Yes sir, working groups are the rage at OSAF these days. I ran in to Mitchell Baker in the hall and we started talking about community oriented stuff at OSAF. By the end of the conversation, I was a member of the community working group. This is a good idea for OSAF because community/open source development issues are a cross cutting concern for the organization. It does feel a little weird to be part of such a group. At the ASF, its part of the culture that all developers are involved in building and maintaining the community. The idea that you would partition the community activites and the development activities is a foreign one. So for me its no big deal to be part of the community workgroup, because its part of my definition of what it means to be an open source developer. Mitchell and I also talked about all sorts of issues related to the future of open source. They are going to be trying some interesting ideas at Mozilla.org, so keep you eye on what they are doing.

Today I attended the first meeting on the query system for Chandler. At times we wandered down to the implementation level, which is a bit early. We need to get a very clear grip on the Chandler data model, so we can use that to design a suitably capable query language. I also need a bunch of sample queries or use cases that would result in queries. In addition to traveral of sets of items in an item graph, we'll also be integrating a full text searching capability. Another unusual feature that we talked about is a notification system. What you want is to be notified when other threads update the contents of the database in such a way that the contents of the a query result would change. Good thing I did some reading on active databases when I was a grad student. That stuff might come in handy.

I also went to the Apps working group meeting. These are the people who are bringing you all stuff that you can see. One of the more notable things that I learned was that the RSS block/parcel is going to be used as an example of how to write a block, use an agent, and access the repository. I think that this is great, and will get us to the point where you could imagine building a really cool aggregator as a Chandler block.

Morgen Sagen was in the office today working on the build. The bad news is that the Panther build on MacOS X is broken -- it builds but doesn't run, and it's nasty after that. In an effort to help out, I checked out a completely clean copy of Chandler (I had copied over my Linux checkout to the Powerbook and done a clean build) and discovered that while Chandler won't run, at least the unit tests (such as they are) will run. That unblocks me so that I can get back to work on them, although working on the tests is a lower priority compared to maximizing face time this week. In sad Powerbook news, the new machine appears to be slower than the 1 GHz Linux box that I was using for Chandler builds. Hope is not completely lost, though, because I have 512MB of RAM waiting at home, and the 7200RPM disk that we ordered for the Powerbook arrived. So when I get home I can put it in a Firewire enclosure and see if it makes a difference. Last night I located a copy of the Apple Service Manual for the FW800 Powerbook, so in theory I know how to replace the disk if it works out. Whether I'll actually do it or not is another story.

Tonight I took my friend Rachel Reinitz out for "a very nice dinner" at Venticello's. Rachel was one of my references for the OSAF job, and we've done a bunch of consulting work together in the XML and Web Services space. It was good to catch up with her and find out what she's been up to. We have a system where big favors end up being appreciated with "a very nice dinner". Lately, I seem to be doing all the dinner buying...