couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paul Davis <>
Subject Re: CouchOne + Membase = Couchbase
Date Tue, 08 Feb 2011 18:26:57 GMT
> A few months ago, I read about BigCouch:
>    BigCouch = (CouchDB + Amazon Dynamo clustering theory)
> That sounds neat, and I was getting the impression this was done in Erlang in such a
way it could become part of core and was excited for it.

It is and there's still a general agreement that it will be. The
issues holding it back are mostly technical in that we need to
refactor our source tree which is being held up by things like waiting
for the new replicator to land.

> Now I'm trying to figure out what Couchbase is, and my reading indicates:
>    Couchbase = (Memcached + magic) + (CouchDB + ponies)
> Will these new dependencies make CouchDB harder to compile and use for personal deployments?
How does merging in an in-memory cache provide the clustered resiliency I was hoping would
be possible by using BigCouch? I'd never heard of Membase before last night, so I guess I'm
just feeling a bit like a nervous IT guy hearing the platform he relies on is about to change
in ways he doesn't understand.

Couchbase is not Apache CouchDB. There are no new dependencies that
are suddenly going to be required by CouchDB.

> I'd feel more comfortable if I knew what the magic and ponies really were at a code base
level, so I could understand better how they will change things for me and my little Couch
apps. I can tell the CouchOne guys are excited about this, though, and trust it means good
things for the CouchDB community.

For the foreseeable future, absolutely nothing is going to change for
you. Its always possible that we'll start to see some patches coming
back from Couchbase employees into Apache CouchDB, but as with all
other development they'll go through the same development processes
we've been using since CouchDB came to Apache.

View raw message