I'm not sure if it's about competitiveness. Communicating is a lot slower and more frustrating than hacking. In the old days, forking came at a cost; keeping up with upstream changes to a project that was actively developed was a huge pain, so it was easier to try to get your changes accepted upstream, even if you had to do some work to convince them to take it. With git, it's only inconvenient when upstream changes something you changed too, and that probably doesn't happen often, so the easier thing to do is just maintain your fork.
This is also something I thought about. It's really easy with git to just fork a project and do a pull request upstream if you wanted to contribute something to the original project. I get the feeling too many people would bypass that latter part and just hope their changes catch on to some as-of-yet unfulfilled niche of users of the original to keep their fork maintained - needlessly fragmenting the program and its userbase as a whole so they can get their 15 minutes of fame and satisfied delusions of grandeur. Maybe I'm just too cynical...
I don't think you're being cynical, but I do think everyone (including the author of the article) might be looking at the past through rose colored glasses. People are not always kind on mailing lists, and that can turn a lot of people off from software. Theo De Raadt forked OpenBSD from NetBSD specifically because of personality clashes iirc. You can also take a look at Eric Raymond's "How to ask questions the smart way" http://www.catb.org/esr/faqs/smart-questions.html and find little bits of passive aggressiveness and misanthropy. I submit that open source has always been a mix of competition and cooperation in almost equal parts. That the only real change here has been that the number of active users in the open source realm has grown. Also I have issues with this: I don't care if someone forks software as long as the license permits it. Why would I? If you don't want someone to be able to do that, relicense it! There isn't much fragmentation when a fork is created, and usually unmaintained software will be given up on in the short future. needlessly fragmenting the program and its userbase as a whole so they can get their 15 minutes of fame and satisfied delusions of grandeur
Yeah after thinking about it I make it sound like the whole premise of the hundreds of *nix flavors weren't all conceived under basically identical circumstances and that's by no means a bad thing, so I suppose it can certainly go both ways. Speaking of asshole programmers, I found out the guy that wrote the Learn Code the Hard Way series (which I actually like a lot) is a huge dick, which is disappointing. I dunno what it is about this profession that makes people act like that. I was going to link to a bunch of examples of his Good Behavior from Twitter but it seems he's made that private now, with separate "news" and "PR" accounts. Ha ha! That's a new development.