consider SLEEP

[max-mapper]

http://www.dataprotocols.org/en/latest/syncing.html#sleep

its essentially a generic version of couchdbs multi-master streaming replication protocol without all the complicated MVCC stuff.

[max-mapper]

more detail: from a data storage perspective this simply means you would be storing a "transaction log" that is exposed over HTTP as JSON so that it can be used to quickly do queries like this:

get all posts from max@maxogden.com since {{myMostRecentPostID}}

[max-mapper]

one more thing: having a underlying data storage model that supports this makes it easy to build push and pull webservices on top. doesn't necessarily have to be SLEEP but I would highly recommend researching this

[titanous]

This isn't going to be part of the initial protocol, but it's an interesting approach.