About the Karma algebra...

[Stuart Bishop]

Matthew Paul Thomas wrote:
> On Mar 26, 2006, at 10:44 PM, Lionel Dricot wrote:
>>> ...
>>> 1) The decrease rate must only be proportionnal to your actual karma
>>> and nothing else. It must not take care about when you've got those
>>> point.
> 
> Why?
> 
>>> ATM, it seems that one karma point lost for a given time and decrease.
>>> It's not really funny because it means that if you had a big burst of
>>> activity 2 months ago (and then a big burst of karma), you will have a
>>> big fall of karma today.
> 
> No, the karma gain algorithm had some heavy adjustments last week that
> made it jump around. The decay algorithm is still the same, though: when
> you earn some karma, it decays at a constant rate, reaching zero a year
> after you earned it.
> 
>>> ...
>>> 2) Displaying only the logX of the Karma.
>>> ...
> 
> That's a nifty idea, though it would hide bugs in the karma code, so we
> might not want to do it for a while yet. Perhaps you could report it as
> a bug on the Launchpad product.

I did some experimentation with the real data, and I think it will work
well. log(karma)*1000 or similar.

-- 
Stuart Bishop <stuart.bishop at canonical.com>   http://www.canonical.com/
Canonical Ltd.                                http://www.ubuntu.com/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 191 bytes
Desc: OpenPGP digital signature
Url : https://lists.ubuntu.com/archives/launchpad-users/attachments/20060330/5f76afc2/signature.pgp