JJZolx wrote:
But the idea of multiple libraries is something that wouldn't have to depend on user profiles. It might be a browse level above Browse Music.
true, it doesn't have to depend on user profile, but it helps. some of the user issues that were driving the requests for multiple libraries would be automatically solved if libraries were attached to users. i'm thinking of the kids music polluting the parent's library, adult themed music being accessible to kids, roommates who don't want to expose their libraries to each other.
i guess in my mind the concept of library has more to do with ownership and management responsibility than anything else. it's /personal/, kind of like the data on your iPod, cellphone or PDA.
given this, the idea of having your own music directory makes sense. OS-level rights to that directory can easily exist. differences in tagging and format preferences can be honored. and segregation in the web interface where playlists and favorites can be managed also seems to make sense.
although libraries could be useful for genre segregation we already have mechanisms for that.
The question I'd have is how do you designate what is contained within a given library? Setting different music library root folders is one way, but would require quite a bit of creative use of links/shortcuts unless you accepted having multiple copies of a track in each library to which it belongs.
overlap may well occur, but could be managed a number of ways. as you point out, rather than a copy being placed in more than one library aliases could be used. as long as SS was adept at recognizing and dealing with these it shouldn't pose a problem. alternatively the duplicates (and near duplicates) could either be presented distinctly or merged in the UI by some sort of rule.
One approach I could see is having a "master" library with a single music folder like there is now, then creating a web interface where you could build a library by browsing the music directory and saying "Add to Library". This might also make (re)scanning easier, although when doing a scan you'd have to maintain library db links so that work done to create the custom libraries aren't lost.
the problem of persisting the library-to-track association on rescan and wipes is one thing that got my thinking headed in the direction of directory based libraries.
--rt _______________________________________________ Discuss mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/discuss
