[mb-users] extending the tagger for music organization

[Paul Taylor]

Philipp Wolfer wrote:
> On Tue, Jul 15, 2008 at 12:45 AM, Alan <alan at ufies.org 
> <mailto:alan at ufies.org>> wrote:
>
>     Hey all, just wanted to throw a couple of things out about being
>     able to
>     use the tagger a bit more efficiently for collection organization.
>      I've
>     been re-doing my 33k or so tracks and have had time to reflect on the
>     tagger as a music organization aid as well.
>
>
> It might interest you that there was a discussion about this just 
> yesterday on IRC:
> http://chatlogs.musicbrainz.org/2008/2008-07/2008-07-14.html#T09-34-18-414342
>  
> Apart from this I had a discussion about such functionality with 
> Alexander (who is doing the Picard Wizard GSoC project) some time ago. 
> Our main ideas were:
>
> 1. Have a watch folder for Picard, which Picard watches for newly 
> added files which would automatically be loaded. This would make it 
> easy to integrate Picard nicely into the ripping and tagging workflow.
> 2. Let Picard maintain an index of your music collection (located in 
> the watch folder or separate folder?)
> 3. Offer an update utility that goes over your collection performing 
> certain tasks, especially:
>   a: Update the metadata from MB
>   b: Calculate missing PUIDs
>   c: Submit PUIDs to MB
>   d: Load the files not already tagged with Picard into the Picard GUI 
> for tagging
>
> As the MB server should soon support change dates for the data such an 
> update functionallity should become possible without causing too much 
> hits and traffic on the server.
>
> If I find the time I will probably outline those ideas in more detail, 
> any ideas are welcome. Maybe it will get implemented some day :)
>
> Phil

I've come to a similar conclusion with Jaikoz development. The average 
customer wants to be able to access their whole collection including 
tagged files, but they dont want to wait long for the files to be 
loaded. The only realistic way to achieve this is to maintain som kind 
of database or index, (but it shouldn't contain any essential metadata 
that is not also in the files itself.)

Paul