GNU C Library Looking To Drop FSF Copyright Assignment Policy

[coder]

If you have nothing but your hate and anger to spread then you do not reply.

Stop redirecting. It's a simple request I shouldn't even have to make.

And I wasn't angry until you started projecting things onto me that had nothing to do with my reason for asking that question.

[sdack]

OMG. WTF? I asked a very straight forward question. If you're not going to address it, don't fucking reply.

If you have nothing but your hate and anger to spread then you do not reply.

[coder]

Rather than drawing a line, should one get over the differences and put them aside. A lot of nonsense has gone viral on the Internet and not all of it was certainly good. The issue with Stallman was born from a single email,

OMG. WTF? I asked a very straight forward question. If you're not going to address it, don't fucking reply. You don't need to springboard off my post to air your ideological theatrics.

opening up licenses to make contributing to a project easier is certainly a good idea.

You should well know that two people can have a very different take on what "opening up licenses" means. The licenses are different because the underlying motives are different. You're not going to "fix" that.

[sdack]

Can anyone meaningfully speak to the amount of GNU software left in Android? I think it's not much.

Rather than drawing a line, should one get over the differences and put them aside. A lot of nonsense has gone viral on the Internet and not all of it was certainly good. The issue with Stallman was born from a single email, that was misinterpreted, either accidentally or intentionally, and then used to feed into the cancel culture. Cancel culture is no different from racism in that it is used to exclude and separate people rather than to communicate and to unite. And we have seen these senseless acts of polarisation lead to a riot at the US capitol. And people are still not done with this nonsense.

Assuming we can get past all this will there be no need to separate GNU, UNIX and Linux from another. That said, opening up licenses to make contributing to a project easier is certainly a good idea.

Of course we can also talk about sugar, and how it was the single biggest driver for slavery and racism, and that we need to remove it from all our foods. Instead, do we eat it more than ever before, such much that it makes us fat and sick. Cancel sugar!

[Filiprino]

Some consider that a feature. Newer versions of GPL have some controversial aspects that people don't necessarily want to apply to their contributions.

IMO, this is actually the biggest point in favor of letting contributors retain their copyrights.

Only corporations prefer GPLv2 over GPLv3. Only corporate entities (and some people working for/with them) prefer copyright vs copyleft. Copyright retention weakens the project. Its only use is to defend patent holders.

This is a flaw in the license. They should've added a clause where ownership reverts to the Linux Foundation, if the original owner is deceased or defunct, and no attempt is made to reassign it to another entity for like 6 months or whatever.

They could always adopt such a clause, for new contributions.

Or they could assign their copyright from the beginning. But corporations prefer to retain copyright because that way they can defend their patent portfolio. That's why they choose (L)GPLv2 too.


Over there in the USA you have a great roadblock with software patents. And the worst thing is that's viewed as legitimate.

[coder]

Look at Linux. That project is stuck at GPLv2 forever. No one can convince every single contributor of such a big project to upgrade the license.

Some consider that a feature. Newer versions of GPL have some controversial aspects that people don't necessarily want to apply to their contributions.

IMO, this is actually the biggest point in favor of letting contributors retain their copyrights.

some Copyright holders no longer exist, good luck with those

This is a flaw in the license. They should've added a clause where ownership reverts to the Linux Foundation, if the original owner is deceased or defunct, and no attempt is made to reassign it to another entity for like 6 months or whatever.

They could always adopt such a clause, for new contributions.

[Filiprino]

Phoronix article is pure sensationalism.

From the article:

I find this oddly amusing. They want the benefits of a fork without any of the responsibilities.

https://sourceware.org/pipermail/lib...ne/127616.html
Yeah. It's easier for them. This move is a bad decision with zero future-proofing. Look at Linux. That project is stuck at GPLv2 forever. No one can convince every single contributor of such a big project to upgrade the license. The same will happen to Glibc and GCC in case a GPLv4 is released and an upgrade is wanted.

In the end, private corporations have taken over and their capitalist interests are a priority. Do the contributors working for this companies belong to a union or something? Do they thing they are not being exploited by capitalism as the rest of the proletariat? These changes benefit corporations, not people.

There are already 704 source files from various corporate copyright owners. And they don't care to change those sources (some Copyright holders no longer exist, good luck with those) or improve their bureaucracy and reassign copyright. Corporations will increase their footprint in free software projects instead of reducing them making future changes difficult, which in turn will block innovation and put barriers in their business, too.

Even a single author has bureaucracy problems: https://sourceware.org/pipermail/lib...ne/127631.html

A better approach would be a promise to assign copyright within a defined time frame once the patch has been reviewed and accepted. It would mean another patch changing copyright only. Or even better: a CLA or an agreement to allow the GNU Toolchain to assign copyright in name of the author. That would allow bulk copyright assignments.

A DCO is worse for the project.

If I ever contribute to these projects I'd assign the Copyright to the FSF, something which is still allowed. It's not like they're banning it.

[GI_Jack]

A few points.

1. for those who like public domain:
Check out CC0 by creative commons

https://creativecommons.org/share-your-work/public-domain/cc0/

2. Regardless of what RMS has done in previous years, he is woefully out of touch with the needs of both tech activism, and needs most users have with modern computing devices. he still spends most of the time on the command line, sans X, only starting X for things like web browsers and such.
3. The FSF has been lost, and close to rudderless for decades, and most of the slack has been picked up elsewhere. The campaign against unfree javascript while crucial technologies like Replicant languish and are backburnered. It might be true while a lot of these people where inspired by previous actions of the FSF in previous DECADES, they have hardly been on the front lines of tech for a very very long time.

There is a lot of trolling, copypasta, and shitposting, of which is to be expected in a high profile drama thread. I am going to post this for the uniniated reading this who may want a straight answer.

1. UNIX came first. Lets be really honest about what UNIX was, and why it grew. UNIX was developed at AT&T bell labs. It was NOT originally a commercial project because AT&T was not allowed to sell computer stuff being a state sanctioned monopoly. It was made because it was useful internally. AT&T shared the source code because it couldn't sell it. Often, because everyone had the source code, people would fix bugs and sent back fixes to AT&T, and each other. One such big user of UNIX was Berkeley University, a public university that developed fixes and tools that ran on UNIX and distributed those for Free. Much of this software got included in later AT&T versions of UNIX. This software package was known as the Berkley Software Distribution. You likely know it by the acronym BSD.
2. There was a culture of sharing software that revolved around UNIX. Humans don't live in a vacuum. There is a relation between culture and tech. Long have cultures formed around users of similar machines. So even if this was made largely by a company. It was distributed with source code, and there where a community of programmers that enjoyed sharing code that came from the environment.
3. UNIX was made close source in the 80s, and AT&T was allowed to license and sell UNIX so it did. Many companies at this stage sold computers with UNIX derived operating systems. BSD itself was still free, but need an operating system to run on, of which was no longer free.
4. There were a lot of people at this time that enjoyed and wanted to continue the community that revolved around UNIX now that AT&T was no longer offering UNIX for free. Many of these people are programmers, just as good as AT&T, and have provided code that was later included in UNIX which AT&T profited off, of which they received no money for. These innovations include the TCP/IP stack everyone is familiar with.
5. There where many attempts to release a community UNIX continuation, including some of the BSD people, porting BSD with the AT&T parts replaced to the then new 386, which later became FreeBSD
6. That gets us to RMS and the FSF which started earlier, but instead of porting existing software, tried an ambitious re-write including centering around the mach microkernel. It turns out that such a project was far more ambitious then they anticipated, and went on for years with no real usable kernel
7. Later, about 7-8 years later, a grad student wants to run UNIX on his PC, but the best he gets is MINIX, which is a simplified version for teaching OS design. So he re-works the kernel to be more powerful.
8. He finds Minix to be quite limited, so he ports to GNU which had every intention on being a full strength production OS, but lacked a kernel. Lets be clear that GNU existed for several years at this point. Linux adopted the GPL which linux said "was the best thing to ever happen to linux", and targeted the GNU toolchain for builds. Yes, it is in fact a seperate project, but it targets GNU. Even if that changed now, it wouldn't erase the decades of history where linux exists because the GNU toolchain used to build it.
9. Operating system history is better documented than compiler and toolchain history. But from what I can gather, gcc was by far the best Free compiler/toolchain, and its single handedly responsible for giving Free operating systems and other software a leg up in the early years. FreeBSD used it after switching away from AT&T's pcc. Free software in modern form really wouldn't exist without GCC + toolchain

The beauty of Free software as a concept, is just because the original author had one idea about the software, you don't have to share their vision, you can fork it, and then add/remove features as you see fit.

So, RMS lost touch a very long time ago, but because of the IDEALs of Free software, its possible for everyone else to just pick it up and go in whatever direction is needed. It removes the single point of failure.

I can acknowledge his past contributions and admire his vision of software ideology, but he really needed to retire 10 or even 15 years ago, at least.

[coder]

Linux has made itself a slave of GNU, whether one likes it or not. One cannot wish it away, but one has to work in order to remove it. Only then will Linux be free from GNU.

Can anyone meaningfully speak to the amount of GNU software left in Android? I think it's not much.

[coder]

I am not nitpicking, I am pointing out your factual error. You spoke about Unix itself, not about Unix-like. And you did spoke using words that strongly imply all Unices were using FSF software - which is as untruth as can be.

Thanks for explaining your objection. I apologize for my sloppy wording, but I figured everyone here would understand what I meant. I don't mind the correction, as long as it's clear.

Let's revisit your ode of praise to FSF

Not a single true certified Unix has ever been built upon FSF-sponsored software. And when it comes to "Unix-like" software - it's just GNU/Linux and Hurd built purely on those.

I think you're being too parsimonious, here. Quite a lot of UNIX-relative (and other) operating systems have used FSF software. What about FreeBSD and its cousins, for instance? Haven't they long used gcc, GNU's libc, and certain other GNU tools?

With the way Linux has been ignoring POSIX standards you cannot call Linux "Unix-like" any more. It's just Linux nowadays, going it's own way.. Getting shit built on Linux (or for Linux) ported over to actual "Unix-like" OS'es has become major effort or may be flat out impossible when your porting effort happens to meet purely Linux-specific implementations. "Unix-likeness" assumes compatibility.

Fair point, although my statements were primarily retrospective, since I was replying to a post concerned with the creation of Linux.

While your corrections are welcome, it's not my goal to get into a big battle over semantics. However, if you feel there's an important distinction worth making, go ahead and make it. I'm sure I could learn a thing or two.