Anthony Towns wrote: > On Tue, Apr 24, 2001 at 11:28:09AM +1000, Edward C. Lang wrote: > > On Mon, Apr 23, 2001 at 02:27:47PM -0500, Jimmie Houchin wrote: > > > I have experienced roughly the same things. > > Same here - should it be recommended that users first upgrade to woody, and > > then to sid/unstable? > > I'd go so far as to say it's not clear that it should be recommended > that people upgrade to woody or sid at all, even.
Personally I think a distribution/version between stable and testing which was based on testing but the packaging was complete in as such no dependencies were broken on any package would/could be a popular distribution. I have no knowledge as to difficulties or challenges in doing such so I speak naively. The software in such a distribution would be reasonably current. Some as current as testing/unstable. I don't if testing is supposed to already work that way and it just happens to be a bug that it didn't. I am not knowledgeable about the criteria for testing. There seems to be a great difference between software in Potato and software in Woody. Just a simple glance at packages of wine and alsa* reveal tremendous differences. If a user developer desires to be reasonably current with certain software or kernels then Woody seems to be the best option. I want to use the 2.4.x kernel, alsa and wine. Yes running Woody can offer challenges. Are those challenges significantly more than trying to keep up to date on software such as above on Potato. It seems Potato in such a manner is also challenging. This is why I choose Woody. I could run RedHat 7.x but I think Woody still sounds like the better way to go. :) I also use software I want/need to compile with the newer libraries. I will be learning how to compile kernels and software the Debian Way tm. I like what I read about the benefits of such. > But this is the -testing group, after all, so if anyone can make that > recommendation it's you guys. > > It'd be nice if such recommendations were based on facts rather than > general feelings, and make them sensible qualified statements rather > than sensationalist headlines. But you take what you get. > [snip] > One of the things that'd be *really* helpful is logs of the upgrades. If > you're using apt-get at the commandline, then it's trivially easy to > get good logs: just run "script" first. This is probably good enough > for people using dselect too, although the result'll be a bit harder to > interpret. Dunno any good way of logging what's going on with the apt > frontends, or whatever else is out there. I did my upgrades from apt-get. I am on a PC at work. What are you referring to here as "script". Is this a commandline option for apt-get? I am more than happy to learn what I as a user can do to help those who are developers. What I know in the process I took over several complete reinstalls on a clean drive is this, dist-upgrade to Woody left me with an unusable system due to dependencies not met. Or should I say Gnome/X11 were in a state of disrepair. Dist-upgrade to Sid and I now have a working system. I am not saying nothing will crash with Sid but that I can handle. From my naive perspective in doing a simple dist-upgrade from a basic, very minimal Potato there were enough differences between Woody and Sid to warrant the experiences. I am about to get my system configured and am currently pleased with what I have. I would rather be a positive contributor than a negative detractor. I'll be a good guy and go read the bug page. Is there any other documentation on how I can provide usefull data when something fails? Anyway any comments you or any has to help us become constructive citizens of the Debian community please do so. Thanks, Jimmie Houchin
