I don't think using a float is intuitive. To play half a song twice doesn't mean that the song was played one time.
But I liked the idea of using a threshold (which could be configured via gconf). Regards, On Tue, Jul 22, 2008 at 13:56, Steven Brown <[EMAIL PROTECTED]> wrote: > I like the double float idea... Then you could just accumulate the > percentage of a song played. Play through half a song, add 0.5 to the > playcount. Play through 10%, add 0.1. 90%? 0.9. > > 0.5 + 0.1 + 0.9 = 1.5 -> rounds to 2 times. > > On Tue, Jul 22, 2008 at 9:25 AM, John Daiker <[EMAIL PROTECTED]> wrote: >> >> Peter wrote: >>> >>> Arthens wrote: >>> >>>> >>>> I'm going to suggest a little (I think, I hope) change to the way >>>> rhythmbox >>>> counts how many times you have listened a song. >>>> >>>> At the moment (version 0.11.5 on Ubuntu 8.04) the counter is updated >>>> exactly >>>> when the song ends. >>>> IMHO this is really annoying... I just noticed that a lot of my >>>> favourites >>>> songs have the counter at 0 or few more because often when the song it's >>>> almost finished I switch to another one and the counter is not trigged. >>>> Consider for example that a lot of songs have a long ending, like a >>>> speech >>>> or several seconds of silence. >>>> >>>> My suggest is to move the trigger from the end to about 3/4. >>>> >>> >>> I like this idea, but I would suggest maybe at the 90% or 95% point >>> rather than at 75% through the song. >>> >>> Peter >> >> Having used Rhythmbox for quite some time (over 3 years now, I think), I >> would *definitely* support this change. It is frustrating to see very >> popular songs with unfairly lower playcounts. >> >> As a tangent to this idea, maybe the playcount could be stored as a >> double/float? If you played a song 50%-75% of the way through, then the >> playcount could be incremented by .5 If you played a song over 75%, then >> the playcount could be incremented by 1.0. >> >> Storing playcount as a double could be a lot of code, though... so this >> particular tangent could prove unfeasable. >> >> John >> _______________________________________________ >> rhythmbox-devel mailing list >> [email protected] >> http://mail.gnome.org/mailman/listinfo/rhythmbox-devel > > > _______________________________________________ > rhythmbox-devel mailing list > [email protected] > http://mail.gnome.org/mailman/listinfo/rhythmbox-devel > > -- Alan Godoy Souza Mello Engenheiro de Computação LBiC - Laboratório de Bioinformática e Computação Bio-Inspirada DCA / FEEC / Unicamp _______________________________________________ rhythmbox-devel mailing list [email protected] http://mail.gnome.org/mailman/listinfo/rhythmbox-devel
