On Mon, Aug 17, 2009 at 23:02, Alexey Zakhlestin<[email protected]> wrote:
>
> I have first results regarding 9.06-07 sprint-task. We have 19
> days of stats-history, so, I was able to get 5 days of actual
> karma-digits.
>
> Formula used is:
>
> $karma = ($yesterday * 0.999)
>        + 0.1*($avg_d_week - $avg_d_last_week)
>        + 0.1*($avg_c_week - $avg_c_last_week)
>        + 0.1*($avg_v_week - $avg_v_last_week);
>
> d = downloads
> c = comments
> v = rating-votes

This'd be the only element in our current karma system which decays
over time, then (IIRC)? Now, I'm all in favour of karma (as a whole)
decaying, but I'm not sure about doing that on an individual element.

Is this solely to be used for the authors' karma rating, or is it also
going to be used for identifying "Hot" downloads on maemo.org? And the
aim is to track the "vitality" of an application?

> there is a large group of apps, which have highest "karma"

Indeed. It seems very very bunched up at the top there. Perhaps some
kind of non-linear function needs to be introduced to space things out
a little?

Certainly a very good starting point, and the Flash demo gives an easy
way of visualising.

Cheers,

Andrew

-- 
Andrew Flegg -- mailto:[email protected]  |  http://www.bleb.org/
Maemo Community Council chair
_______________________________________________
maemo-community mailing list
[email protected]
https://lists.maemo.org/mailman/listinfo/maemo-community

Reply via email to