After loads of editing and discussion on this list since NGS started I
still can't see any clear direction. People have very divergent ideas
about what we should make of NGS, and the style guidelines aren't
providing much guidance.

Before diving into improving small parts of the puzzle I think we
really need to reach some kind of conclusion about what level of
normalization we want and where. Points to consider:

 * difficulty of entering new releases in line with the guidelines

 * for a huge portion of our data, we will never have the chance to
see the cover again, so it will remain as it is now (normalized)

 * usefulness of the data for tagging

The point where people seem to disagree most is the release-level
data: release titles, track titles and track artists. For the purposes
of discussion, I'll make an initial "no normalization" suggestion to
work from:

"Release-level data should be entered as close as possible to what is
on the release. All captialization should be preserved. All typos
should be preserved. Anything written in proximity to the track title
should be entered, no matter how long-winded or irrelevant. Track
artists should only diverge from the release artist if there is a
clear split between titles and artists on the cover and the way the
artist is credited can be recreated exactly with an artist credit.
Otherwise, the artist should be part of the track title."

http://wiki.musicbrainz.org/Style/Principle needs to be rewritten for
NGS, since there can't possibly be artist intent for both the release
level and the recording/work levels. I'm suggesting (in a devil's
advocate kind of way) that it updated to define release-level data and
say that "No normalization whatsoever should be applied to
release-level data".

Now, please disagree, and tell me what we should do instead.

-- 
Philip Jägenstedt

_______________________________________________
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Reply via email to