On Thu, Dec 4, 2008 at 11:18 AM, Loui Chang <louipc.ist@gmail.com> wrote:
On Thu, Dec 04, 2008 at 10:47:24AM -0700, w9ya wrote:Things change Bob. Nothing in this world is set in stone, and most of us
> Well IF you go back far enough into the mail archives (which may NOT be
> possible at this time because of current issues with that system) you WILL
> run across those discussions about the voting being added to the TU/Aur
> system. At that time we were SPECIFICALLY told that this would not be used
> for restrictions in the future. Writing about that now is NOT being
> dishonest. It is rather DIRECTLY related to what you propose.
probably never made those promises. We can't be held to them.
You've proposed nothing that accomplishes the same result.
> As for not giving your proposal a chance. Your very correct about that. I
> CHOOSE not to give it a chance. That is NOT however dishonest either. It is
> not a bad thing to speak out about a proposal one does not like and sees
> other way to accomplish the same result.
I have no problem with using votes as a metric.
> I know it seems like a circular argument, but that is because your proposal
> BEGS for a reason, and simply putting faith into a faithless entity like an
> exceptional poor tool like the aurvotes is the heart of the matter with your
> proposal. Even people supporting your proposal are quick to point out that
> the aurvotes stinks as any form of metric.
>
> You should FIRST come up with a useful tool and then a need to make the repo
> more "efficient", THEN and ONLY THEN shoudl you be asking us to consider
> such a proposal.
Three stats have been proposed: votes, pkgstats, and downloads.
We are using two of those three. Downloads aren't quite feasible because
they raise privacy concerns and there are technical problems in counting
them. They'd probably show similar results anyways.
When I see a problem I do what's in my power to correct it.
If you have a problem with any of the stats that we are using, then you
should suggest something else. Then again, your issue isn't really with
the metric, it's with the proposal itself. That's why you haven't
offered anything in cooperation to this discussion.
Considering your opinion of votes I wonder why you were so concerned
about votes here:
>From http://archlinux.org/pipermail/aur-general/2008-February/000741.html
> P.S.... Will it be possible to retain or reinstate the 250-odd votes
> this
> package received ?; as it is now NOT extent in either unsupported OR the
> community repo, and it would be nice to be able to properly reflect the
> voting.