My opinion on (3): I think you're overlooking the importance of sheer dumb luck. For entertainment products like Twitter, you have to somehow hook a core, influ

My opinion on (3): I think you're overlooking the importance of sheer dumb luck. For entertainment products like Twitter, you have to somehow hook a core, influential audience. Then it takes marketing talent and more luck to expand into a mainstream audience. I don't think you can plan it, even with deep pockets (see Wave, Orkut, Buzz). I don't use it nor understand the appeal, but I think of Twitter abstractly as an Internet-scale multicast message queue. You can blast any type of message out to a set of subscribers very quickly. That is a very interesting technology with potential value. It just happens to be used primarily as a chat application today. For example, you could trivially implement rssCloud on top of it.