On 06/14/2012 03:41 PM, Stephan Kulow wrote:
On 14.06.2012 14:00, Angelos Tzotsos wrote:
I am not trying to add noise to this discussion, I just thought that this suggestion should have more attention. Sorry if I didn't make any arguments to support this, so here they are: Thank you for investing some more time to give your arguments.
I have tried beta1 on both virtual environments and through live cd's and I realize that there are many problems and issues unresolved at this point (zypper-yast, grub 2, kde). The most annoying for me was the grub installation that failed everytime under virtual box. I totally agree with you about not wanting to release RC1 yet.
At the same time I must share the feeling I have that Tumbleweed seems to be very stable for a long time. At least to the part I am actively involved (Application:GEO) I can see that it could be used as a base to a new release, since we are getting fewer failures on the long run.
I understand on the other hand that having a move to a rolling model would be more demanding, and as GregH stated he would need more man power on this. Plus there might be an issue on kernel updates versus application updates. I am also worried how this would work for server installations (that I also have to maintain). This is a *bit* unfair, no? Of course the zypp in tumbleweed continues to work, because it didn't change - nor did the toolchain or anything but the kernel below it changed.
So yes, tumbleweed works and tumbleweed is a very good replacement for people who want up to date software while not suffering the 100 repos problem, it's still not a replacment for openSUSE releases with new toolchains.
Greetings, Stephan
Hi Stephan, Yes, it is unfair since major packages have not been replaced in Tumbleweed. We have 2 ways of seeing this: 1. When we have major changes, things will break fast and ugly. But this is how we can have latest versions. This applies to Factory the same way we have to upgrade a bunch of stuff in a smaller community repository, when a major library upgrades. I want to be fair here and thank you for the hard work you do, fixing stuff after all these changes. My point here is that since this is expected, we should not be sensitive, asking for a new development model even it makes the next version slip the release date. We should just accept that it will not be ready on time, and that is fine. 2. If we believe that slipping release date is very important even if it happens rarely, then we should consider not breaking everything at once (as suggested in another thread) and/or changing the development model. Cheers, Angelos -- Angelos Tzotsos Remote Sensing Laboratory National Technical University of Athens http://users.ntua.gr/tzotsos -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org