Not to sound like a broken record, but I think it's still unclear what 1% usage from pkgstats (or 10 AUR votes) actually means in terms of true package usage. I didn't know pkgstats existed until recently, and I suspect there are many other Arch users who still don't. I am speculating, but it may be that a 1% usage rating from pkgstats really means "popular with power users" rather than "popular in general". Also, there may be some packages that have 0% usage on pkgstats (and/or 0 votes) but enjoy relatively high actual usage among users who don't know about voting or pkgstats -- probably some easy-to-use GUI applications that the more hardcore crowd would never touch, but the casual crowd depends on. Even if it is determined that the statistics are skewed as I suggest they might be, they still might be deemed useful. This goes back to the purpose of the [community] repo; maybe it's just not important to cater to the needs of those who don't invest the time to find out about voting and pkgstats, but I think that's a decision that should be made explicitly rather than just assumed. Drew On Tue, Dec 2, 2008 at 4:30 PM, Geoffroy Carrier <geoffroy.carrier@koon.fr> wrote:
On Wed, Dec 3, 2008 at 00:39, Allan McRae <allan@archlinux.org> wrote:
* Only "popular" packages may enter the repo, as defined by 1% usage from pkgstats or 10 votes on the AUR.
We really gotta make pkgstats results official, or at least provide the URL: FIY, it's http://www.archlinux.de/?page=PackageStatistics
It just took me ages to find it again :)
-- Geoffroy Carrier