Offline capable peer to peer replication was the main reason I chose CouchDB - we needed something that would realistically run on clients, even mobile devices. NoSQL we mainly chose because we needed schema-less data (the whole system relies on ad-hoc design updates). It's basically an information system IDE with rapid application development.
I immediately wondered why Diaspora didn't try CouchDB, since replication seems to be one of the key features they were after.
In Diaspora as it exists now, replication - really, federation - is between pods. There's a protocol for transferring data between pods that is deliberately database-agnostic:
https://wiki.diasporafoundation.org/Federation_protocol_over...
So CouchDB's replication doesn't really help.
If the day comes that any single pod is big enough to need replication between clustered machines within it, then CouchDB should certainly be a contender for storing its data.
Plus change feeds and peer to peer replication are first class citizens in the CouchDB. Once you start having large number of clients needing realtime updates, having to periodically poll for data updates can get very expensive.