Personally, I wouldn't have put the people calling it into question and then the fact that it is unproven legally at the end. It might be best for the flow, but to conclude "All of this might be meaningless, as it hasn't been proven to work!" is slightly strange to me.

I didn't learn anything new from reading this, but there are those that might, as was stated by the submitter. In my opinion, telling the uninitiated simply that the GPL grants the right to use the code however you like, but if you decide to distribute it you have to agree to the contract stating that you'll give the source as well. If they have other questions, send them to the GNU.org philosophy page [gnu.org]

Creating a presentation on GPL Enforceability.... in Powerpoint.

So...

GNU make available the main elements of an OS for free (or some useful helper utilities, make whatever distinction you want, it doesn't really matter for the purpose of my argument) and they request that you include them in the name of the OS.

Now, you have several choices.

Ethically "pure" ones:

* Gratefully use the software they've provided for free and agree to their modest request

* Decide that they're on a power trip and decline to use their software

Ethically slightly grey:

* Use their software, call it what you want anyway (probably most, er, "Linux" users)

Ungrateful:

* Use their software (given away for free!), call it what you want, and flame them in public for having the audacity to ask for a name modification

Ungrateful and annoying:

* As above, but also berate others for agreeing to their request (this is you, presumably!)

There is one other option, that of troll: don't use their software and flame them anyway - in which case why would you care what other people call it?

The whole paper appears to be centred around the idea that the GPL is a contract. Most people seem to think the GPL is actually a licence, not a contract - this kind of makes the whole paper useless, in my opinion, and is what led to the confused conclusion.

http://lwn.net/Articles/61292/ [lwn.net] is a useful discussion of the difference, once you accept it as a licence the whole discussion of "enforcability" kind of goes out of the window.

It's too simplistic to say that a court case will either confirm or deny the validity of the GPL. As always, the court will focus on the minutiae of the situation. It's very rare for a court to say "this contract is not legally enforceable in any situation" (unless the contract is for an illegal purpose or is so vague as to be meaningless). More likely, if a court rules against enforcement of the GPL in a case, it will cite some factual item in support. E.g. that the recipient of the software did not have proper notice of the license - quite possible if a company hires an outside programmer to do some work on some GPL'd code. Sure the GPL would bind the programmer, but it might not bind the company if they'd never looked at the source code or even realised the software was GPL.

It's no use (as is claimed in the paper) to say that the company had no right to distribute the software except under the GPL. That doesn't mean it's accepted the GPL by the act of distributing the software, if it has no knowledge of the GPL. That argument is like saying if I steal something, I must have agreed to buy it (since otherwise it would not be mine) and am in breach of contract for not paying for it. While someone distributing software without permission is in breach of copyright, it's different to distributing software in breach of the GPL. In the former, the copyright owner can stop distribution/use of the code, but he can't compel the distributer to release the source code, nor to hand over copyright in the modifications made to the code.

There are other reasons [uklinux.net] to stop prepending things to the OS name, by the way...

My take on this : Stallman (who recently did a speech in my area, and didn't forget to tell us that ``GNU is the operating system and Linux is its kernel'' amongst other questionable things, like Open Source is not about freedom. He must have overlooked the Open Source Definition [opensource.org]...) has seen the Linux OS gain lots of attention, and he's disgruntled people talk less about the GNU project. Still, he's wrong in thinking this causes problems for GNU or the FSF : quite to the contrary, the number of people that have started writing (or even hearing about) free software because of Linux has greatly helped the GNU project approach the goal of having a computing environment made only of free software. He should be grateful of this. Instead, he's obsessed with safeguarding a message he thinks is ``suppressed'' and churns out non-free licenses (like the GFDL) for this purpose (I'd be insterested in seeing only one occurrence of a distributor removing philosophical texts from the GNU documentations. I don't think this ever happened). That's sad, and not very positive, IMHO.