Re: [mb-users] AcoustID issue

On Sun, 2012-09-23 at 23:50 +0200, Frederic Da Vitoria wrote:
> 2012/9/23 Nicolás Tamargo de Eguren <[email protected]>
> >> http://acoustid.org/track/36963f6c-8cc9-4225-b9c5-1a3b684cfa0b displays
> >> one track with a completely different duration and a different performer. I
> >> thought that AcoustID "knew" the track duration. Is the
> >> 3ddf7440-bfb5-4dd0-8cec-33229efe9de5 wrong (and if so how could this
> >> happen?) or is this kind of collision possible with AcoustID?
> >
> >
> > I'm not that familiar with acoustID but it seems like a case of someone
> > having the wrong tags and submitting the info - I'd just unlink the
> > Nikolayeva since this looks to clearly be the Gould.
> >
> 
> I guess you are right (if collisions can happen, what are the chances it
> would happen with the same composer?) and I'll do it. I am still interested
> in knowing if AcoustID indeed "knows" about recording duration, this would
> be a useful information when stumbling on similar issues.

In cases like this with mistagged files, doing something to find
Acoustids matched to multiple recordings with different lengths on
Musicbrainz could be interesting. It would have to a fairly loose match
however, given the types of inaccuracies that sneak into our Musicbrainz
recording lengths... and the largish number of recordings that simply
don't have lengths set at all.

-- 
Calvin Walton <[email protected]>


_______________________________________________
MusicBrainz-users mailing list
[email protected]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users