Putting the Apache back into CouchDB

Over the last year, the Apache CouchDB project feels like it’s been in a holding pattern between missing Benevolent Dictator and ASF-style meritocracy.

It hasn’t helped that Damien Katz, the original visionary developer of CouchDB, has been largely absent in the community since the CouchOne and MemBase merger. And that’s a shame - few know CouchDB better than him.

His blog post about the future of CouchBase comes as no surprise to anybody who followed the merger; CouchOne was funded by venture capital, and as such the owners will be pushing for a return - and there’s nothing wrong with following the money.


An important goal is to keep interfaces in CouchDB simple enough that creating compatible implementations on other platforms is feasible.

– @DamienKatz FAQ about CouchDB and its new IBM overlords

Despite that, I still think that Apache CouchDB is not only relevant, but indeed essential in the increasingly internet-connected world. The trend is for people to put their personal data into a corporate cloud, where it is sliced, diced and ultimately resold. For most users, there’s simply no alternative, and CouchDB is to my mind the best way to develop a peer-to-peer internet.

I’d like to see CouchDB as being the enabler in open data, in breaking open the web for joe & jane user, and enabling interoperability of large data sets especially in research and for government. And an independent replication protocol, with a reference implementation, well supported and documented, is critical to that end - even if the back end isn’t actually CouchDB.

Community and Commitment

The points made by both Mikeal and Damien, and others, are valid, however. CouchDB needs to have both focus and drive, to keep up its active current developer base, and to grow the user community. And it needs to happen this year or it will likely be too late.

Healing the Wounds

So where to from here? What’s really needed? The top IRC peeves are not too hard to address:

But clearly that’s not sufficient – in terms of restoring confidence, and increasing mind-share, things are not so straightforwards.

Fundamentally the Apache CouchDB community needs to get it together – to start demonstrating progress to the wider coder community, and do so in a way that articulates the project’s objectives, and restores faith that the Apache Incubator process really did create a community around the code, that can survive the clumsy departure of its inventor. Here’s a suggestion:

Communicate the Way Forwards:

Make it easy to get started:

Continue to Build the Community:

Interestingly enough, there’s not a lot of code required, but there is a fair bit required from the community.

This is a great opportunity for people who’d like to contribute, and who are keen to learn more about CouchDB along the way.

Are you joining me?