[vorbis] Re: The authority of Jonathan

[Monty]

On Sat, Dec 08, 2001 at 09:20:28PM -0800, Craig Dickson wrote:
> > Monty has final say.  Theres never been any question about that.
> > 
> > When I have rejected peoples suggestions flat out, its not because I am
> > saying it, it is generally because Monty has already said, in email or
> > on IRC, "no way!".  I have been working within the constraints of what
> > Monty has already said he would accept.
> 
> You may have a clearer sense of Monty's views than I do; he hasn't said
> much in this discussion. 

There have been many previous rounds of this discussion.  Also note
that when I don't respond it's because I get several hundred emails a
day and am also swamped by a very late rc3.  I don't have a few hours
to spare every day to the flaming.  

I am watching, and I will make comments here in the near future; some
of the suggestions have been good.  Most have not for some immovable
reason.

> > All proposals for keeping the
> > ARTIST tag have violated his dictum about subtagging and imposing
> > structure on the tag data.
> 
> Not all of them. Some people have simply rejected your idea of having
> separate PERFORMER and ENSEMBLE tags, and letting ARTIST remain in their
> place. (Aside from the name, this is essentially the same as just
> dropping ENSEMBLE from your proposal.)

It is also the case that we're looking for a happy, informal, system
that is perhaps erring *slightly* on the side of overkill, and that
most people can ignore safely.  Remember; the vast majority of these
tags will not be populated beyond what an automated ripper gets back
from a CDDB or FreeBD query.

> > When there is a sensible proposal, and a sizeable number of people have
> > sniffed it over and said "this is good", its likely Monty or Jack will
> > quietly stick it up on the website and say "this is a standard we
> > approve of".
> 
> That's what I expect too.

It's gone a bit too far astray for that, but the original statement of
'the current crop of tags is too pop centric; here's a few more people
can use' is one I agree with.  The idea is neither to formally
standardize tags people *must* use, or try to modify user behavior to
fill in 25 tags correctly; it's to make a list of tags such that
several unconnected, motivated groups who *do* want more complete
information, by default, end up using mostly the same things.

I do not want a complex or programmatic tag system with rules like
'presence of ARTIST and absence of COMPOSER shall imply that GENRE is
to be modified to mean...'.  Any such proposal gets vetoed.
Subtagging is an interesting idea, but overboard.  It also cannot be
implemented without changing the 1.0 tag spec.  So it is also veteoed.

> band! You've said previously that a one-man band should be considered a
> PERFORMER, not an ENSEMBLE, so would you say that the new ELO record
> should be tagged as PERFORMER="Electric Light Orchestra"? In which case,
> any file search for ENSEMBLE="Electric Light Orchestra" will find all
> the old songs, but not the new ones, which seems like a bad thing to me.

Let's not go too far with dogma here, or mandate that the tagging
system can't make sense without it on either side of this argument.

Here's another rule of thumb: It should be perfectly reasonable for a
player to support tags no further than dumping them as text to stderr,
and the user should instantly know what it all means. The tags are for
people first.  Make them for machines second.

I'll go into more detail on the original proposal soon; I couldn't
fetch the proposal from the web earlier due to the aforementioned DNS
problem.

Last request; let's please try to drop the egos.  Debate is fine;
heated debate isn't really called for on the topic of tags.

Monty

--- >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.