Back from FOSDEM in Brussels. In their own words on the information booklet: "4000+ geeks, 200+ lectures, 2 days, 0 EURO". I had two motivations to go there: Brussels is close to my home in Amsterdam and always nice to visit; and I could sit and listen a whole day to Drupal presentations, and get the opportunity to check out some other projects too.
As the FOSDEM booklet says, it’s a gathering focused on lectures, so I got my portion of sitting still and getting powerpoint-poisened in over-crowded and under-ventilated rooms. A lot of people breaking the "show me, don’t tell me" rule. But Brussels was nice, and some of the presentations on Drupal and Thunderbird were useful for me:
Dries Buytaert presented the general direction for Drupal 7: taking the top feature requests from end users and developers from a survey as suggested priorities to work on by the community, to hopefully deliver a killer release by 2009 to succeed to recently released version 6.
The long-term focus will shift from modules and functionality towards data and inter-operability, being able to integrate content from different sources, allow others to reuse, and decentralisation: a general move towards RDF and its "subject-predicate-object" triples, and towards XML to better define web services interfaces, as well as allowing for object validation and so on.
"Search" in Drupal
Robert Douglass spoke about his ApacheSolr module: searching on Drupal is a disaster still, and he nicely demonstrated that by showing he couldn’t find a post of him complaining about this on drupal.org, but showed it was the first result via Google. Also, the search function on drupal.org had to be disabled a few times at high loads, because it would bring down the whole server.
So Robert built the ApacheSolr module to work with the Solr web service (which in turns works with Lucene for the indexing), and actually get better and more useful results on the content of the Drupal site. And Solr scales perfectly, and offers interesting extra options that make search a lot more useful.
There also was a presentation on the recent launch of Mozilla Messaging, and the interest of Mozilla to make get to a staff of around 10 people working on Thunderbird as influential mail and calendar application by the end of 2008. The challenge is finding the right people to grow the team in the right way, funding seems to be no problem.
With a possible version 3 by the end of the year, including a calendar and better workflow, this all makes me feel a bit more confident again about the future of Thunderbird, after discussions a while ago seemed to suggest it might be abandoned.
I found it interesting how people in some projects seem to look at the current/"old" situation with a relatively inward-looking orientation, and define themselves as an improvement over that; while others look more at "best of breed" practices, and seem to be more competitive and/or open. Drupal in my perception fits a bit in the former category, I had a bit the same feeling I had at BarCamp in Amsterdam in 2005: a very strong drive to "do it yourself" and a bit of a barrier of entry for outsiders.