Email or username:

Password:

Forgot your password?
Top-level
Glyph

@mcc Did copyleft licenses ever meaningfully restrict the behavior of large corporations? Licenses are effectively a statement of intent with respect to future litigation, and if the copyright holder is not willing or able to actually *perform* that litigation, everyone gradually understands that this is a Mexican standoff where one side's guns aren't loaded.

45 comments
mcc

@glyph Redhat establishing that its GPLed software is no longer available under the GPL seems to indicate the answer to your question is "no".

Glyph

@mcc IIRC, Sony did it much earlier. I cannot even find any record of this, but as I recall, Sony distributed a modified version of GCC as part of their early Playstation SDKs, in a way which clearly violated the GPL. FSF found out somehow, and the result was just that Sony said "oops, our bad, we forgot to contractually forbid members of our SDK program from talking to you" and then later switched to LLVM.

Glyph

@mcc Searching for this today only finds stories about the GPL code included in the Sony-published game ICO, which also didn't result in litigation.

Glyph

@mcc gosh I hope I'm not spreading misinformation here, my memory of this was that it was a pretty pedestrian corporate interaction not some weird urban legend, but it's bothering me that I can't find _any_ news coverage

Farce Majeure

@glyph @mcc So the PlayStation thing was basically the way Cygnus did compiler port sales. The Long and short of it is they'd do a gcc target for a given CPU and agree to not release it, then distribute it only to the customer, but still under the GPL, with the caveat that the contracts said they retained all the rights and could use the source code in other ports, and I think it may have also had a timeout. Then, \

Farce Majeure replied to Farce

@glyph @mcc Sony distributed it to game devlopers, who were under their own contracts and NDAs with Sony and received this compiler as a binary and with no right to distribute it. The contractors didn't receive it under the GPL, but Cygnus still had their own copy that would eventually be upstreamed.

This sort of thing has always been part of the business of making and selling GPLed software.

And also the dig at Red Hat above is still just plain not what actually happened.

Farce Majeure replied to Farce

@glyph @mcc (incidentally I think we stopped doing this in the compiler business not long after Red Hat bought Cygnus.)

Glyph replied to Farce

@vathpela @mcc “didn’t receive it under the GPL” is doing a lot of work there. Like presumably portions of it were copyrighted by the FSF? Were they a party to this contract somehow? Were they okay with binary proprietary licensing?

Glyph replied to Glyph

@vathpela @mcc (thank you SO MUCH for pointing me towards the factual history here!)

Farce Majeure replied to Glyph

@glyph @mcc so there are two parts here - 1) Cygnus owned the code they wrote (which wound up being a strikingly large amount of the compiler) until they contributed it back to the FSF, (though they did have a blanket agreement which did that), and ... \

Farce Majeure replied to Farce

@glyph @mcc 2) in terms of software licensing, as I understand it (and let's be sure: IANAL) if you work for a company and they for instance get a copy of Visual Studio and give it to you, Microsoft has given it to them under whatever license, but your employer has more or less just loaned it to you. You don't have that software licensed to you, your employer does. And that's something that can be established through non-employment contracts as well.

mcc replied to Farce

@vathpela @glyph This (conveying code from company A to B to C while claiming it was not "distributed") seems to violate the plain language of "copying" or "distribution" as it would have been understood by people who place their code under GPL2, and I doubt a court would sign on with it if this were tested. In GPL3 "distribution" is replaced with a more fine-grained definition of terms concerning copying, and so if this loophole was ever real, it's probably shut now.

mcc replied to mcc

@vathpela @glyph At any rate if the behavior in question has been halted for unrelated reasons it seems hard to come to a firm conclusion about whether it was entirely legal?

Farce Majeure replied to mcc

@mcc @glyph I think that's a fair conclusion. Also worth noting that there are other slightly different ways they could have done it that are even more confusing (and maybe they did, I wasn't ever in that part of the business). Like, what if they only ever gave the customers patches, and then rented them out a consultant who downloaded, patched, and compiled gcc for them? It's exactly the same in the end, but a largely different thing being handed over.

Glyph replied to Farce

@vathpela @mcc thanks again for stepping in so I wasn't spreading around false claims. I still think the whole scheme sounds confusing and dubious, but now at least I know where to go to do research to understand it better in the future!

Farce Majeure replied to Glyph

@glyph @mcc Also worth keeping in mind that this was like 25 years ago, I didn't work in that part of the company (or indeed for Cygnus ever), I've never paid someone to write a compiler port for me, and it would be remarkable if I didn't have at least some part of this history wrong ;)

mcc replied to Farce

@vathpela @glyph Hrm, well, now that you mention it, I think there might actually *be* a very dangerous patch distribution loophole to the GPL, I'm willing to hear that argument. Especially since I've ALSO seen an almost identical loophole used by people who ADVOCATE the gpl (see "LAME"..)

Farce Majeure replied to mcc

@mcc @glyph I think the argument against that is that a patch to a piece of software must be, at least in some part, a derived work. (Also I have written a lot of GPL'd software. ;)

Farce Majeure replied to mcc

@mcc @glyph Well, I don't think lawyers or the courts see it that way, but again I am not a lawyer so I won't belabor that point any more. Maybe as @luis_in_brief, I hear he is a lawyer, just not your lawyer or my lawyer ;)

Luis Villa replied to Farce

@vathpela @mcc @glyph ugh, I’m a lawyer but despite recent appearances I have a day job, so I should really get back to my work of (checks todo list) writing up more opinionating on xz, oh gooooodddddddddd

Glyph replied to Luis

@luis_in_brief @vathpela @mcc we'd all better stop posting so we can get back to our respective jobs, which are *checks notes* also posting, mostly about the same topics, just with a slightly different emphasis

Glyph replied to Farce

@vathpela @mcc @luis_in_brief ultimately the nature of the debate here just proves the point I was making :). We are arguing over what precise powers that a particular genie would have if it were ever released, but the reality is that the genie gets its power from the tens of millions of dollars that would need to be spent to release it in the first place, and we will never find out so it doesn't matter in any practical sense

Geoffrey Thomas replied to mcc

@mcc @vathpela @glyph I don't know how to conclude whether this article is correct or biased but it talks about the question of whether GPL triggers between e.g. a company and its independent contractors, or when a company is acquired: jolts.world/index.php/jolts/ar

Glyph replied to Farce

@vathpela @mcc also not a lawyer, but this sounds novel and bizarre to me. "licensed to" is a shorthand; the license is a limited right to perform copies that would otherwise be considered infringement, which presumably an employee acting on behalf of a company would need in exactly the same way as the company itself would; everybody involved needs the right to make these copies

Glyph replied to Glyph

@vathpela @mcc (I mean, just kidding, as long as the MAI v. Peak / Cartoon Network v. CSC circuit split holds this is all gibberish in US law and I have _zero_ idea how it would hold up in Japan)

mcc replied to Farce

@vathpela @glyph I find in the text of GPL2: "You may not impose any further restrictions on the recipients' exercise of the rights granted herein."

If Sony [or whoever else] made someone sign an NDA saying they could not distribute the GPLed software they gave them, Sony was violating the GPL, and therefore did not have the right to distribute the modified GPLed code in question.

mcc replied to mcc

@vathpela @glyph (The GPL3 has a more complex clause. The fact I find the GPL3 harder to interpret as a non-lawyer is a reason I prefer GPL2.)

keithzg replied to mcc
@mcc @vathpela @glyph I'm a big fan of copyleft-next for this very reason, it's even easier to read as a non-lawyer while still having upsides from the GPLv3.
Farce Majeure replied to mcc

@mcc @glyph right, but that only applies if you're actually distributing the software, which isn't as straightforward as you may hope (see my other reply to Glyph).

Miguel de Icaza

@glyph @mcc folks with secret platforms have been able to distribute GPL code under assorted NDAs, as an external governing contract covers that - and the GPL covers that case.

Luis Villa

@mcc @glyph I think that’s a little unfair to red hat (it’s just *more inconveniently* available now) but as a general matter, yeah. In particular:
1️⃣ copyleft has always been mostly in practice less legal contract, more social contract (though there are important exceptions); and
2️⃣ because it is ~ impossible to revise, over time copylefts almost always become *less* contractual because lawyers for users are constantly shrinking them, while lawyers for creators are rarely expanding them.

Luis Villa

@mcc @glyph like, GPLv2 to lawyers in the late 1990s: THIS IS A BOOGEYMAN THAT WILL EAT YOUR CHILDREN.

GPLv2 to lawyers in the present: oh, yeah, I learned that in second-year copyright, including the standard weaknesses/limitations, and I know who to speed-dial if I need to get details on how to exploit it.

Exact same text, very different real world impact.

Luis Villa

@amgine @mcc @glyph @kat probably so often she’s tired of it, yes ;)

Glyph

@luis_in_brief @mcc I think the best understanding of the Red Hat issue is r0ml's "freedom from source code":

""" the largest Free Software business, Red Hat, has always had “freedom from source code” as its business model. A business pays Red Hat with the same licensing scheme as they would for any proprietary commercial operating system, in exchange for which Red Hat frees them from the inconvenience of needing to be exposed to the source code """

r0ml.medium.com/free-software-

@luis_in_brief @mcc I think the best understanding of the Red Hat issue is r0ml's "freedom from source code":

""" the largest Free Software business, Red Hat, has always had “freedom from source code” as its business model. A business pays Red Hat with the same licensing scheme as they would for any proprietary commercial operating system, in exchange for which Red Hat frees them from the inconvenience of needing to be exposed to the source code """

Glyph

@luis_in_brief @mcc bonus, a lot of the "bad actor" stuff in that essay is even funnier in light of xz

Farce Majeure

@glyph @luis_in_brief @mcc I think this is an interesting take, but ~wrongish in two ways. 1) probably the best understanding of how Red Hat conducts business is the parts of "Spacesuit" (mitpress.mit.edu/9780262015202 ) about sewing patterns as they relate to engineering drawings; 2) the customers' understanding of their relationship with source code varies quite a bit more than that.

Christopher Neugebauer

@glyph
Yes: it's why Linux got massively popular quickly, with open source contributions from notable large-corps, while BSD derivatives just spawned proprietary forks
@mcc

Glyph

@chrisjrn @mcc Is there a good history of this somewhere? My impression of this process is that it was way more complicated than that, particularly that it had way more to do with process than licensing. But I would certainly love to be wrong about this

mcc

@glyph @chrisjrn I do think there were some other complications pushing people away from serious use of BSD during this time. But I think Christopher's argument could be made convincingly.

Christopher Neugebauer

@mcc
The fact that those complications were overcome (eventually) suggests that with the right early adoption and contribution patterns they could have been overcome (expediently), but _something_ got in the way of people wanting to contribute 🤔
@glyph

mcc

@chrisjrn @glyph Well, the "complications" I'm referring to were the lawsuit, and the extremely awkward "advertising" clause in the BSD license that was not removed until 1999. Those things were just going to happen over a certain time period either way (you could maybe even argue without Linux blowing up the advertising clause might have stayed in BSD longer…)

Christopher Neugebauer replied to mcc

@mcc
Fair assessment. Certainly those affected community BSD forks, but I find myself wondering how that threat would have impacted IBM (who were distributing BSD-compatible AIX and still contributing to Linux at the same time).
@glyph

Christopher Neugebauer

@glyph
I'd need to go looking for sources, but my recollection was that Linus was convinced to change licenses from something hand-spun to GPLv2 very early in the process, because legally unsound licensing was a barrier to contribution.

Process was a differentiator, but FreeBSD took a similar process approach and got a headstart from a more technically mature initial codebase.
@mcc

Christopher Neugebauer

@glyph
Wikipedia's source is in Finnish, published by the Finnish Karl Marx Society (en.m.wikipedia.org/wiki/Histor), but pre-GPL, Linux used a copyleftish noncommercial licence.
@mcc

Go Up