CouchDB is a distributed document-oriented database system written in Erlang. The project graduated to TLP in November 2008 and Damien Katz was approved as
the Apache CouchDB PMC .

- Community

The community around CouchDB grew in the last couple of months quite a bit. We gain attention in early adaptor markets and can look at the first production grade setups running on CouchDB (iWantMyName, pcapr.net<http://www.pcapr.net >).

The mailing list traffic grew and so did the number of third party library implementations that interact with CouchDB. More and more support on the mailing list comes from outside the core team which proves that there is an active
community growing.

Multiple open source projects and commercial projects and products built on CouchDB have been released or launched in the last few weeks. We are trying to keep track of them onhttp://wiki.apache.org/couchdb/ CouchDB_in_the_wild on
github reveals around 30 projects around CouchDB.

The three committers Chris Anderson, Noah Slater and Jan Lehnardt started releasing chapters from their work-in-progress book "CouchDB: The Definitive Guide" on http://books.couchdb.org/ The content will be licensed under Apache 2.0 and published in print by O'Reilly Media. The book mailing list has 340+
subscribers and the feedback is immense.

For the book, Chris and Jan are working on CouchApp, a toolkit for writing
standalone CouchDB applications that is gaining traction on its own.


- Development

We are now looking to release CouchDB 0.9.0 in January, our first as TLP and our
first BETA release .

New features for 0.9.0:
- Security and data validation functionality
- View server internals re-architecture to allow the ability to quickly read
    stale views as indexes are being updated.
- Multi-key view lookup enhancements
- Include_docs. Ability to load full documents with view rows.
- Btree performance improvements
- Faster and more compact internal JSON term format, that is also the new Erlang canonical JSON term format. Additionally, when a low level JSON parser is implemented in core Erlang, it is will use this format and we'll gain more
    performance for free.
- Ability to run a CouchDB server right from the project source directory,
    for development purposes
- Streaming attachment writes
- fliters for converting json documents to other formats (XML, HTML, etc) - Deferred commits (optional acid) for faster update and replication speed


- Infrastructure
We have now transfered all couchdb assets over to the final TLP locations,
including the couchdb.apache.org site, the mailing lists and the SVN
repositories.

Reply via email to