pidgin.next.minor: 9e31989c: A patch from Chris Connett to change the...
kevin at simguy.net
Mon Jan 26 14:52:05 EST 2009
Etan Reisner wrote:
> On Mon, Jan 26, 2009 at 12:55:39PM -0600, Kevin Stange wrote:
>>> Here's the relevant description from #5447:
>>> It implements a new buddy list sorting method. Buddies are
>>> sorted by their recent activity, i.e., how much text you have
>>> exchanged recently.
>>> It works similarly to the log size sorting method, but instead
>>> of counting number of bytes in a particular conversation, it
>>> counts bytes in the conversation decayed exponentially by age,
>>> with a half-life of 14 days. Thus, people you've started talking
>>> to a lot in recent days will climb to the top of the buddy list.
>>> Since recent text counts more, new friends you've been talking
>>> with recently will claim the top spots, instead of the person
>>> you've known a long time and who has the largest log.
>>> Note, the patch as committed *replaces* the existing sort method. I'm
>>> especially interested to know how people feel about that.
>> Why did you replace the existing sort method instead of adding an
>> additional option to the menu?
> We couldn't think of a reason to have/need both. I think the fall-off
> behaviour is most likely what people are going to want and I really don't
> want to have to come up with explanatory text which explains the
> difference well enough for people to understand what they each were if we
> kept both.
> Does anyone have a reason they think we should have both (and ideally some
> good text to explain them)?
I can't think of a compelling reason to keep them both, since chances
are you sort by log size to keep people you talk to a lot near the top.
If you don't talk to someone a lot, but used to, keeping them at the
top forever probably doesn't serve much other than nostalgia.
If you wanted to, you could call the:
Sort by Log Size
Sort by Recent Activity Level
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 260 bytes
Desc: OpenPGP digital signature
More information about the Devel