On Sat, Nov 17, 2007 at 08:10:37PM +0100, Gudjon I. Gudjonsson wrote: > Hi > I read your arguments Ron. You mention the frequent upgrade of Wx 2.8.
And some things about bugs, the lack of a real upstream freeze, the lack of any viable transition plan for existing apps, and its impending obsolescence when wx3 arrives supposedly in the next month or two -- just to mention a few that I think are greater and more specific showstoppers than just 'frequent updates'. > This probably means that Wx 2.8 should stay in experimental but I don't see > it as an argument to keep it out of Debian. I don't really follow that. Saying it should stay in experimental is saying it will not be a part of any Debian release. If it will not be a part of any Debian release, then it is not part of the equation for "what we want in Lenny" -- which is the problem that we actually need to solve. So I don't really see how busywork on this is wise if we are already getting tight on time for the important stuff to be done. > I really need Wx 2.8 into Debian now to be able to release a new > version of GSpiceUi I've said previously that I quite sympathise with people stuck on what we might call the hemorrhaging edge here, but no amount of sympathy can change the facts on the ground. Until there is a new stable release suitable for the distro, the release date of their 2.8-only app revisions is blocked waiting for a viable wx release. That's the way release cycles work. We had similar trouble when 'stable' wxPython released depending 'unstable' (meaning its API changed nearly every day still) wx. A whole pile of 'stable' apps were all teetering on an unstable daily snapshot of a fundamental dependency. But re GSpiceUi specifically: what exactly do you 'really need'? I see we have a version in the distro headed for testing that has no serious bugs reported against it... Does the new version really have something so cool that its worth the chaos to everyone else to rush introducing it? > and guess the Wx team would be happy to get feedback from the > Debian users. They are already getting bug reports faster than they are resolving them. If getting even more of them even faster would make them 'happy', then perhaps we are in deeper trouble than I had first thought ... ;) > I do understand that it is too much work for you but since there is a lot > of people willing to maintain the Wx 2.8 version in some collaborative > maintenance, please let them do that. It cannot do any harm :) Please. End this myth that I'm doing anything to stop anyone who actually wants to help from doing so and that hordes of people who would do so are lurking in the wings. How many times do I have to repeat this: We have had collaborative tools for years now. To date, everyone who has ever asked for access to them has got it. If nothing appears to be happening out of that, then I hope you can also understand that I don't feel it is my place to nag them about it. My #1 job here is to try and assess what our best bet for Lenny will be. So far that's looking like stay where we are, or make the leap to 3.x, but its still too early to say which. If people have other ideas, it doesn't really help to just profess them, they actually need to do the work to make them viable in the time that we have to us. So again, and hopefully for the last time, people who want to help and have a plan for how they can should ping me privately about it -- and it would be a great help too if the people who can't or won't or don't, can just let us get on with that without too many regular showers of magic bullets to dodge. A big part of this whole collaboration thing is _listening_. It's very hard to propose a credible Better Solution if you're not paying attention to what the big problems really are and/or shortsightedly blaming the wrong things for them. I know this is an awful mess, but it really can get worse if people try to force-fit solutions that blindly assume it isn't. Ron -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
