[mb-style] [Clean up CSG] Classical and Release Language

[Lauri Watts]

On Jan 27, 2008 11:05 PM, Leiv Hellebo <leiv.hellebo at gmail.com> wrote:
> Jim DeLaHunt wrote:
> >
> > Leiv Hellebo-2 wrote:
>
> (Huh? If I am LH-2, then who is LH-1?)
>
> >> Jim DeLaHunt wrote:
> >>> My big obstacle in discussing ReleaseLanguage is that I don't have a clue
> >>> why we're recording that data, what problem it's trying to solve.
> >> I guess you would have problems with distinguishing between Danish and
> >> Norwegian, or between various Slavic languages... But with the help of the
> >> attribute ReleaseLanguage, for the purposes of MB, you don't have to.
> >>
> >
> > Yes, I would have problems distinguishing between Danish and Norwegian.
> > Absolutely. The attribute ReleaseLanguage would tell me which language the
> > release text was in. But why would I care?  If I can't read either language,
> > it doesn't seem to matter much if I know which language it is. I still can't
> > read it properly.  I still get the text, whatever language it is, in my
> > track titles (for pop/rock/everything else outside the CSG), where I stare
> > at it and try to make sense of it.

MB the database is an evolving thing.  Just because we don't have
alternate titles _now_ doesn't mean we never will.  Just because we
don't have a method of display on pages for multilingual artists to
display 'only the albums primarily in Norwegian' doesn't mean we never
will.  Just because we don't currently run reports to pick up
'probably has spelling errors for the chosen language' or 'probably
needs capitalisation fixes' for native speakers to review, doesn't
mean we won't next week.  I regularly use the compact listing of some
artists to check through their albums for typo/capitalisation fixes
language by language, simply to save me having to switch the guess
case settings back and forth after each album.  There's a whole pile
of uses, current and only dreamed of for knowing what language the
track titles are written in.

But here's one immediate and common usage: I know for certain that
several people regularly review open edits for a particular language,
because they are more likely than the average user to spot a silly
typo, bad capitalisation, unlikely collaborations., incorrect artist
attributions and the like.

In any case, it's a relatively painless (in fact, nearly automated)
piece of data to collect for the majority of releases, and it's a lot
easier to come up with uses for data we have, than retroactively
collect it when at some future point we discover we need it.

I have to agree with what was already said though: It is to describe
the language of the track listing, as it's listed in MB (assuming that
the track listing has already been normalised to whatever other rules
apply.)  If CSG needs further language attributes, that's a different
issue.

On that note, even non classical could conceivably use a 'vocal
language' to capture what language tracks are sung in, because english
titles on non-english songs, to name one case, is very common.

-- 
Lauri Watts