this is universal.

    This seems to fit the kind of situation described in your earlier post about “sharecropping” to a tee, but while you said it’s not ‘real’ open-source, you didn’t state whether you think the people who build such things qualify as hackers.

  101. @Patrick Maupin:

    “But for the sake of argument, I’ll agree and play along: Why on earth would you want to give the government more power? Conversely, why would you make it difficult for companies to fire people?”

    I don’t, and that’s not my goal. My observation was cultural, not legal. Sociologists complain about Henry Ford going around and inspecting the homes and lives of his workers, insisting on certain personal behavior. Yet doing the same thing today based on political commentary or participation is somehow appropriate. That’s where I see the problem occurring.

  102. By your tenth open-source project or so, denying that you’re a hacker is about as futile as collecting Star Wars figurines and denying that you’re a geek…

    What? Those are unique collectible items that appreciate in value. Do you know what a pristine 1978 Darth with a telescoping saber is worth? It’s like those plates that people collect and hang on their walls. Nothing geeky about that at all. Nope.

  103. >This seems to fit the kind of situation described in your earlier post about “sharecropping” to a tee, but while you said it’s not ‘real’ open-source, you didn’t state whether you think the people who build such things qualify

    As hackers, you mean? Interesting borderline case.

    In the historical examples, the answer is often “yes”, but I wouldn’t want to say it always is without looking at individual cases. Certainly some of the early sharecropper communities left their DNA in today’s hacker culture. Of these, the one around the DEC PDP-10 is the most obvious example.

    The Minecraft community is a tougher call because it reflects a choice to stay closed-source that was made well after “free software” (and later open source) became a central, organizing idea in the hacker culture. I’m certain that the concept of a non-open-source hacker subculture made sense as late as 1985 (GNU Manifesto), but whether that was still true starts to get iffy after about 1992 (early free Unixes) and highly questionable after the gathering of the hacker chieftains that Tim O’Reilly arranged in 1998.

    I’m going to say I don’t know yet. Dang, you’ve handed me another research project – I may have to immerse myself in the Minecraft culture just to answer this question. I think it might turn on whether they laugh at the same jokes we do. No, I’m not kidding….

  104. There are nine original production cels from Animaniacs, as well as a variety of other collectibles associated with the series, around my house. Yes, I’m an Animaniacs geek.

  105. @ESR, Not that I am arguing your premise, but here you are potentially confusing causation versus correlation. For example, you could make a case that a more restrained NSA that was prevalent for the majority of emacs life-cycles led to less concern over non-obfuscatory naming amongst open source contributors.

    A more interesting methodology would be frequency analysis of contributor naming controlling for age, as I would suspect older, more established folks are less likely to use obfuscatory handles in more recent builds. Even better would be to figure out corporate sponsorship.

  106. > I’m going to say I don’t know yet. Dang, you’ve handed me another research project – I may have to immerse myself in the Minecraft culture just to answer this question. I think it might turn on whether they laugh at the same jokes we do. No, I’m not kidding….

    If you do, you might want to make a side-trip to the people who design and build stuff with Redstone (an in-game, maybe theoretically turing-complete, automation system). Some of their stuff is crazy.

  107. @Garrett:

    Sociologists complain

    Ah, there’s your problem right there — giving a rat’s ass about what sociologists think.

  108. I disagree with the premise. Anybody who isn’t deeply scared of the government’s powers is not paying attention, and anybody who is scared of HR obviously has employability issues.

    An interesting discussion given Brenden Eich’s recent ouster. Blacklisting is an HR sorta thing and not a governmental sort of thing.

  109. @Nigel:

    Please tell me you’re not seriously comparing a company’s ability to terminate its relationship with an employee to what a government can do to a person.

    Never mind — for the sake of argument, I’ll play along with you, too.

    Are you arguing that we somehow need to take away the ability of the public to boycott a company, or are you suggesting that any company which is the subject of such a boycott should always be legally required to suck it up and keep the offender employed?

    Because if you’re not arguing either of those, then you’re allowing for the possibility of at-will employment, which allows for the possibility that the employee can fuck it up. In this case, the current possibilities would seem to be that (a) Brendan Eich has plenty of other stuff he can go and do (so shouldn’t be scared of losing his employer) or (b) he has employability issues.

    But if it’s the latter, he could probably fix those issues by moving away from California to go work as an IT guy for a Southern Baptist church somewhere.

  110. @patrick

    I’m saying that I believe that the Hollywood blacklist was as wrong as prop 8.

    How did people lose their livelihoods? Via legislation to prohibit communists from holding jobs? Or through boycotts and intimidation by special interest groups (the American Legion in this case)?

    I think that communism is stupid. I think supporting communism is stupid. I even think at some point that supporting communism could result in treasonous behaviors. I don’t think that many of the folks blacklisted as commies by Hollywood actually were commies or actually did any harm to the US.

    If prop 8 was introduced in my state I’d have opposed it. But that wouldn’t preclude me from working with folks on the other side on unrelated issues. I don’t like litmus tests with such low thresholds that simply taking part of the democratic process results in blacklisting.

    I don’t like mob rule…and yes, I fear mobs more than I fear the US government. Historically they have been more dangerous.

    I dunno…for some of you guys everything seems so black and white. The government is bad. Free market is good. Legislation bad. Boycotting good. And yet there are clear examples in history where no, not really.

    I feel that demonizing people that disagree with you is usually wrong. I haven’t ever met the guy but I think some of the things said to vilify him are factually incorrect and very insidious.

    And your response that he could just move out of California to work IT for a Baptist church is telling in its callousness. That’s the same thing that many of the Hollywood blacklist folks had to do. Move to Europe to make a living at their profession.

  111. @Nigel:

    > How did people lose their livelihoods? Via legislation to prohibit communists from holding jobs?

    If you are trying to insinuate that Congress wasn’t involved at the highest level of this, you should go study history.

    > I don’t like mob rule…and yes, I fear mobs more than I fear the US government.

    This is stupid. The mob _was_ the government.

    > I dunno…for some of you guys everything seems so black and white.

    Some things are, including expected consequences of actions.

    > And your response that he could just move out of California to work IT for a Baptist church is telling in its callousness.

    yeah, and when I tell my daughters not to walk around bad neighborhoods at night wearing skimpy clothes, I’m being similarly callous.

    Seriously, Brendan Eich is theoretically a grown man who is theoretically relatively intelligent. If he wants to be a reactionary activist, he should be capable of figuring out some of the potential outcomes, especially after studying the very history you are describing.

  112. > I don’t think that many of the folks blacklisted as commies by Hollywood actually were commies or actually did any harm to the US.

    Whether they harmed the U.S. is debatable. That most were Communists is, unfortunately, fact.

    With a few minutes of web-searching I was able to verify that seven of the Hollywood Ten were members of the CPUSA. An eighth was plausibly accused of pressuring screenwriters into inserting Communist propaganda into their work.

    The Wikipedia entry on the “Hollywood Ten” fails to mention this inconvenient truth. Which is entirely typical of the way the blacklist has been misrepresented and and Soviet subversion in its era whitewashed.

  113. The Wikipedia entry on the “Hollywood Ten” fails to mention this inconvenient truth. Which is entirely typical of the way the blacklist has been misrepresented and and Soviet subversion in its era whitewashed.

    I’d be interested in some reverse engineering in this area. Specifically, communists have long expressed willingness to play the long con, slowly subverting social structures to move gradually toward totalitarian states; viz. the American shift in discourse you’ve noted before where communist duckspeak is everyday. (I’m not quite yet willing to believe the claim that the political piece of the collapse of the Soviet Union was a conscious ploy to institute a market economy so corrupt that not only could communist propaganda more easily attach to it but even that many of the communist failings could retroactively blamed on it in preparation for a reinstitution of formal communism, but it’s looking decreasingly far-fetched.) Similarly, even the institution of Russian communism itself was a deliberate memetic infection by the Kaiser to destabilize the eastern front.

    But why is the human brain so susceptible to that memetic infection, especially since communist memes have a number of rather basic contradictions are are openly built on absurd premises (such as the idea that all people are cooperative, generous, and inherently hardworking absent incentives)? Why does that long con work at all, and why do the people who practice it dedicate so much of their lives to harming society? How many of them are deluded true believers, and how many are just plusdoublethinkers?

  114. With a few minutes of web-searching I was able to verify that seven of the Hollywood Ten were members of the CPUSA. An eighth was plausibly accused of pressuring screenwriters into inserting Communist propaganda into their work.

    The Hollywood blacklist extended FAR beyond the ten. If it were just the ten it would have been one thing.

  115. @patrick

    You can easily read about the history by going no further than wikipedia (not the best source but an easily accessed one). The congress subpoenaed ten. Some went to jail for contempt.

    The private special interest groups produced large lists of people that were suspected of being communists. The mob targeted these folks with boycotts and letter campaigns based on those lists. The studios caved and caved quickly. Those folks got fired even it was a case of mistaken identity. This is all documented.

  116. @Nigel:

    I don’t think you are giving enough credit to the spectre of jail time, and I don’t think you’ve followed the money. Disney wasn’t following along — he was right out in front, because it was advantageous to call anybody who dared to strike a communist.

  117. Eric screws up the logic on anonymity.

    It is not personal identity that is important, but rather reputation. Reputation can be separate from personal identity so that we can’t be enslaved by totalitarianism, debt rating agencies, etc.. We can be reborn and creative at will. He is trying to say that reputation is a corner case. No it is the case.

  118. >He is trying to say that reputation is a corner case.

    That’s an astoundingly dim misreading. Stable non-repudiable identities are important precisely because they carry more complete information about reputation and your future expectations about same than a repudiable handle does.

  119. I believe the “modern hackers” concealment might have more to do with paranoia than anonymity.This might not have been the case 20 or even 15 years ago but right now, we know the US government and others are keeping tabs on everyone and anyone. You might have nothing to hide, and might not even be worried about reputational issues, but making the job easier for people who are very much interested in denying you your privacy, will definitely set most hackers on this course. Exceptions have to be made for this as well.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>