If the past week has been any indication, he's going to build a small army of web API related infrastructure projects: http://www.scriptlets.org/ http://www.postbin.org/ http://github.com/progrium/hookah/
I'm actually very surprised at the number of people happy with DevjaVu as a service since after the first year or so I spent very little time working on it. Except when something went wrong and then I spent a lot of time to plug the hole (and not properly fix the problem like I should have--that would require even more time).
The whole hosted project/version control business is tough. For starters the market is crowded. Additionally, a lot of people do want it all for free. To top it off, it's uber resource (space + bandwidth) intensive.
There are some really good folks out there who provide a great service and devjavu's one of them. Its sad to see it shutting down, and I can empathize with Jeff on his decision to terminate the service.
I think we did okay for what was ultimately a one-man show. But I'd much rather put the kind of investment required for running a business into something more innovative.
Do keep all of us posted on what you're doing next!
You guys could always get a slice at linode and save even more.
Since EY's customers are mostly Ruby developers, it must have been hard competing with GitHub + Lighthouse:
http://www.reuters.com/article/pressRelease/idUS145946+29-Ap...
