Martin Steigerwald ha scritto: > Am Samstag, 9. Juli 2016, 12:25:14 CEST schrieb Andrej Kacian: >> On Sat, 09 Jul 2016 12:00:20 +0200 >> >> Martin Steigerwald <mar...@lichtvoll.de> wrote: >>> Diederik, I think this is about "I want an always stable and releaseable >>> testing" again. >> >> I disagree. This is more about knowing if the package combination >> I am currently running is worth testing and reporting found bugs on, or >> whether I am in some temporary state where some packages I should >> have are still waiting in the queue, and any bug report I make would >> only waste time of the packagers, or whoever else will be reviewing it. > > That is part of what this list is for. > > Or do you think Debian/Kubuntu Qt/KDE packagers always know which versions of > which packages go together nicely or not? These effects of mixed KDE > Frameworks packages between 5.22 and 5.23 have been unknown to them as well. > > And heck, if you only want to test when all KF packages are at 5.23 thats > easy > enough to tell as well.
Which does not mean that testing users should find their distribution broken every week because some random packages from a block (Frameworks, Plasma) which was supposed to stay together were upgraded randomly. > >> If the KDE packaging layout was simpler, and all packages followed >> e.g. same naming policy or same versioning scheme, it wouldn't be an >> issue, as it would be easy to verify whether I still have some packages >> waiting for an upgrade. But there are various sets of packages >> (around kf5, plasma, kdepim, ...), each having various and different >> version numbers, and it is quite taxing trying to make heads or tails >> of it all. > > Feel free to bring this up with upstreams distributions mailing list. > > Just bringing this up here is – again – a waste of energy. It is not. We can identify what the problem is: some blocks of packages needs to be handled together. Fedora does it, for example. > >>> And, yes, while I think some improvements in the order packages enter >>> testing, would certainly be nice, I also think that the main purpose of >>> testing is just that: Preparing Debian for release by testing the heck >>> out of it. That it is kind of a rolling distro, is in my eyes just a side >>> effect of it. And I think for a rolling distro unstable is more suitable >>> anyway. >>> >>> So that is where I will focus my energy: Testing out the new packages >>> while >>> fully understanding that there can be transitional states of breakage that >>> do not even matter for the final stable distribution as it won´t have KF >>> 5.22 packages. >> >> Fully agreed here, but first you have to be able to identify which are >> the transitional states, otherwise you're just bringing more noise to >> the packagers, as I described above. > > Then help to make this possible. > > Otherwise its a complaint. Its as easy as that. In my oppinion a complaint is > usually something that is not intended to create a real change. And thats > usually exactly the effect of it. > > All of this discussion is a huge waste of time unless someone in here chooses > to *act*. See, we had this before. Every two months an epic discussion here. > Any real change up to now? *None*. Throwing in new packages does not help. Let's start from here. For example, Plasma 5.7 packages should have gone in until at least all 5.23 was in testing. First solve the known issues, then adds new thing. -- Luigi
