Friends of Armed & Dangerous 2019

Once again I will be at Penguicon and hosting a party for all friends of this blog. This coming Friday evening, room number not yet known, it will be posted at the con.

Those of you who participated in the design of the Great Beast may be interested to know that I expect to receive its successor at Penguicon – a Greater Beast built from a 64-core Threadripper chip. The machine might well be at the party.

UPDATE: room 507, 9pm, Friday

Published
Categorized as General

52 comments

    1. >Tempting. Checking my calendar….

      If you can be there Friday morning we night be able to get you into Geeks With Guns.

      In any case, Cathy and I would be very pleased if you attended.

      1. ESR,

        Check your e-mail please.

        I want to confirm that I’m saving the last two GwG spots for you and Cathy. (Or would it be just you?)

        R

        1. >(Or would it be just you?)

          Just me this time. Though, if Mark Atwood shows up I hope we’ll find room for him.

          1. I have you down. If you bring Mark we’ll fit him in, even if he’s 21 of our planned 20 attendees.

            Eastpoint Action Impact, same as last year. (NOT Southfield)

            E-mail sent with address and phone of range and info

            1 pm safety brief. 1:30 range time start

    1. I doubt it. Penguicon disappointed my wife and I so badly 4 years ago and we haven’t been back since. Every year I check the programming and every year I don’t feel one bit disappointed that we don’t go anymore.

  1. It’s a shame that Penguicon appeals to me not in the slightest….Geeks With Guns seems like the place to be ;)

    How does one even attend GwG anyway? It seems very limited (20 slots?), so is there some selection process? First come first served?

    1. >How does one even attend GwG anyway? It seems very limited (20 slots?), so is there some selection process? First come first served?

      Basically first come first served, yeah. We do try to hold open a few slots for newbies, because acculturating new gunfolks is part of the point.

    2. We have to cap Geeks with Guns for a couple reasons. The first is due to the range. We rent a 10 lane bay at an indoor range that we have all to ourselves. Their range rules limit it to two shooters per bay so that caps it at 20.

      Aside from that, about 20 is the max to have it run efficiently and safely. We have held it at an outdoor range in the past, and may do so again, and will likely allow a few more people then the indoor range.

      The con is also paying the range fee and 20 is the number they’ve set in the budget.

    3. Oh, and I forgot:

      To register contact me at guns@penguicon.org.

      This year is full but I can put you on the “please contact” list for next year.

      The e mail address goes to whoever runs the event so if I step down (unlikely as this is my first year running it) someone else will have access

      1. The e mail address goes to whoever runs the event so if I step down (unlikely as this is my first year running it) someone else will have access

        And I just wanted to say here that Rob graciously agreed to take it over from me after I was unable to attend the ‘con this year. I appreciated all his help in the past, and am very glad that he was able to carry it forward now and into the future.

        Thank you!

    4. Thanks for all the feedback guys :)

      Since I am not inclined to attend Penguicon, I will not be making the long drive for GwG alone. If you ever do a GwG event elsewhere, I’ll see if I can swing it, and drop you a line if so.

      I’m far from a newbie anyway, so it is far better that I do not occupy a slot that a far more deserving (and needy) person would benefit from.

      But it would be cool to hang out with you geeks and make some noise ;)

  2. I managed to make it to Penguicon and FAD exactly once. Penguicon was fun, but mostly because I hung out around the gaming area and went to only one panel (on slide show improvisation, which was just good old-fashioned fun). FAD was fun in the same way it sounds like GwG would be fun. Alas, I can make neither, because I’m working at Baltimore’s Kinetic Sculpture Race this weekend.

    That said, it’s starting to sound like GwG, not Penguicon, is the place to be.

  3. We now have a definite room location: 507.

    Party starts 9pm Friday, ends when I throw the last of y’all out.

    1. If they’re right, then everything the FSF has ever said about the GPL is wrong. The GPL explicitly says that a licensee need not have paid a dime to the licensor, and it’s always been promoted under the idea that the benefit contributors receive is the value of others’ contributions, so they can’t claim there was no consideration and therefore no contract.

      1. Actions speak louder than words. The FSF has always and continues to require copyright assignment for any “contributions” of code they will accept.

        >”The GPL explicitly says that a licensee need not have paid a dime to the licensor,”

        Indeed: the licensor has not required consideration from the licensee.

        >”and it’s always been promoted under the idea that the benefit contributors receive is the value of others’ contributions”

        The licensees are not required to create derivative works as “payment” for the permissions given in the license. There is no bargained-for-consideration here; just speculative future benefits not required to take under the license.
        Sapna Kumar’s paper touches on this, read it.

        >”, so they can’t claim there was no consideration and therefore no contract.”

        They can claim exactly that: since it is true. The licensee is not required to perform any act benefiting the copyright holder. He is not required to pay the copyright holder, he is not required to do any work for the copyright owner, nothing, (in order to receive these permissions and “promises”)

        In return he gets: the permission to use the software, modify the software, create derivative works of the software, and even DISTRIBUTE derivative works of the software alongside the license text under the same permissions specified by the copyright holder, and DISTRIBUTE modified or original versions of the work of authorship under the same permissions specified by the copyright holder.

        All for nothing. Absolutely nothing.
        And he(the licensee) doesn’t have to do any of it.

        It’s not consideration.

        1. This whole discussion is a waste of time, but I have time to waste tonight, so here goes (with standard not-a-lawyer disclaimer)…

          When you received a copy of $project, you did so under the GPL. That (and only that) gave you licence to distribute derived works (which includes patches), and only subject to the condition that said derived works also fell under the GPL. Therefore, (a) the consideration you received was the original licence, and (b) if you revoke the GPL on your contributed patches, you are yourself in violation of the GPL on the version against which those patches were developed.

          Of course, talk of ‘consideration’ is itself dubious when bearing in mind that the GPL is a copyright licence and not a contract.

          There is also a further question of legal interpretation here which might independently break your argument: when you receive a licence to a derived work, do you also require licences to every ancestor work, or does the final licence “stand on its own” as a grant? (The former possibility is analogous to “chain of title” in real property, the latter more closely to the law of personal property. I’m sure Eric will correct me here with some finer details.)

          If a chain of title is required, then each act of distribution is subject to all the ancestor licences, and thus revocation of any one licence suffices to enjoin. But if each patch application (creation of a derived work) creates a new licence which stands on its own, then a previous contributor’s revocation could not revoke that new licence save by retroactive effect (i.e. making the past grant of the licence to the derived work illegal), which is contrary to principles of law.

          My first argument, then, suggests that only the first commit of a project can be revoked (all subsequent contributors are bound by the GPL to release their contributions under the GPL), while my second argument suggests only revocation of the last commit is effective (revocation does not propagate). Thus, excepting the edge case of a project with only one commit of history (or, more precisely, a project with only one contributor), revocation is an empty threat.

          (And of course that’s before we get into reliance, promissory estoppel, etcetera…)

          1. >reliance
            No contract, no reliance damages.

            You don’t get to attack people that granted you free permission to use a work.

            “Hey, I get to enter your property and drink from your well FOREVER, for NOTHING, and if you stop me: YOU’LL HAVE HELL TO PAY” – It does not work that way. The owner can rescind that free license at any time.

            >promissory estoppel
            Has been applied to situations where a parent promised a heir land, heir built upon the land, estate tried to withhold land after death of progenitor. Essentially a modern take on the livery of seizen, and would not have been accepted otherwise. Other times promissory estoppel has been found is when a father promised his daughter something: again an explicit one-to-one promise to a heir or heiress; similar to a promised advance on inheritance. A third area that promissory estoppel has been found is when a worker has performed a task and the business owner is attempting to take the work for free arguing a failure of contract: the courts save the worker from being stiffed. It is not known why the courts used promissory estoppel here: they could have used quazi contract, quantum meruit, etc – it’s similar in this case.

            So we see that promissory estoppel is used, essentially, as a substitute for the livery of seizen, or to enforce promised advances on inheritance, or as a substitute for quazi contract and quantum meruit.

            We do not see it ever being used to enrich unidentified alien free takers of non-exclusive copyright licenses.

            The rest of your argument is… simply misinformed regarding what a non-exclusive license actually is: because you are not a lawyer and do not know what you do not know.

            Oh and you shouldn’t separate promissory estoppel and reliance like you do, as if they were alternative theories: they are not really separate: one comes after the other: one is a claim, the other is a remedy. Don’t you know this? If you were a lawyer you would know that. But you are not so you simply repeat garbage other lay idiots have repeated to you.

      2. If the revokists are right, think of the implications: The maintainers of Node.js, Go, Rust, or Linux may decide to apply their CoC to all users of their software, and revoke the license to use the software from persons or entities they deem to be violating, or at risk of violating, the CoC. (That’s the new hotness: preëmptive disciplinary action against persons deemed to be “at risk” of CoC violations.)

        Do you still want to open that can of worms, mikee? Because I don’t think things will turn out the way you hope they will.

        1. > The maintainers of Node.js, Go, Rust, or Linux may decide to apply their CoC to all users of their software

          At which point any sane developer forks the project from just before the CoC-Doom, and carries on as if nothing had happened.

          1. If license revocation works the way the anti-CoC revokers hope, you won’t have the right to fork. Any license you may have had to copy, distribute, or modify the software is revoked. That’s what revoking means.

            This is why revoking is bad juju, no matter which side you fall on.

            1. If license revocation works the way the anti-CoC revokers hope, you won’t have the right to fork.

              However, they can’t do that to a contributor, because their own license to distribute the combined work is chained through the licenses of each and every contributor to distribute derivative works.

              This is why revoking is an asymmetric (and scary) weapon: if either side revokes, then upstream is dead.

              Of course, I still don’t believe that revocation/”rescission” does work that way, but I’ll expand on that question separately.

              1. Indeed. I’m presupposing a lot of ifs here: IF people with the copyright on critical kernel code decide to revoke AND it’s taken seriously by the community AND someone initiates a lawsuit AND it goes before a judge AND the judge, or the appeals judges, favor the revokers, then we’ve got a real pickle on our hands — and one of the consequences of that is if political conservatives feel excluded from the open source process now, they ain’t seen nothing yet compared to what would transpire if all of the above held. And it’s the conservatives (at this point, one ultra-far-right kookster) who are banging the revoke drum.

                1. Jeff Read: the aforementioned “ultra-far-right kookster” is a licensed attorney, which is why he is able to present legal arguments in ways the lay-person can understand. The first battle is to inform the rights-holders that they are indeed rights-holders.

              2. Edward Cree: Starting to understand why the FSF did what it did from the very start? Their lawyers knew all about this.

            2. It does work that way. A license is permission. All non-exclusive unsecured (free) licenses are revocable, they are not grants, they are not covenants, they are simple permission: “You can do this that and the other thing with my property, this that and this other way, but not this forth way: that’s not allowed”.

              Everyone loves free because they do not have to pay anything for the permission.

    2. This again? I see mikeeusa is getting involved. Supporting my suspicion that the anti-CoC revokist crowd mainly consists of trolls and griefers with little if any actual code in the kernel and almost no skin in the game.

      1. I wonder how much of this is false-flag ops on the part of the SJWs. I suspect that a lot of the people who don’t support this madness mostly just keep their heads down and try to keep working, praying that some harpy doesn’t decide to descend upon them.

        1. My guess: none of it. I think most if not all of the revokist shitposting being conducted here and on lkml has been mikeeUSA himself. Note that when one of the revokist posts appears here, it often gets sidetracked into ramblings about how men built open source and how an adult man should be able to marry, rape, or rape and then marry a girl as young as 12 or 13 per Biblical law. These are all mikee talking points. Have been since 2005. They could be a really good faker, but I doubt anyone would risk shitting up lkml with nonsense, trigger the more senstive readers, or just act like an unprofessional ass if they want to remain in good standing.

          Besides, Mikee put his revokist theories to the test earlier this year, by attempting to rescind modification and distribution rights under the GPL from Geek Feminism wiki members to the only significant code he’s ever written: a shitty text mode slot machine game.

          Nothing more to see here. Revokism is a failed revolution by a single man — a man who has made no substantive contribution to open source and who believes that God gave him the right to fuck kids.

          1. >Besides, Mikee put his revokist theories to the test earlier this year, by attempting to rescind modification and distribution rights under the GPL from Geek Feminism wiki members to the only significant code he’s ever written: a shitty text mode slot machine game.

            Show us the court case Jeff. Show us the “failure”. The rescission was not contested by the Geek Feminists, and no one was sued yet. Also your information is 10 years out of date: MikeeUSA now codes QuakeC for a certain project and has for the last decade. SLOC calculator puts the code added alone at worth between 4 and 6 million dollars (hey they Linux Team uses said calculator…), not including the 3d models, textures, 3d maps, etc added by him.

            The project has grown from 16 weapons to over 200, from 0 spells to over 50, from 3 vehicles, to over 30. It includes city generation, with interiors (not just boxes you can’t enter), RTS style build-able buildings, etc.

            Oh, and MikeeUSA is a licensed attorney, what are you?

          2. Lawsuits take money and alot of time, Jeff. No suit has yet been filed: why do you decree that an issue that is only a few months old is “settled”? You may have noticed that the Geek Feminists, as far as I know, have not chosen to re-post the work of authorship in question, nor do they, as far as I know, distribute it at all, post-revocation. Tell me, Jeff, what would they be sued for? Their license was revoked and they have not chosen to redistribute the work or make derivatives post-revocation.

            There was no “attempt”: the license _was_ revoked.
            It seems you are ignorant of property law, Jeff, along with licensing and the differences between licenses and license contracts.

        2. >mostly just keep their heads down and try to keep working, praying that some harpy doesn’t decide to descend upon them.

          How humiliating. All these men bending the knee to women, just as they’ve been taught all their lives as “white men”: “this is what makes you better than the others: you RESPECT (and defer to ) women!”
          (These men, legally holding all the cards, but begging outsiders to not punish them. If only there was some way to fight back.)

      1. Anime poster: is it true what they say about you?
        (Your avatar is of a red headed anime girl or woman gravatar.com/70ed18649ead5aba74f877cfde56114f )

  4. How is the “Greater Beast”? I thought the Great Beast was pretty amazing.

    Will we get a detailed post on all the other goodies that came along with the 32 cores?

    Will the Great Beast go into retirement, or is there a cluster connection with it’s new bigger sibling in the future?

    1. >How is the “Greater Beast”? I thought the Great Beast was pretty amazing.

      Post upcoming once its teething troubles have resolved.

      1. Relatedly, how are you liking Devuan? I’ve tried it myself and find it quite nice — my only real concern being long-term support. Many forks of popular distros — including, most recently, Antergos — vanished into the aether with a snap of Linux-distro Thanos’s fingers. But Devuan may yet be spared this fate because it’s maintained by a dedicated group — and they’ve collaborated with Debian developers in the past on addressing security vulnerabilities and the like so they’re not just haters.

        1. >Relatedly, how are you liking Devuan?

          Can’t say yet. Have it installed on a fanless brick downstairs but haven’t used it for production.

  5. My favorite (Internet) author writing about ESR’s favorite (novel) author: https://www.takimag.com/article/heinlein_in_hindsight_the_moses_of_nerds1/

    Of Heinleins many ideas, the one that impressed me most was that you should learn math and languages because these two enable you to learn everything else. This made me realize why we consider humanities easy today and STEM hard: because people in humanities no longer learn much in the way of languages. Humanities weren’t that easy back when it was expected that, say, philosophers will read both Plato and Kant in the original. And that wasn’t even that long ago, Umberto Eco’s How To Write A Thesis, 1977 was very much in this spirit.

    1. That’s not the only reason humanities are “easier” than STEM. Another issue is that STEM graduates are judged by their employers on the quality (and volume) of their knowledge of the subject of their degree, while most humanities graduates aren’t. In the sorts of jobs that History majors get, it doesn’t matter whether they remember what happened at Canossa, and employers won’t really care if they think Shakespeare’s plays were written by a Jewish woman. So the History (and other Humanities) faculty can slowly reduce the amount of work (they have to grade) to get a degree without getting lots of complaints from the outside, while the Engineering faculty can’t.

  6. ESR: Eben Moglen suggested that he was going to write a paper refuting the notion that “The GPL is revocable”:

    >I think the best procedure would be for me to publish my analysis and
    >for you then to tell me what is wrong with it.
    https://lkml.org/lkml/2018/10/26/718

    Why has no such paper from him materialized?

    1. >Why has no such paper from him materialized?

      Because Moglen is either wrong about or deliberately misrepresenting the law and the “revokists” have it right. Thank you, I had already figured this out from his (non)responses on lkml.

      1. Again, if the revokists have it right, I don’t think you’ll like how that will actually play out. The people (person!) calling for exercise of revocation rights over the CoC have close to zero skin in the game, and thus zero leverage over kernel leadership.

        However, future kernel leadership now has a shiny new toy to play with, and may well decide to revoke the rights to use, copy, or distribute Linux from anyone found to be violating the CoC — or even thought to be “at risk” of violating it.

        1. Remember, individual contributors to Linux retain copyright on the code they contribute. I feel like this could easily lead to a “Mutually Assured Destruction” scenario, where people on both sides decide to revoke because they don’t like what the other side is up to. The end result then is not merely “badthinkers” being punished, but the entire project shutting down. This is what Cree was getting at above.

          1. Actual hackers would pick up the pieces then and continue the project elsewhere, and anonymously (as it was in the beginning before the “real name” policy took effect): hackers are not so concerned with US legalities and never were. A gentleman’s agreement is good enough (what the open licenses actually are). Yes the corporate backed “project” may shut down: but that is no-longer a hacker project anyway: it’s hacker hostile.

            There wouldn’t be a CoC either, and useful patches like GRSecurity and PaX would be accepted. Garbage like systemd (which is not a kernel patch but wishes it were) would be seen for what they were rather than forced down everyone’s throat on pain of banishment.

            Patches from men like Hans Reiser wouldn’t be shunned: the quality of the work is what defines the patch; not the person. Engineers working for the defense industry are but-for causes of countless murders anyway; it’s always been part-and-parcel of the profession (The CIA actually thanked the developer of one of the drone subroutines for the slight inaccuracy – helped kill more dependents of the target).

        2. >Again, if the revokists have it right, I don’t think you’ll like how that will actually play out.

          Maybe I won’t, but the law is what it is. We can’t solve any problem by denying that reality.

Leave a comment

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