To be honest, I felt the same way you do about the BSD licenses until I saw what closing ZDaemon's source did to the competitive Doom community. That's really the "real-world" basis for my argument.
You have a very small sample for a real-world basis. Plus, I don't think it was the closing of the source that caused - well, whatever you claim it caused - so much as the extreme reaction to that process. After all, if it was the closing of the source, where's all the ZDaemon 1.06 forks? It seems to me, that the only thing the GPL would have done would have been to chase away the existing developers on the project (who were getting tired of the issues associated with having all the sources in the wild), leaving the project to die.
I think it's unacceptable that this is even possible, whether or not it requires people to be shitty or whatever is immaterial to me. In fact I think if a license allows openness to be defeated by shittiness, then the license itself is shitty.
So, since we're allowed to do something 'bad' with it....
It seems silly to me to try and enforce the morality of code-sharing (something which isn't detrimental to another person's health & well-being) by legal document. Like you said later in your post, programmers will write software, one way or another. Not all of them will share.
We could probably argue the merits about whether ZDaemon's community is worse/smaller than it was pre-1.07, or whether closing the source had anything to do with it
We can't. I won't argue this point with you, because quite frankly - regardless of what the truth is - I enjoy it more. Apparently I long to have more disagreements, hence me coming on doomworld to continue a war. Obviously, the only definition of a truce to said war is acknowledging that the other party simply won't bend, and call it at that.
But along the Linux/FreeBSD lines, FreeBSD 2.0 (the first "unencumbered" version) was released 11/29/1994. Linux 1.0 was released 3/14/1994, but Linux 1.0 sucked balls. It wasn't until the 2.0 series it really took off, and that wasn't released until 6/9/1996. So really, I would say that just from a timing standpoint, the lawsuit couldn't have had an effect on competition for developer effort between Linux & FreeBSD. If anything, FreeBSD had a head start.
But you can go further and say that the lawsuit was a result of AT&T's shitty licensing agreement with Berkeley. If they had licensed their code to Berkeley under the GPL, that lawsuit never would have happened. You might argue back, "but then AT&T would never have invested resources into UNIX in the first place, and we wouldn't have had UNIX and therefore no BSD, Linux, GNU, and so on." I would say that's not true at all, Andrew Tenenbaum developed MINIX, Linus Torvalds developed Linux, Syllable Desktop exists, Genode OS Framework exists, etc. Programmers program, whether or not they get paid for it.
http://en.wikipedia.org/wiki/History_of_Linux - See paragraph 3. Even wikipedia cites the lawsuit (apparently from AT&T, not Novell - sorry) as stymieing BSD, with sources. Yet somehow you overlook this in favor of your "the license caused it" argument. People never forget the bad things. A lawsuit against an OS is a bad thing. It takes a long time to recover from that.
Additionally, the GPL wouldn't protect against a lawsuit, much like it had not protected against SCO's lawsuit. The shit still hits the fan.
You'd have better luck attributing Linux's success on that it written from the ground-up as an open-source OS. Which, after the debacle with AT&T and BSD, was appealing to many people.
Also see, chronology.
Really though, I never would have brought AT&T's work on UNIX into the mix, but since you did....
Granted, we might still have Tenenbaum, Torvalds, etc, develop OSs, but they had something to mimic. In the world we live in, this was UNIX.
Could you imagine Linux if it were based off MS-DOS?