” As an example, GCC and other GNU tools are included in the OpenBSD tool chain. However, it is quite possible to distribute a system for many applications without a tool chain, or the distributor can choose to include a tool chain as an optional bundle which conforms to the GPL terms.”

Now, what fascinates me here is that the BSD folks “get it” — that the “GNU Operating System” never happened, but that the VARIOUS PARTS of what MIGHT HAVE EVENTUALLY BEEN the “GNU operating system” DID. The claim that “most of what we actually use is GNU-based” is outright false, when one considers that it is all “based” on the kernel (in that everything else “runs on top of” that.)

So, systems which use the “Linux Kernel” are by definition Linux-based systems. Stallman is the one who keeps copping the attitude over the fact that GNU doesn’t get mentioned every second breath.

Stallman should be applauded for what he did in 1984. For what he’s doing NOWADAYS, he should be regarded as a laughing-stock, and a hasbeen, until (and unless) he — and his organization — can come to grips with the fact that the “movement” is WAY bigger than one “spokesman”.

(Personally, I find the whole topic of “software licenses” to be somewhat pointless, seeing as — as with movies and mp3’s — people will do whatever the hell they WANT to do with it, regardless of whether they’re “legally permitted” to do so, or not.

Isn’t it a little ironic to anybody that the same populace who agressively advocate p2p apps ALSO tend to be very into the idea of “software licenses?” There’s some kind of doublethink going on here — an unwillingness to go “hey, wait a minute — people don’t give a shit whether they’re “allowed” to copy music/DVD movies/take a camcorder into a movie-theater/etc. Why do we think they’ll give a shit about SOFTWARE licenses?”

Some kind of doublethink here, as I said, but hey….whatever.

The GPL is, at heart, every bit as “restrictive” as all other “User-Agreements”, in that for it to be legally-useful at all, it is dependent on the present copyright structure (all of that “copyleft” blather notwithstanding. This is what nobody — least of all Stallman — seem to want to admit.

Relevant GPL quote: “Developers that use the GNU GPL protect your rights with two steps: (1) assert copyright on the software, and (2) offer you this License giving you legal permission to copy, distribute and/or modify it.”

So from this we can gather two things:
1. The GPL’s ability to “protect your rights” is completely dependent on the existing copyright structure (in that it is copyright which empowers the “owner” to place restrictions on what is done with the copyrighted material.
And,
2. The GPL is totally groovy because it imposes such restrictions “for a good cause.”

It all makes sense now: the GPL is about “forcing you to be free” (in that you are permitted to ‘be free’ by complying with it — doing what it PERMITS you to do — so why not centralize everything under the rubric of ONE organization, so that such “forced freedom” can be done more efficiently, and make things “move forward”?

Yeah, baby — that’s a bandwagon I want to get on.

The GPL is, at heart, every bit as “restrictive” as all other “User-Agreements”, in that for it to be legally-useful at all, it is dependent on the present copyright structure (all of that “copyleft” blather notwithstanding. This is what nobody — least of all Stallman — seem to want to admit.

False and deliberately idiotic:
They should have control because now they have to get a DEMOCRATIC VOTE BEFORE THEY TAKE ACTION? This is the “basic truth” that all defenders of fascist/totalitarian hierarchies depend on — the fact that those ‘in control’ can basically ram things through above and beyond anyone elses wishes. Fine — you want dictatorial control by a centralized monopoly, then Microsoft really ISN’T the enemy, right?
Further (last I checked, which was about a month ago) several distros are already using “unofficial” kernels — nonstandard ones with their own enhancements/hacks whatever. That’s another advantage of decentralization — they can do that.

Yeah, things really “move forward” in a centralized development model — witness the fact that Micro$ucks eagerly-awaited ‘longhorn’ release was delayed for THREE YEARS, and then still sucked.
So no, coercive centralization is a really bad idea.

“The kernel is actually only a small portion of the operating system, which performs the basic functions. The real guts of a GNU / Linux OS, is of course, the gnu compiler collection (aka gcc), glibc, gnu coreutils. Then the other hundred utilities that provide basic services, such as web browsing (elinks, links, lynx), package managers, emacs, nano, vim. Then there is the X window system, which is invaluable. All these things make up an operating system. The kernel is really a very small part of it. Most of the system we use is actually GNU based, that’s why debian chose the moniker GNU / Linux.”

Uh-huh. So how far do you take this “credit where credit is due” thing? “I’m running debian GNU/Linux/X/KDE/(ad nauseum for every software package and project used in the thing? Bullshit.
The GNU “tools” can be used on top of anything from Minix to BSD — doesn’t mean that they themselves constitute an “operating system” in and of themselves. Further, do you hear ANY other “spokesman” from any other project bleating about how THEY don’t get mentioned in the name of distributions? No.
Stallman is the ONLY person making pissy little plugs for his “Not-quite-ready-and-never-will-be” implimentation. He’s the one who can’t get over the fact that the “GNU project” — big and vibrant and damn useful as the stuff is creates is, don’t get me wrong — does not, in and of itself, constitute a complete OS developed “in-house”.

Let’s see ‘em do it: let’s see ‘em create their OWN (working) kernel — with actual support for actual devices that people will actually use. Let’s see ‘em do their OWN variant of the X-windows system (like xfree86 and x.org already do.) Let’s SEE ‘em pony up, and run in-house versions of EVERY COMPONENT and EVERY userland program. THEN Stallman has a leg to stand on with his pissy little “GNU/Linux” fetish.

If you’re using Gcc (one of the GNU tools) on Minix, then does that mean you’re using “GNU/Minix?” What about those who use GNU tools on BSD systems (there’s quite a bit of that going on as well.) Does that immediately mandate that the system is “GNU/BSD?”
Untill — and unless — Stallman et. al. can get over the fact that the FLOSS (”Free-Libre-Opensource Software”) movement is WAY bigger than any particular project, foundation, or “spokesman”, it’ll continue to resemble a religious cult more than a serious movement for genuine change, and for anybody to even PROPOSE the idea that the FSF/Stallman/whoever be granted the power to “FORCE” change through in the interst of “making things move forward” maybe SHOULD raise some eyebrows, or is it just me?

Hey, maybe I’m wrong: Hitler DID make the trains run on time. (grin.)

That said, Richard Stallman is no longer the only or even the most important spokesperson for the Global Free Software movement. If people want to learn about the importance of this movement, they may want to listen to (watch/etc) one or two recent lectures by Eben Moglen. As an active member of the Free Software movement I find that Eben’s words and way of thinking match me more than other spokespersons such as RMS.

The Global Software Industry in Transformation: After GPLv3
http://www.digital-copyright.ca/node/4055

The GPL is by far the most popular software license used for any software (by projects, by lines of code, and in the future by number of users of code). Eben Moglen puts this work in the proper context of possibly being as important in our current world as the decisions about how science would work that brought western countries out of the dark ages. He discusses one of the important moral questions of our time — a question that should be adequately thought about by anyone reading this P2P proponent news site.

Most people don’t understand how copyright licensing works. I’ve given many presentations over the years to explain how large software projects work that have components contributed from many places.

Lets take the GPLv2 only question and Linux kernel as an example. This is a single component of a Linux operating system (It is usually 1 or two package out of hundreds for an average “Linux” installation). This single component has literally hundreds upon hundreds of contributors to it. (A typical Linux distribution has contributors numbering in the tens or hundreds of thousands of individual authors. Far more brain-power than any corporation could ever hire).

Some of the Linux kernel contributors have no version of the GPL listed, which means that these components can be understood as licensed under any version (at the discretion of the recipient). Some of the contributors use the default “either version 2 of the License, or (at your option) any later version”, and others (like Linus) removed that clause and contributed under GPLv2 only.

There is additionally code under various GPLv2 compatible licenses, such as code shared with the BSD kernels. A full list of licenses and whether they are GPLv2 compatible is at http://www.fsf.org/licensing/licenses/ .

Components can be combined into a larger project as long as the licenses of all the components are able to be simultaneously honored. The project as a whole is then, for the sake of simplicity, said to be under the license which is the “common denominator” of all the individual licenses used for all the components.

One informal study of the kernel suggested that contributions that are GPL version 2 comprise far less than half of the Linux kernel. This means that the majority of the components within this single program are able to be extracted from the Linux kernel and used in GPLv3 projects.

It will be quite possible that a growing amount of code in the Linux kernel will be re-licensed by their authors to be GPLv3. There is nothing stopping individual kernel developers from contributing their own code in any license that is GPLv2 compatible, which includes the “GPLv2 and higher versions” clause.

The Linux kernel as a whole will still be said to be under the “lowest common denominator” license which is GPLv2 given there are some contributors that don’t want to upgrade, but that doesn’t mean that specific components can’t be shared.

Clear as mud? For most people this won’t make sense, which is why Jon should be excused for being in the vast majority that don’t understand how this all works.

As he points out, when I was editing his blog article, I included the wrong “they,” ie:

“One notable exception is the Linux kernel that, because they removed this option, doesn’t realistically have the option to change licenses
(without trying to get the approval of thousands of contributors).”

As opposed to ….

“One notable exception is the Linux kernel which, because the FSF removed it, doesn’t realistically allow changes to licenses (without
trying to get the approval of thousands of contributors).”

The ‘they’ was the Linux kernel community, not the FSF.

And, “It is also the Linux kernel community that would be wanting to change the license they use for their project, not make “changes to licenses,” he says.

So my sincere apologies to Russell, and to anyone else who may have been discombobulated by my post.

By way of explanation, but not as an excuse, I’m not a technical expert, and I’ve only just started familiarising myself with Linux. I’m experimenting with Kubunto, but I still have a very long way to go, as you can obviously see.

I do my best to present news on open source developments and I’m only too glad when my mistakes are corrected.

Cheers! And thanks …’
Jon

It’s the FSF, not the EFF. And the reason why the FSF should have primary control over it is because, as it stands at the moment, to switch the licence of the kernel, they would have to get a yes vote from every person who wrote the code. Which is crazy. It’s completely unmanageable and things can never move forward.

The kernel is actually only a small portion of the operating system, which performs the basic functions. The real guts of a GNU / Linux OS, is of course, the gnu compiler collection (aka gcc), glibc, gnu coreutils. Then the other hundred utilities that provide basic services, such as web browsing (elinks, links, lynx), package managers, emacs, nano, vim. Then there is the X window system, which is invaluable. All these things make up an operating system. The kernel is really a very small part of it. Most of the system we use is actually GNU based, that’s why debian chose the moniker GNU / Linux.

Torvalds took the easiest solution, a monolithic kernel. This design has been in use since the first kernels appeared. The hurd is trying to revolutionize this, which i laud. Of course it would be nice if the hurd had full hardware support.

The GNU system (also called GNU/Hurd) is completely self-contained (you can compile all parts of it using GNU itself). You can run several instances of the Hurd in parallel, and debug even critical servers in one Hurd instance with gdb running on another Hurd instance. You can run the X window system, applications that use it, and advanced server applications like the Apache webserver.

On the negative side, the support for character devices (like sound cards) and other hardware is mostly missing. Although the POSIX interface is provided, some additional interfaces like POSIX shared memory or semaphores are still under development. ”

It’s been in development for well over 10 years and still doesn’t even support character devices. No sound, no dvb. So i don’t think it’s appropriate to say it’s complete. It has only basic driver support.

I actually like the concept of the hurd, where each server is isolated. Stability would be enhanced greatly by this model, as apposed to the monolithic linux kernel. But unfortunately, it’s no where near usable on a daily basis in it’s current state.

WHY should even MORE software be under the “control” of the EFF? Stallman is the one who “loves to exert his control” — why else the pissy tirades about “GNU/Linux” — which Debian gladly obey. STALLMAN is the one who can’t get it through his head that the “GNU Operating system” never actually happened, and that he and his crew were bested by Torvalds and totally beaten out in development of a working kernel?

Oh yeah, “GNU” is a “full operating system” — even though the wikipedia article on it admits that it uses numerous so-called “Third-party” projects (Like the X graphics system, for one.)

Stallman et. al. WERE useful in that they did create the GPL, but there is NO BASIS for this rabid Stallman-worship that’s been increasing of late. I detest the “Free Software song” — it sucks. Stallman spends too much time prancing around on stage and bleating about how it’s so “unfortunate” that all us pathetic infidels dare use the term “Linux” for the whole distro. Fine — you want “A kernel that EFF controls?” They have an (approximately) working version of the HURD — although there’s almost nobody using it, primarily because their “ambition design” is barely even functional.

The real strength of the FLOSS scene (”Free-Libre-Opensource), is that the type of “control” the Stallmanites are advocating is very literally IMPOSSIBLE for any organization, project, or even individual to wield.

This whole mentality of Stallman-As-Guru needs to be bitchslapped: the FLOSS scene is MUCH bigger than the EFF, or Stallman’s fruity little pet peeves.

(Need any more proof that he’s a has-been? He insists on numbering the “four freedoms” as 0 through 3. That’s not clever. It’s not funny, and it makes the whole scene look stupid. Eric Raymond was right to form the Open-source initiative (Opensource.org) as a perventative counterweight to “Stallman’s holy Mother-church.”

When did we all start thinking that aggressive “centralization” was a good idea? I thought that was the PROBLEM with Microsoft — that they had so-called “monopoly control”, and could use that ‘power’ to bludgeon everybody else into doing what they said?

But mysteriously, when “Rich and the boyz” over at EFF start trying that shit, it’s supposedly all good — he wrote EMACS and the “Hacker” song……pffff.

http://www.gnu.org/software/hurd/hurd.html

It was not the “FSF” that removed the ‘either version 2 of the License, or (at your option) any later version’ clause from GPLv2 used in the Linux Kernel, but a few Linux kernel developers such as Linus.

As Eben Moglen said in a recent speech “The Global Software Industry in Transformation: After GPLv3″ http://www.digital-copyright.ca/node/4055 , that “later version” clause in GPLv2 came down to a trust of the Free Software Foundation. Because of political differences with the FSF, Linus didn’t have that trust.

The issue at this point is that the Linux kernel community doesn’t have any organization, FSF or otherwise, who are able to relicense the kernel. The only way to move to GPLv3 would be to get individual permission from every author who contributed code and who have removed the “and future versions” clause. This is extremely hard to do, and would require rewriting any code where the author was unlocateable, and thus is unlikely to happen.

One of the many features of the enhanced GPLv3 is the suggestion that people can delegate the question about future versions to a third party to handle this in trust. Had this option been more clear back in 1991 this is likely what Linus and other developers would have done, and upgrading the license without trusting the FSF would have been possible.

It is the practical issue of not being able to relicense the Linux kernel that Bottomley was speaking of. He did co-author an article back in 2006 that was critical of an old draft of GPLv3, but that was an old draft and not the released GPLv3. More recently he has offered a “Tip of the hat to the FSF,” adding that “they listened to our [Linux developers] complaints about the license.”
http://www.linux-watch.com/news/NS8688613635.html

The reality: While the Linux kernel will stay GPLv2 for both practical and political reasons, subsystems can be GPLv3 compatible and thus be able to be shared in common with other future kernel projects. As the importance of the upgrade of the GPL becomes to the larger community, it may be that the importance of the GPLv2 licensed Linux kernel will be diminished considerably as people move to replacements.