Thought it an interesting discussion, so I will also contribute my 2 cents worth of open-source speculation;
Open source is thriving. Open source is a funny hobby, that allows individuals to gain experience, improve their social networks and have fun while doing it.
Open source does not depend on the succes of Linux or of any other "open" platform, because the two things are fundamentally different things.
Linux and all its related projects inspire lots of aspiring young programmers to get into the open source world. Linux is a crusade, a quest, a holy grail, that provide a focus- point for a lot of programming activity.
Linux makes it possible for all the distributed decision makers to make autonomous decisions that have as a focus- point what efforts will be to the good of the Linux community. However, due to the scattered activity and the absence of any real authority in the community, most or all decisions are made with a lack of information that prevents any decision from being optimal.
Of course there are several kinds of levels of decisions that are continually made. Decisions are made for a specific project, regarding how feature X should be implemented, whether feature X should be implemented at all, and decisions about who writes the documentation, or maintains the website or provides user support etc etc. I think this level of decision making is working flawlessly. All well-functioning open source projects have this level of decision making taken care of, that's what the project is all about.
However, the project doesn't contemplate the fact that 5 other projects exist that are spending time doing the exact same thing. All 6 projects are convinced that they're creating the best solution to the "problem" that this class of projects aim for, and none or very few of the projects cooperate amongst themselves. This is a new level of decision making. An absent level of decision making, because no decisions are made. The decision is left to evolution; the most stubborn project wins (or best if you're an optimist). This kind of decision making incurs a horrendous amount of replicated work, that is useless if any long-time success of the "open source crusade" is intentioned.
There is no way to make people give up this mode of operation, because the open source world is fundamentally an ego-boosting business. It's just not as cool to work on someone else's project. You can do the stuff he's doing better, and you're going to show it to the World. Linux is about ego boosting, and many developers work for projects related to the primary open source platforms to be a part of, and thereby gain part of the prestige that the platforms provide. Ego boost is measured in number of users, but most developers are kidding themselves if they say that they are working on their project for their users. We're only human.
At the same time, because the whole thing is a voluntary "for fun" hobby, no one but the most visionary/obstinate is interested in working on all the dull bits and pieces that are essential for the mass of projects to be able to function together.
This is why the open source community has got to come to terms with the fact that open source software projects like Linux will always be a niche, because it's made for the good of the developers, not for the good of the users. I think thats only fair, because it's the developers who have spent countless hours of their life working on cool software.
To conclude; open source as a creative hobby is far from dead, but a user-focused open source platform is a dead end.