Hi, On Tuesday 14 October 2008 14:03, Christian Lohmaier wrote: [...] > You also say that making a release is too much work., but I bet > by the time you have been arguing against a new release, you > would have been able to prepare on just as easily. I agree, releases aren't that easy to make. But IMO they worth it. Putting aside all that waffle about "responsibility" and "stability", there are my considerations why releases should exist: 1) It is a good point to cleanup, to fix long standing issues, bugs and so on. Of course it isn't possible to fix 'em all for any particular release. But release creation incites developers to put things in order, to sync and converge changes, to sync/finish translations at least (but not at last!). 2) Despite the fact that SVN versions of mplayer are quite stable and nice, most of users use a versions from their distro's repositories. And most packagers do not care enough to update something from SVN (gentoo is the only known exception for me). Well, I do not say that I like that, but this is how things are done in Linux oftenly, most of the time. Well, to take stock off: release creation will make newer version of mplayer available to the much wider community of people, most of which are not qualified enough to use SVN or are lasy enough to do this (and do not forget, every good programmer is somewhat lasy 8-))). Look for the mplayer-users list archives. You will see a huge burst of bugreports and feedbacks after each release. And this is why releases worth all work done to make them. Do not underestimate the value of the user feedback. Even the best developers are not able to test their program thoroughly due to limited numbers. And some good projects died just because they thought they're irreplaceable, do not repeat their mistakes please, I really want to see radiant future of MPlayer. At least once a year we can spend some of our time to make a release. So, to summarise. Release creation isn't easy thing, but they have cogent advantages. That's why IMHO releases should be created in the interval somewhere between half a year and year. Too often releases take too much time, OTOH too rare may drop the quality of code (due to lack of cleanup and pending bugs) and distract some users. Well, I can help with release creation with what I have. Unfortunately, me free time is limited (well, perhaps the same as for all of you), but still I want to do something, perhaps, some fixes )). However, in my opinion someone of experienced veterans should take the leadership of this release creation. Perhaps, Diego? Best regards, Andrew -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.mplayerhq.hu/pipermail/mplayer-users/attachments/20081014/a6ade3ab/attachment.pgp
