Marketing is a huge part of the success of a new paradigm. App.net was "marketed" with a practically blank home page ("one login many applications" - which applications?), and a nondescript niche Twitter clone on one of its subdomains.

It had no chance at all. Where was the evangelism? Where were the presentations, demos, and the other useful apps built on this platform?

I'm a developer and I did hear about App.net when the buzz was at its peak. I remember going to their home page and wondering "wtf is this?", then checking alpha.app.net on a couple of occasions, finding nothing of interest and that's it.

That wonderful arrogance that the world will care about your project because you're somehow "right" has killed many tech efforts and will kill a lot to come yet.

This. So much of this.

I see real value in using App.net as an identity provider so that I'm not relying on something shaky like Twitter or Facebook but I can still get the benefits that come with an external provider (lower signup friction, 2-factor auth for users that want it, less overall damage if site is hacked). But (a) their marketing is so weird that I don't really know that I can do that, and (b) their onboarding for end users sucks, so I'm losing people telling them to sign up through them.

Those are really good points.

I think part of the problem was the initial strategy that required all accounts to be paid limited the userbase to the point that it wasn't worth experimenting with.

I don't know how you get around that challenge of building a mass userbase, keeping the framework open and yet still being sustainable. But that's a separate challenge from what you described.