Some members of the Twisted development team, including myself, attended IPC10 this last week. ( http://www.python10.com/ ) In general, the conference was a huge success. The talk that Moshe Zadka and I gave was heavily attended. Various materials related to the conference, including photos, the paper, and erratta are available at: http://twistedmatrix.com/users/glyph/ipc10 There will probably be a few more additions, since I've been really busy after getting back from the conference. Also, most of the Twisted development team met in meatspace for the first time. If you were there, either at the paper presentation or at the Birds of a Feather session, thanks for dropping by! We got some good feedback from the people who attended both of these, and that was mainly that our key problem is (surprise) documentation. It was hard to decide what features of the conference to describe in detail here, since I had a lot of fun and met a lot of people, so I'll focus on what impact these things will have on the future development of Twisted itself. There was an incredible amount of interest in Twisted; I found that often, rather than trying to generate interest, I was having to keep people's expectations low. The chief reason was that Twisted doesn't have very good documentation, and so it's difficult for end-users to get into it. I had underestimated how useful something like Twisted could be to the average web developer. Currently I'm trying to develop a documentation strategy, and I'm still not sure what form it will take. If you are interested in writing documentation, but don't understand Twisted well enough to do so, please contact me and let me know what information you need to get started. The file doc/index.html in the distribution is an outline of the docs we need -- there is nobody signed up to work on them yet. A few people have mentioned their interest, and if we're lucky, enough time will clear up in all of our schedules to permit working on that. While Developer Day was going on, the Twisted team met outside the conference to talk about the Version 1.0 release strategy. I'd like to get to 1.0 by about april, since I think we've already got a lot of functionality, and one of the things that's making it difficult to document is that it changes all the time. It is easiest to define release criterea when your test suites are written by someone else. We hope to leverage that by using use-cases derived from other sorts of server software for those parts of Twisted which are not really integrated: we can tell when twisted.mail is at the postfix/qmail level of mail application, when twisted.web can compete with apache, and when twisted.names can replace BIND. Those are all going to be release criterea, but we also need some way to decide when our integration between services is good enough. Since including them in this email was making it rediculously long, I've put up a webpage that is a first draft at figuring out what we need for 1.0: http://twistedmatrix.com/users/glyph/twisted/one-point-oh.html Twisted Developers (especially those not at IPC10), your comments are appreciated. Hope to see you all there next year! -- "Cannot stand to be one of many -- I'm not what they are." -Guster, "Rocketship" glyph lefkowitz; ninjaneer, freelance demiurge glyph @ [ninjaneering|twistedmatrix].com