Twitter Futures
tbray.org
tbray.org
We are at a point where small teams can get a lot done (see Gas app). This has all sorts of ramifications, the major one is staying agile and nimble for far longer. I mean, Clayton Christensen wrote books on the topic.
In an age where you can outsource expertise to BigCloud or NewCloud (sure, the costs are higher than if you run your own DCs and in-house expertise matters), you can avoid building large orgs, which are also a drag on the business and come with non-trivial amount of latent cost. Besides, tech is a knowledge-based field. Engs who want to improve have the opportunity do so (cue Ousterhout's a bit of slope makes for a lot of y-intercept).
The industry is just coming off of a major layoff season. An estimated 150k very capable engs now out of job. I hope some of these engs start companies (think: Jan Koum), but many others simply wouldn't have that luxury. While the best (or rather, popular) amongst us may have a pick of their choice (incl starting companies), the rest of us, as the saying goes... we cannot be choosers.
Twitter may yet fall apart like MySpace, Digg, Path, Yahoo et al, but I do think it stands a better chance than its predecessors did despite Elon.
I'm not sure we've only now arrived at that point. Instagram allegedly only had 12 employees when it was acquired by Facebook, for instance. The bloated amount of employees in many tech companies seems to many other reasons too.