Tell HN: I built a push service for HTTP APIs and want beta testers
What makes Fanout special is its versatility. As an API/protocol guy, it occurred to me that existing cloud push solutions aren't well-suited for building APIs. For example, if you want to create and offer a real-time API like Twitter, Yammer, Digg, etc, then you pretty much have to set up your own infrastructure. Also, migrating to or between any of the existing solutions means having to update all client code to match. That's a very heavy transition cost, especially if you have any third-party consumers.
I figured there were probably other developers out there considering the cloud for push but having similar concerns about API feasibility and transition cost, so I've spent the last few months developing a solution that I now feel is ready for some beta testers.
Why should you be interested in Fanout?
Fanout allows for customization of the client/server interactions. This makes it possible to:
* Develop real-time APIs using HTTP streaming or HTTP long-polling. * Migrate APIs to the cloud without breaking existing consumers. * Keep using your existing client libraries. * Avoid proprietary protocols and associated vendor lock-in.
Of course, if you don't want to futz with designing protocol, Fanout additionally has its own JSON-based protocol and complementary JavaScript library that'll let you get off the ground quickly and is on par with existing solutions.
I'm currently letting a few developers access the beta, so if you're interested, please visit http://fanout.io/ and submit your email address. I look forward to hearing from you.