Hacker News Reader

Correct, However, assuming this is all true, the developers (whose work is being infringed[a]) could get an injunction saying Cheat Engine can't release the infringing versions anymore, and must remove the infringing code going forward. And if Cheat Engine refuses, they could get an order against GitHub to remove the infringing works (the releases and possibly the entire repo).

[a]: According to a comment on issue #60[0], this would be any of: the FSF, Fabrice Bellard, Timppa, Martin Waldenburg, Jean-Franois Goulet, or Bo Brantén

[0]: https://github.com/cheat-engine/cheat-engine/issues/60#issue...

(The only actually infringing file was authored by Bo Brantén, and it has been removed from the repository as of 4 years ago)

I wonder how many threads patience and reading comprehension would kill outright?

> Importantly, the new code in the infringing product is still copyrighted, and that copyright is owned by the person who wrote that code even though they also violated the GPL license of other code.

I thought the "one GPL file touched these other files and made them all GPL too!" vibe coming off the thread felt a bit weird! But I'd read something about the GPL being a viral license somewhere so wasn't quite twigging that no that's not what that means. Thanks.

I've incidentally been looking for a good dense steak to chew through on software licensing that's maybe dense but leaves me with a properly calibrated mental model that lets me answer simple non-normative questions without pining for counsel :). Something with good signal-to-noise ratio that eschews circumstantial drama to focus on theory and how things (have) play(ed) out in practice in nuanced and unintuitive situations. I wonder if something like this even exists...?

"Viral" just means that if you use GPL code, you need to release derivative work as GPL as well. If you refuse to do so, you are in violation of the license, but you still retain copyright on your code.

Microsoft FUD morphed that into "if you use GPL code, all other code you write automatically becomes GPL (and you lose copyright)", which is wrong.

I would suggest finding a law professor who writes about open source and asking them if they recommend any casebooks (voluminous textbook full of long excerpts from judicial opinions) or hornbooks (more succinct summaries of an area of law, like textbooks in other fields of study) on open source and/or software copyright generally.

To really get the issues discussed in this sub-thread and at issue in the SFC/Vizio case, you would want to know about contract remedies, privity of contract, third party beneficiaries, license agreements as a special species of contract, standing, and maybe some other areas too. But I think software engineers are actually well suited to understanding law, especially if they also have an innate interest in political/moral philosophical questions.

Wow, thanks very much for this info. I was honestly just chucking the question out there out of general curiosity - TIL what I was actually asking for :'), once I get the lid back on the mental vertigo I'm definitely going to be following up on this in the long term. I've been wanting to learn more about the subject for a while.

Be careful when choosing which versions of which licences you use, if you are mixing them with other licences, as this can alter your options if found to be (unintentionally or otherwise) in breach of them. GPLv3 and CC v3 & later licences have a cure provision clause which makes it clear that you can simply stop using the code with 30 days as one of your options to rectify the breach (the other two key options being to negotiate different licensing terms, or in the case of GPL go the GPL route yourself). CCv2 and GPL2 did not have these clauses and this can lead to copy-left trolling as discussed in https://archive.ph/N84oj

That has been true so far, but do note that Software Freedom Conservancy is currently involved in a lawsuit with Vizio in which they hope to establish precedent that recipients have standing for GPL violations.

> No one is going to touch GPL code with a ten foot pole if it gives the entire public standing to sue you.

Then just... actually follow the license? The only way for it to be a problem is if you're using GPL code and not releasing it under the GPL; the only change is who can try and force you to actually follow the license, not the actual exposure.

(IANAL, but to the best of my amateur knowledge this isn't particularly complex or controversial)

That's not how the law actually works. It's hard to know what is or isn't fair use, or even what is or isn't infringement. What exact legal effect the GPL, LGPL, and AGPL have are hotly contested matters too. If you know who holds the copyright and can sue you, you know who to negotiate with and if a dispute does occur, you can settle it for a reasonable amount of money. If anyone in the public can sue you, those options disappear and your legal peril can be dramatically higher.

I now see how the Software Freedom Conservatory's lawsuit against Vizio is notable in that it seems to do most these things.

> Let me clear up a huge and very common misconception: an (alleged) GPL violation like this doesn't mean that you, the recipient, suddenly has the right to demand that the rest of the code be released as GPL

That's not clear.

> All it does is give the copyright holder of the GPL'd code a legal claim against the developer of the derivative product.

That's a hotly debated topic, because of the “intended third-party beneficiary” doctrine in contract law, which clearly gives people other than parties a right to sue (why it is debated is because it is debated how the doctrine applies to the GPL specifically, and it may differ among different particular cases where the license is used, since it isn't just a question of the bare contract language.)