Bas Zoetekouw wrote: > Hi Luk! Hi Bas > You wrote: > >>> In other words, if almost no-one uses them, does it matter if the packages >>> are of decent quality? Also, if almost no-one uses them, how do you know >>> they're of bad quality? >> It does matter if they are of decent quality as we need to support them >> (mirrors, infrastructure, security support etc.). I don't say they are >> of bad quality, don't turn my words... > > ATM, the support costs of removing the packages are much bigger than > just letting them sit in the archive. > >>> I think you should be looking for some additional metrics for package >>> removal, such as age, date of last upload, for example, in addition to size >>> of install base, before deciding that a package is stale. (Because IMHO, if >>> a package has no bugs filed against it, you can't honestly say you want to >>> file for removal because it's potentially buggy.) >> Note that besides not being used by many people, the packages are >> orphaned and will be tested to the points: >> >>>>> (a) aren't ITAed, and >>>>> (b) have been orphaned for more than, say, three months >>>>> (c) don't have some special reason why popcon would be unrepresentative >>>>> (d) don't have any other special reasons to stay in Debian > > Well, I could agree with removing packages if, in addition to the cited > citeria: > > (i) there are RC or important bugs (or many normal bugs) > > or > > (iia) there are very few users (in the bottom 1% of popcon number or > so, not sure how much votes that would translate to, 20 sounds a > bit high IMO), and It's not about votes, but about installations, votes should be less or equal than the amount of installations... 20 is less than 1 in 500, so that is also less than 1% in my world... > (iib) there are replacement packages in the archive that deliver > similar functionality. This could be looked at, though I don't know if it's worth the trouble for almost unused packages. Though I'm sure ftp-master would decide on that if we would ask for removal... > In other cases, IMO, it's a lot of work to check and remove the > packages (both for the QA and the FTP teams), without any real gain for > the project. I don't see how it can be still a lot of work now? The real gain is less packages to support which many people already told us would be a good thing... Cheers Luk -- Luk Claes - http://people.debian.org/~luk - GPG key 1024D/9B7C328D Fingerprint: D5AF 25FB 316B 53BB 08E7 F999 E544 DE07 9B7C 328D
Attachment:
signature.asc
Description: OpenPGP digital signature