[vorbis] (Classical) Request for Standardization of expanded TAGS

[Craig Dickson]

Jonathan Walther wrote:

> I imagine you think that classical music listeners should just go off in
> a corner and play by themselves?  Maybe they should use their own
> "special" ghettoized version of ogg123 and the other ogg tools?  Maybe
> use their own "special" encoders too?  All hail King Pop!  I guess if
> people want to know what that mystery file they are listening to is,
> they'll jolly well have to guess!

You keep fantasizing about this imaginary conspiracy against classical
music listeners. I've pointed out in the past (check your archives) that
all tags that are meaningful for classical music are also meaningful for
pop music. I was just re-reading the old threads last night, so it's
fresh in my memory at the moment.

The worst problems of your proposal come from it being too complex and
arbitrary. There is zero chance that people will use it in a consistent
way even if they try, and most people won't bother trying. That makes it
a poor standard.

>>This is not going to work well in practice. There are too many gray-area
>>cases. Is "Crosby, Stills and Nash" three PERFORMERs or an ENSEMBLE?
>
>I see no grey area.  The answer is
>
>ENSEMBLE=Crosby, Stills and Nash
>
>because they are known as a group by that name, not "3 guys who occasionally
>sing together".

Actually, they've broken up and reformed, and played in different
sub-partnerships (Crosby and Nash, Crosby and Stills, Nash and Stills,
Crosby, Stills, Nash and Young, etc.) so much that I don't think that
argument is very strong. They've never gone more than two albums in a
row without breaking up or somebody leaving or rejoining.

Additionally, your judgment is arbitrary and lacks obviousness.

>>And what about a "band" like the Durutti Column, which is really just
>>one guy overdubbing everything?
>
>PERFORMER=Durutti Column

You can say that, but I suspect about 50% of their fans who use ogg
would get it wrong, because it's just not obvious enough. It's too
subtle a distinction.

>>And what about The Fireman, which sounds like one person, but is
>>actually duo performing under a "common name"?
>
>ENSEMBLE=The Fireman

Same again. The name sounds singular, so users will treat it that way.

>>And what about something like "Buddy Holly and the Crickets" -- is that
>>PERFORMER="Buddy Holly" and ENSEMBLE="The Crickets"? Logically, you
>
>ENSEMBLE=Buddy Holly and The Crickets

Why? Buddy Holly is really the artist. The Crickets were just his hired
backing band. Another arbitrary, non-obvious decision.

>>So even if you could come up with clean definitions that handle all the
>>gray-area cases, you will _never_ manage to get all the ogg users doing
>
>You haven't shown me any gray area cases.

Yes, I have. You've just refused to admit that they're there. You don't
seem able to accept that these things will not be anywhere near as
simple and obvious to others as they are to you. You think that because
you can decide for yourself what seems right, that everyone else will
reach the same conclusions. You're wrong.

>>And I don't see the
>>point of "standardizing" something if most users won't conform to (or
>>even know about, in most cases) the "standard".
>
>You won't know until you try.

I already know. The scheme is too complicated, the distinctions you're
making between "performers" and "ensembles" are too subtle, and experience
with MP3 trading proves that most users don't bother with tags, or get
them wrong even in the simple ID3 scheme (I've seen "title" and "album"
interchanged, and hard rock songs with "genre=disco", and just about any
other error imaginable -- and I'm not talking about once in a blue moon,
these are _normal_, _everyday_ things), that this has zero chance of
having any impact beyond the microscopically small set of obsessives who
will actually bother entering everything and checking it for accuracy.

Craig

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.