I think the key points here - that it won't get critical mass and that people aren't going to host their own are both demonstrably solvable. Hosting: there will

I think the key points here - that it won't get critical mass and that people aren't going to host their own are both demonstrably solvable. Hosting: there will be an open market for people to set up hosted seeds of every level and flavour and Flickr showed that people will pay for a social service. There's nothing stopping someone offering free seeds with advertising. And nothing stopping someone selling pre-configured VMs with gobs of storage. The network effect is more interesting. Diaspora is designed to hoover content from other networks, I can see how I could use it as much as a distribution and management system as a network in and of itself. It's a meta-network, an identity framework on which services can be built. That mechanism will hopefully provide base utility from day one, as well as an open platform on which people can easily build new and innovative services without those ideas being dependent on either a 3rd party's infrastructure or building one themselves. I'm not convinced it's going to succeed but I'm hopeful.