2008/9/27 Gregory Maxwell gmaxwell@gmail.com:
The existing system correctly plays back for an overwhelming majority of users (I posted some stats on this two years ago or so)
Could you give a link, just for reference purposes?
I think this discussion needs to be informed by as much hard data as possible: Do users manage to successfully and repeatedly play back a full video with sound, and how does their success rate compare with other solutions? How do they behave when they cannot play the video? I'll look into what we can do to get up-to-date user experiences. Things should improve significantly with Firefox 3.1, so we may want to hold off any new tests until it has gained some adoption.
I don't think there's any question that we are committed to making sure that every piece of user-facing content, including interactivity, can be accessed using 100% open source software. That's an important consensus. The primary issue is the question of parallel distribution, which is one on which reasonable people can disagree. We should collect as much data as possible to help the Board reach a decision on that question.
BTW: I do think the question about supporting OOXML is a relevant and reasonable one. PediaPress has developed an open source wiki-to-ODT converter as part of their wiki-to-print technology, and I think one can reasonably ask whether we want to support OOXML in the future as an output format. An example use scenario would be a teacher downloading a collection of Wikipedia articles to edit in Microsoft Word. Again, it raises the same issue of competing interests (primarily a belief in short term impact vs. one in long term transformative changes).