The GPL has been validated in court many times by now. That doesn't mean they'll win this case but there's a significant amount of precedent protecting it at this point.

Edit: It's remarkable to see a raft of uninformed comments here. What's up with that? Once most people in HN knew the story of Linux, Gnu and so forth.

The more topics you are knowledgeable about on HN, the more glaring the mass lack of competence the comments generally have except for topics like algorithms. Especially on anything bordering economics, medicine, and some others. It's similar to knoll's law. The libertarian / austrian skews are sometimes wild, and equally wildly inaccurate, sometimes bordering on conspiracy / delusional thinking.

> The libertarian / austrian

I am Austrian and I am a bit confused...

sorry, meant specifically austrian school of economics there. Austrians the people are cool :)

No worries! TIL

apparently he has a problem with your schools though.

I don't think I've gotten that much more knowledgeable in the last few years but maybe I should take that as the explanation for HN seeming to decline - it's certainly the most flattering possibility.

Every now and then when I've got some time to kill I'll go look at old comment sections (particularly when Dang links previous iterations of a discussion) and compare it to what I see today. It turns out that the quality hasn't changed greatly, though viewpoints, popular gripes, etc may have. I've discovered that a lot of my thoughts about "hn is going to shit" are usually better phrased as "I dislike that this viewpoint is popular now" or "I've come around to seeing that differently". Particularly when it turns out I preserved my old viewpoint in the form of a comment on the old article.

A related thing is: look through places where the crowd skews young/newb like r/programmerhumor. You'll see arguments, discussions, etc about things that might make you think "I thought that was settled ages ago" but in reality were just things you got experienced perspective on, and/or grew out of.

People grow and change - usually in a slow and steady way that's hard to see in the mirror. I suspect you've picked up a lot of knowledge and some new perspectives - a few years is a long time!

Who's Unix? I only know Steve Jobs and Jim Cramer.

> That doesn't mean they'll win this case

As of yet there is no case to win or lose.

HN is becoming the tech sibling of 4chan

hn is much more ... extreme

It's only a bluff if we do nothing about it. Maybe the government doesn't have the enforcement spine for the job, but if they can make it clear who is in violation, well now we have a list of targets. We just need to create some incentives to not be on that list.

GPL has always been mostly a bluff, it's occasionally attempted to be enforced but its a drop in the bucket compared to the ocean of infringement.

This gives me hope for using Arduino and LGPL.

I want to make a product, I don't even mind giving away the code. I'm just worried since its an embedded system, its hard to update or whatever. As many times as I read the legal stuff, its over my head. I can't understand how you can let someone update part of the code on something compiled. Its not like I'm going to have a screen on my device to let users read the code either.

If someone wants to fight a legal battle, its really due to my incompetence, not malice.

For [L]GPLv2 it's super simple - release any changed code and anything that links closely to it. If you publish all your source you basically can't fail to do this right.

For GPLv3, it's that simple most of the time. You talk about embedded which suggests you may be thinking primarily of the "tivoization" clauses designed to prevent user lockout. Again, if you're a fully open source project this will be hard to fail because the user should be able to just copy whatever you do in development. It's when you try to prevent user updates while allowing your own that this applies.

That was basically my thought, I'm just worried that if I link to the code online, that someone could claim 'Well I don't have internet'. Bam lawsuit.

I suspect that publishing it public to the internet is sufficient. If you really wanna cover your ass offer to mail someone the source on a CD for "reasonable handling fees"... which can reasonably be fairly high if someone demands such service.

This was changed in GPLv3. To my reading, section 6(b)(2) means that if you're distributing a physical product, you can just include a link to a URL where the source can be downloaded for free.

You could read 6(b) to mean that you have to offer consumers the choice between physical-delivery-at-cost and for-free-download of the source... but I don't think that's the right reading.

Maybe offer the source code at the point of purchase? Could even sell USB/SD with source code on it. Folks who want the code offline can buy them and copy the code where they want it.

OTOH, the internet is a "medium customarily used for software interchange" so maybe that would be fine, but the internet wouldn't "accompany" the device. Also, in GPLv2 you are allowed to just send a note "you can get source from us delivered on a storage device for $X", so if you do the purchase thing, plus write a note with every device about how to get the code on the internet or on disk, you should be fine.

I think it is their stated aim at sfc to enforce compliance, not to sue. If they see you are not complying they will ask you to comply. Doesn't sound like something to be scared of?

Indeed, and their principles while enforcing compliance are eminently reasonable:

https://sfconservancy.org/copyleft-compliance/principles.htm...

>I can't understand how you can let someone update part of the code on something compiled.

You can provide your closed source part as a precompiled library or object file together with the source code for the LGPL part and build scripts that let the recipient relink it with a (possibly modified version of) the LGPL code to generate a new final binary.

A pain yes but not impossible. Having a screen or not is irrelevant, nothing says the recipient has to access the source on the same device as the binaries run.

IANAL, but I don't think that the part in the GPL about modification makes any requirement that you facilitate modifying compiled binaries. My understanding is that the requirement is essentially that distributing the binary requires you to distribute the source as well; if someone modifies that source code you provide and distributes a binary they compile from that modified code, they need to share the modifications as source as well.

This is also my understanding of why the AGPL was created to try to close the loophole of running GPL code on the backend; since running code on a server doesn't entail distributing the server binary to people who access the service, the GPL didn't enforce any requirements on sharing any modifications to code that was modified and used as part of the backend for the service. Some felt that this went against the spirit of the GPL, so a new license that made this also explicitly disallowed was created.

As for the LGPL, my recollection is that it differs from the GPL in creating an exception for dynamically linking to the compiled binary; the phrasing of the (non-L) GPL requires that any "derivative works" of the GPL-licensed product (I can't remember if that's the exact term, but it's a something similarly general) must abide by the GPL, which could be interpreted as applying to something that dynamically links to a library licensed under the GPL. Licensing something under the LGPL gives permission for users to dynamically link to your product without requiring them to share the code linking to it.

The point of the LGPL is that it allows static or dynamic linking with proprietary software, but it requires that users be able to modify the LGPL part and relink it with the proprietary part and have the result work, as long as the modified part is compatible with the proprietary part. With dynamic linking thats simple, just drop in a new LGPL library, but with static linking you have to share the .o files for the proprietary software to enable user relinking.

I'll chime with my experience in embedded systems. In the last 2-3 companies I've worked with the directive has been "no GPLv3/LGPLv3, period. You will be fired if you ship with it." The legal departments just don't want to deal with it at all.

GPL for Linux is not a Bluff, it is a stated fact they (the Linux Foundation) will not enforce it. Publically stated, and supported by their corporate overlords... err "sponsors"

Can you cite the page?

I found a page talking about how they won’t enforce their trademark against people who use it, but this is not the same as not enforcing copyright.

Thanks in advance!

If you know, and followed the enactment of the "Linux Kernel Community Enforcement Statement " [1] [2] [3] which started out as an effort to stop a "rouge" developer from being a copyright toll, then further devolved with interactions with the core linux group and SFC based on a fundamental difference in the understanding of the GPL problem, and what GPL enforcement should be

SFC wants to use the law as a stick, while trying to get people to comply with out legal action. Many core members prefer to not even engage with the threat of legal action at all, they believe that simply asking nicely will get companies like VmWare to comply. of course this is rarely true but.....

[1] https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/drive...

[2] http://kroah.com/log/blog/2017/10/16/linux-kernel-community-...

[3] http://kroah.com/log/blog/2017/10/16/linux-kernel-community-...

I didn’t know about or follow any of that, so your links were helpful. Thanks!

After looking into this and reading your three links, I couldn’t find a conclusive answer to my follow up question, so maybe you know: did the Linux Foundation ever serve a legal purpose of pursuing copyright violations and take point in defending the copyright holders that contributed to the Linux Kernel?

Near as I can tell, it appears to be largely up to individual copyright holders.

Even that agreement seems to be an article of trust from individuals saying they’ll abide by the extra paragraphs and not doggedly pursue copyright violations so long as violators act in good faith about taking corrective action in a timely manner.

The "leadership" of Linux foundation, have been (and I am paraphrasing) quoted at various talks on the subjects in Q & A;s saying they can not conceive of a scenario for which they would choose legal action over attempting non-legal action (such as Public naming, PR campaign's etc). They simply reject using the legal system to enforce the license.

i am not aware of any legal action the Linux foundation, or anyone in the leadership of said has participated in, in a few instances, like the one that lead to the Enforcement Statement they have actively tried to kill the legal action.

The Enforcement Statement is vague purposefully as to what "good faith", and purposefully never spells out when it is OK to move to the courts... because in their view that answer is never

So in other words the Linux Foundation is not a litigious organization and therefore contributors must enforce their copyrights on their own to the extent that they are interested in them. I’m not intimately familiar with this organization but if they weren’t organized as a legal foundation to pursue copyright enforcement of Linux source code then it makes sense they’re never involved in lawsuits like this. It could very well just be outside its core purpose and well I guess that just makes sense since contribute retain the rights to their code and being a contributor doesn’t automatically establish an attorney-client relationship.

I managed to filter through all the SCO and most of the non-Linux lawsuit stuff gumming up Google and near as I can tell, the only lawsuits involving copyright violations are or Linux specifically are pursued by individuals or agencies like SFC.

So given all that I would read this enforcement statement as a PR move more than a definite article, less “this list of people is effectively giving up their copyright” and more “so you know that one jackass? Well this lot undersigned here won’t be like him. Please don’t stop supporting and using Linux.”

Just as an aside, what constitutes “good faith” is something that can and usually is regardless of whatever contract or license is in effect litigated in court.

This is like saying copyright is a bluff because a corporate violates a person's rights and that person doesn't have the resources of a corporation to immediately go after them.

It will be grim, since everything will be silos, and monetized, and perhaps not even interoperable anymore, to really emphasize the lock-in.

Is there not some sort of legal defense fund organization to help enforce these licenses?

The article this thread is about was posted by Software Freedom Conservancy, who enforce copyleft licenses, for the Linux kernel and other projects. You can help fund their efforts with donations.

IMO, the biggest mistake of the whole FSF, and by extent, GPL way to do things, was expecting that they could 'hack' the intention of copyright and use it against itself as "copyleft"

somebody else said "you cannot fight the master with the tools from their shed" (or something along this line)... IMO, that paraphrased quote points at the same underlying principle that I now consider the reason the FSF project has been slowly failing.

or possibly already failed... nobody is talking about liberty and software anymore, it's now all about the 'open source' of the code....

and guess what? code is now obsolete, GPT and other LLMs are the future of programming languages. this is why e.g. DreamWorks just open sourced whatever, it doesn't matter anymore.

today they teach algorithms and data structures, but it's now all about prompt writing.

in conclusion: LLM PROMTS ARE THE NEW PROGRAMMING LANGUAGES