Why

1. External interest in pylucene is increasing
2. pylucene is of interest independent of Chandler
3. pylucene needs work to become an installable python extension independent of chandler
4. Broader use/adoption of pylucene will help us find bugs
5. pylucene will need to track Lucene releases and improvements

Goals

1. To get practice running a small open source project
2. Develop a community around pylucene
3. Ultimately to attract a non-OSAF staff person as the owner for pylucene [mid to long-term]
4. Increase quality of pylucene by broader exposure and testing
5. Attract people to chandler via association with pylucene
6. Diversify the OSAF portfolio.

Benefits

pylucene is a relatively small code base with limited exposure. Because it is a wrapper around Lucene, much of the work consists in figuring out what a natural binding to Lucene is. Much of this work has already been done. But there is a bunch of work in making it into a separately installable python extension, covering the full Lucene API, and doing more quality / stability work.

Risks:

1. The API grows in a direction that we find unusable - I think this is unlikely. It's more likely that new layers develop atop the existing API.
2. We divert resources from tasks needed for 0.4 - We'll always have this problem.
3. We spend a lot of time fielding questions - This will be true if no community develops.
4. We invest a lot of energy in setting up infrastructure for a subproject (mail, cvs, bugzilla, wiki, etc). - This also has to happen sometime, and it's not that much energy.
5. Nobody comes - There's not a lot we can do about this
6. The project stagnates because Lucene is "mature" - If we get a stable bug-free pylucene then this is probably okay.

How:

• Setup separate version control system, mail, bugzilla
• Create TODO list of tasks remaining to be done
• Mentoring of initial OSAF module owner by Ted - up to .5 day/wk (out of Ted's 1 day/wk)
• Andi as initial owner
• Announce

-- TedLeung - 02 Jun 2004