I wonder if the professional burnout has a similar "contagion" aspect. A bunch of us are getting burned out. People who are borderline burned out, see the similar behaviors, mindsets, etc, and cross over the line. And since we're all super-connected and many of us are super-vocal, the effect is magnified.
I see a similar possibility here. Startup scene got big around early last year say, now it's winding down, people are realizing, once again, that it's the hot shot big hits that get the showtime, not the 90%+ who have a failure and ended up wasting a year or more of their prime living on ramen and "bootstrapping". Take a look at the app store for how that works out.
Edited to add: I have no idea why this is getting downvoted. If you disagree with my assertion then by all means leave a comment. Hitting the down arrow and wandering off does us both a disservice.
Its easy to suggest and might come easy to some, but it's way out of reach for most young developers trying to get on the career track. Making a notable contribution to an impactful project requires a lot of time and understanding. Even if a project could use the help, jumping in head first like that is quite daunting.
I understand what you're saying when you say it's daunting, but this doesn't have to be a big deal. I'm not advocating someone run right out and start cutting kernel patches straight out of school. The Open Source space is enormous and it should be possible to pick a project that's in line with your abilities. Example: there are at least three CMS projects I know of that are well documented and very accessible to new developers looking to help out.
Degree qualified, startup fresh out of university with no post-grad work experience, failed startup, making ends meet with freelancing with little hope of rejoining a career track.
Now don't get me wrong, were all beating the Fuck Big Corp drum here, but you can't beat big corp for a relatively stable income and ratable work experience.
That's why my advice to kids is: Get that degree, get some work experience (1 year say), THEN work on that startup. That way, you have a fallback option.