Am Sonntag, 1. April 2012, 14:08:24 schrieb Ilya Chernykh:
Why do you assume that any user downloaded each of 480 packages in those three weeks? This is nonsence!
- because only a minority of packages were updated in three weeks so that even if somebody has installed all 480 packages they would not update all
- because there are many packages that contain only one window decorartion style and there is no reason in installing all window decorations for anybody
- because there are many users who use another desktop and only use one KDE3 application such as amarok or kmail.
- because the KDE3 applications go in boundles like kdenetwork and kdepim and each bundle contains most of what an average user would need.
Ilya, you missed the point – completely. Once again. And I won't explain it another time after this email since you ignore the simplest facts and email passages to your liking. You do not have any clue how the download numbers come about, i.e. how many users the represent. I do not either. People could bias them with download scripts, there is no info whether they ignore downloads from the same IP etc. You claim that disregarding the number of packages in a repo is more valid than taking them into account when it comes to assessing and comparing repos' downloads. I claim that you are wrong in that regard. BTW: you also ignored the last paragraph of my email, but anyway. You claimed that people only use 3-4 packages from the KDE3 repo, I claimed that this is unlikely since the KDE3 core packages + Qt are already far more than 3-4. And either you are wrong claiming that you maintain those packages, i.e. add patches, or you are wrong claiming that none of the KDE3 users that use the repo want the patched packages. The latter is my bold claim, i.e. people who use a repo want updated packages. :) You forget that whatever you claim for the KDE3 repo is valid for other repos as well. People do not install all packages from KUA, KDE4 or whatever repo either, so user numbers for those increase as well. I never claimed that my figures are quantitatively correct. In fact I came to the conclusion that one cannot show anything with those numbers – which you try to do nevertheless in order to prove how popular KDE3 is. All I claim is that my approach, i.e. taking the number of packages available within a repo into account when assessing the download numbers of a repo makes more sense than your approach of comparing just the numbers without any regard to the repos' properties. As a side note: you make the mistake of comparing one KDE4 (KR48 it was) repo with the KDE3 repo. You would have to sum-up the different flavours of KDE4 + KUA and Extra. While KDE3 has only one repo for all packages, the packages and hence downloads of KDE4 repos are spread across several repos. So please stop using those download numbers as you have no clue what they actually mean, nobody has. I'm not sure why Jos brought them up, he should know better and since you are a smart guy, you as well. Sven -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org