The news here is that, in addition to the huge market for traditional DBs, postgres is going to compete in a serious way on MongoDB's home turf. As that becomes more apparent, it will validate postgres's flexability/adaptability and cast doubt over special-purpose database systems and NoSQL.
MongoDB still has a story around clustering, of course. But that story is somewhat mixed (as all DB clustering stories are); and postgres is not standing still on that front, either.
(Disclaimer: I'm a member of the postgres community.)
Well, mostly. Except for when it doesn't, especially when it comes to partitions.
If I could suggest, please look at ease of configuring replication in MongoDB. If you make it that easy, even if it's just for the JSON stores, I know that maybe you can't make it easy for relational tables, I will move my product to PostgreSQL in a heartbeat.
I have used and promoted PostgreSQL throughout my career, but I'm currently stuck with having to deliver HA in a product for redistribution to customers and Slony can't cut it for our use cases, it's too complicated. I think the key is that we are a vendor and we need to provide a database with our products, so it's not me that's responsible for configuring HA, it's our users. And replication is what I'm concerned about 90% of the time, not necessarily horizontally scaling out (i.e. we're not using sharding in 90% of our cases, we just need HA).
Have you looked at streaming replication in postgres?
I'm not sure what level of expertise your customers are at, so this might not work for you. But it seems like a better fit than slony at least.