William Hooper wrote: >> FC-2 was not sufficiently well tested. > > Works for me. It also works for a number of people. Maybe some cases > were not well tested, but you are making a blanket statement that a number > of people would disagree with. You have to keep in mind when making > conclusions based on mailing lists: People who don't have problems aren't > complaining. The fact that it works on your machine is neither here nor there. "Sufficiently well tested" means "tested on a sufficiently wide range of machines". >> Personally I haven't found FC-2 too much trouble, >> and I will stick to Fedora, and continue to upgrade unless it becomes >> much more difficult. > > So you haven't found it to much trouble, but yet it "was not sufficiently > well tested"? I read this mailing list. The distribution failed disastrously on fairly common machines. In my case, the kernel that came with FC-2 failed on my SCSI-only machine because it did not invoke the (very common) aic7xxx driver. IMHO this was a pretty basic error, but it didn't cause me much trouble because I had an alternative kernel available. >> I think Fedora could and should try to develop >> as a replacement for RH-9 for as wide an audience as possible. > > This is not the stated goal. It never has been. I didn't say it was a "stated goal". I simply suggested that it "could" be a goal. There is a need for such a replacement, and Fedora is the nearest to it. If I had put together a distribution I would want it to be used by as many people as possible. >> It's important in my view to keep a balance >> between adding new features to Linux and at the same time ensuring Linux >> is >> as easy to install and use as possible. > > How does adding new features to Fedora make it hard to install or use? You misunderstood me. What I am saying is that more time and energy should be allotted to simplifying the installation and use of Fedora, and in fact of Linux generally. > How would being able to do an upgrade change you willingness to test? > Test releases are likely to have problems that can cause data loss. > Running them on your "one and only" machine isn't a good idea anyway. I'm actually running FC-2 on several machines, and wouldn't mind occasional problems. To me it is obvious that there was insufficient testing of FC-2, and therefore it is reasonable to enquire how more people could be involved in testing. You talk as though the "Laws of Fedora" were handed down on tablets of stone. Fedora is a fairly young project, and should be able to make sensible adjustments to its methods. -- Timothy Murphy e-mail (<80k only): tim /at/ birdsnest.maths.tcd.ie tel: +353-86-2336090, +353-1-2842366 s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland
