[Pidgin] #5447: Buddy list sort by recent activity

[Pidgin]

#5447: Buddy list sort by recent activity
-------------------------------------+--------------------------------------
 Reporter:  cconnett                 |        Owner:  rlaager     
     Type:  patch                    |       Status:  new         
Milestone:                           |    Component:  pidgin (gtk)
  Version:  2.5.2                    |   Resolution:              
 Keywords:  buddylist sort activity  |  
-------------------------------------+--------------------------------------

Comment(by cconnett):

 Here's a new patch that I think addresses what you brought up in the first
 three points, rlaager.  I added my name to COPYRIGHT (Chris Connett).  If
 you want an email, use chris at connett.net.  The new patch compiles and runs
 cleanly and is running well on my local system.  As for the fourth point,
 I can only offer my thoughts.

 I initially created this patch because I wasn't satisfied with how sorting
 by log size was sorting.  I was really looking for recent activity sort
 when I initially used the log size sort.  I had exactly the problem
 outlined in #263.  Going forward, all I would ever use is the sort
 provided in this patch.  I wonder how many people are using log size sort
 when what they desire is recent activity sort?  I really like the
 difference recent activity sort makes.  I think it is the right thing to
 do.

 As for the problem of breakups or other reasons why one abruptly falls out
 of contact, I think that if one doesn't want to be reminded of a person at
 all any more, it seems to me like they should delete that person from
 their buddy list; to force an immediate drop in sort order, they could
 delete or archive the logs.

 If they just want to change the decay rate, I suppose that could be made a
 configurable option.  I gave significant thought to the 14-day half-life
 that I hard-coded here.  It seemed to be simple and generally appropriate
 to the medium of IM.  I'd be wary of adding a configuration option, as it
 might be hard to explain, and it may not be immediately apparent to a user
 what it does.  Then again it may be simple to explain --- user interfaces
 is not one of my strong suits.  If this patch is accepted as is, a
 configurable option could always be added later: The decay rate is encoded
 in one line of code.

-- 
Ticket URL: <http://developer.pidgin.im/ticket/5447#comment:12>
Pidgin <http://pidgin.im>
Pidgin