Live data from Hacker News

Linux's GPLv2 licence is routinely violated (2015)

devever.net

31–40 of 202 posts

Re: Linux's GPLv2 licence is routinely violated (2015)

#31
post #12

I used to use the GPL license to license my software, but now it seems like it is too much of a legal nightmare, so I've stopped using it.

GPL is a nice idea...but you need lawyers lots of money and time. I hate lawyers and i don't have time for that bs nor do it have/want money to spend for lawyers, i was a diehard gpl evangelist in my 20`s, now it's BSD/MIT/ISC (and not diehard anymore...more like IDGAS)...and i really don't care for religion anymore...just quality and real freedom.

> and i really don't care for religion anymore...just quality and real freedom.

The freedom that GPL is about isn't about the developer's freedom to do whatever they please but about the user's freedom from the developers' whims. This isn't some arbitrary nebulous religion-like dogma but comes from the very practical issue of developers inherently having power over the programs they write and, by extension, power over the users who rely on those programs.

Remember that the entire free software thing didn't start because the Emacs god visited Stallman in his sleep but because he was denied access to the source code that would help him fix his lab's printer.

Re: Linux's GPLv2 licence is routinely violated (2015)

#32

> The userspace ABI is stable and intended not to cause anything targeting it to become a derived work. But once again, this seems to be a way in which the kernel project seems to think it has the power to interpret the GPL. In their minds, targeting the userspace ABI doesn't make a derived work, but writing a module does, unless it only targets GPL_ONLY symbols, in which case for some reason it isn't. I think this w…

> I think this whole part applies as much to the text of the GPL itself as to the common Linux explanation. The GPL draws a distinction between dynamic linking, static linking, and use over a network or through a CLI style interface to decide if a work is a derived work of the GPL program.

I just rechecked and it does not seem the GPLv2 does such things. I've not rechecked for v3, but from memory it does not do that either.

I completely agree with you that clear intent is important though.

Re: Linux's GPLv2 licence is routinely violated (2015)

#33
> Actually, it gets worse. Here's what the GPLv2 says about termination:

> So if you violate the licence, it is terminated. That's it. Given that “you” may refer to a corporation, this creates the following disturbing possibility:

This is called the GPL death penalty, has actually happen in the past even before the article was written.

It is yet another thing fixed by the oft-maligned GPLv3. Samba switched to GPLv3 because of it. https://ftp.samba.org/pub/samba/slides/linuxcollab-why-samba...

Re: Linux's GPLv2 licence is routinely violated (2015)

#34
post #23

> It is rather ironic that in some ways, one has more freedom in how one licences a kernel driver for Windows than the freedom one supposedly has in writing a kernel module for Linux. it is not ironic at all: freedom in Free Software is the freedom of USERS of the software to be able to see and modify the source. This is a freedom granted by the writers of the software, and copyleft means that this right is ensured t…

Yeah this is not a "no license" freedom situation. People routinely get this wrong.

Re: Linux's GPLv2 licence is routinely violated (2015)

#35
I rather use Mozilla Public License 2.0, Eclipse Public License 2.0, or EUPL 1.2 for copyleft purposes.

Strong copyleft as a concept has no legal reality. Linking a program to another don't produce a derivative work.

https://joinup.ec.europa.eu/collection/eupl/news/why-viral-l...

FSF is full of baloney about how static or dynamic linking is different than communicating for example through sockets. To be able to use the program I have to link it. What manner I choose to do that or available to me should not matter. I should also be able to reproduce portions of a program necessary for linking. Interoperability does not interfere with the legitimate interest of the licensor, or conflict with normal usage.

Re: Linux's GPLv2 licence is routinely violated (2015)

#36

Good writeup on the intricacies and consequences of the Linux kernel being GPLv2 licensed. The actual headline is debatable. However, It seems to me that the license doesn't matter that much, since everyone has a vested interest in Linux being exactly the way it is, GPL_ONLY weirdness and all. It's unlikely that someone will show up and fork Linux and explicitly break the license (using GPL_ONLY symbols in a non-GPL…

Except if VMware does it. In that case, they're welcome to join the Linux Foundation. (previously on HN https://news.ycombinator.com/item?id=9151799 )

Re: Linux's GPLv2 licence is routinely violated (2015)

#37

I rather use Mozilla Public License 2.0, Eclipse Public License 2.0, or EUPL 1.2 for copyleft purposes. Strong copyleft as a concept has no legal reality. Linking a program to another don't produce a derivative work. https://joinup.ec.europa.eu/collection/eupl/news/why-viral-l... FSF is full of baloney about how static or dynamic linking is different than communicating for example through sockets. To be able to use t…

That'd be more akin to the Apple Public License, right?

Re: Linux's GPLv2 licence is routinely violated (2015)

#38

I rather use Mozilla Public License 2.0, Eclipse Public License 2.0, or EUPL 1.2 for copyleft purposes. Strong copyleft as a concept has no legal reality. Linking a program to another don't produce a derivative work. https://joinup.ec.europa.eu/collection/eupl/news/why-viral-l... FSF is full of baloney about how static or dynamic linking is different than communicating for example through sockets. To be able to use t…

I wish the FOSS community would rally around this idea.

Re: GNU/FSF, I think they've done a real disservice to the FOSS community misrepresenting the state of American copyright law. I wish someone with credibility with devs/FOSS users would finally say: "This is such an illusion, it's mostly a joke."

Re: Linux's GPLv2 licence is routinely violated (2015)

#39

> The userspace ABI is stable and intended not to cause anything targeting it to become a derived work. But once again, this seems to be a way in which the kernel project seems to think it has the power to interpret the GPL. In their minds, targeting the userspace ABI doesn't make a derived work, but writing a module does, unless it only targets GPL_ONLY symbols, in which case for some reason it isn't. I think this w…

I think there's some confusion in the above post.

The relevant "derived work" definition for the purpose of linking isn't one stemming from copyright law, it's the one in the GPL itself. This is because the GPL defines what licensees are permitted to do with the licensed work, and it imposes restrictions on those rights. Copyright law by default lets you make very little use of the software - the GPL broadens that out to include various forms of use, if (and only if) "derived works" fall within its scope.

You could ignore the GPL and static-link a piece of software to a GPLed library, but if you did so, you'd legally need a different right to use the GPLed library, because you hadn't complied with its license terms and so only have your minimal rights to use someone else's intellectual property.

Said differently, the GPL doesn't try to apply itself to "derived works" because of some contralegal dictum they don't have a different creative origin. It merely says "you have two choices - license the things you link with OUR stuff under OUR terms, or you don't have the right to use our stuff".

Post reply on HN