This is a category error when it comes to copyleft licenses like the GPL. It has nothing to say about usage.
"If, on the other hand, the licensor and licensee both have some obligations (in OSS, this is usually when you modify or redistribute the source or compiled product), then it's basically a contract, no matter what RMS claims."
No it's not. There are no pre-agreed terms, penalties for violation, expected compensation for services provided or anything like that. GPLed software is copyrighted. Copyright law says you have no rights to copy it or make derivative works of it whatsoever. The license permits you to do so.
"Why would you trust what he says about contract law, a field where he has no professional training whatsoever?"
Because, surprise surprise, he has advice from people who ARE trained in the law.
The GPLv2 text: " The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). "
Of course this sentence contradicts the previous sentence in the text, which claims that normal usage "is not covered by this License". I presume you'd argue to support your claim, but seriously, this is bad drafting.
> No it's not. There are no pre-agreed terms, penalties for violation, expected compensation for services provided or anything like that.
"penalties for violation", "expected compensation" are not necessary requirements for formation of a contract. The pre-agreed terms are clearly stated in the license text, or at least as clear as far as they don't contradict each other. By the way, "pre-agreed terms" are not necessary for the formation of a contract either.
> Because, surprise surprise, he has advice from people who ARE trained in the law.
Are you trained in the law? Because if you think the Internet should pay regards to somebody trained in the law (even though they may not have learned the law properly) as opposed to somebody who hasn't, then I don't see why you think you have a standing to speak as though you're an authoritative source on the matter.
Quote: "(of course, free software licenses would still fulfil the important function of contractually requiring the publication of modified source code)"
Even though as many others have pointed out, OSS licenses can be contracts, I'm actually not sure this sentence is correct.
When somebody uses the source code in compliance with the license terms, a contract might be formed to allow both parties to enjoy rights. However, if one party never complied with those terms and breaches them (eg. distributing source without retaining copyright notices), then arguably no contract was ever formed, and the act is a simple matter of copyright violation and not a "breach of contract".
Hope I'm not splitting hairs.
Disclaimer: learned English common law a bit, not a lawyer.
> Eben Moglen is professor of law and legal history at Columbia University Law School.
In Moglen's article about enforcement, I think he's right that where there's a breach of GPL there is no contract. In fact that's what I said also in my follow up reply.
If, on the other hand, the licensor and licensee both have some obligations (in OSS, this is usually when you modify or redistribute the source or compiled product), then it's basically a contract, no matter what RMS claims.
I mean, with all due respect to the guy, he makes controversial claims even in the field of software engineering (and also free software evangelism), his supposed professional field. Why would you trust what he says about contract law, a field where he has no professional training whatsoever?
(That said, GPLv2 is still an ingenious work for many reasons, albeit lawyers probably won't draft it that way)