[vorbis] Tag Proposal -> Tag Standard

Thu Apr 25 01:15:40 PDT 2002

"David K. Gasaway" wrote:
> 
> Alright.  So, the tag proposal that has been labored over for some
> months has now been reclassified a bit.  There new URL is now here:
> 
> http://reactor-core.org/ogg-tag-recommendations.html
> 
> And is prefaced by the following statement:
> 
> "The following recommendations were developed by a community of
> Ogg/Vorbis users for their own use."

So far, so good.

> I believe the implication is that this document is not, and will not,
> be endorsed by xiph or the vorbis team.  This disturbs me a bit, and
> think there should be some discussion on the matter.  Here are some
> points which I feel are pertinent.
> 
> 1) The v-comment.html document (the "Standard") on xiph.org is a
> suitably weak standard.  It was entirely appropriate for the time at
> which it was published.  As I see it, the intent was to develop
> something which at the outset was clearly superior to id3.  This does
> not preclude development of a more concrete standard.  In other words,
> the plan was to leave the Standard was sparse only until a more
> complete standard could be devised and reviewed over time, concurrently
> with the development of ogg vorbis.  The Standard was simply a
> framework on which to build something better.

No.  The Vorbis comments have one goal: provide human-readable comments.
Quoting from the standard:

td> The comment field is meant to be used much like someone jotting a
std> quick note on the bottom of a CDR.

The standard fullfills that goal perfectly.

> 2) As far as traffic to this list, it seems very little has been
> decided about a new meta-data format which would supplant comment tags.
>  Let's see - it won't be in XML; that's about all I can recall from the
> discussion.  In other words, it hasn't come along very far.

Then maybe you should focus on working on it?

> 3) The inadequacy of the id3 system is well-known.  In particular,
> classical music is not given fair treatment.  And so, we have id3v2, a
> conflicting and/or supplemental standard.  While id3v1 support is
> widespread, id3v2 support is still spotty.  Some applications which
> *can* use id3v2 prefer id3v1 when given both! *cough*winamp*cough*
> It's a mess.

How is this relevant to us?

> 4) The need for a concrete standard is very real.  The average user
> needs to be given clear direction.  Hard-core ogg vorbis users have
> been around; they've tried id3, hated it, and have had excellent ideas
> about a superior standard.  I think that's evidenced by the lively
> discussions that have deluged the list from time to time.
> 
> The average GUI, point-and-click user, on the other hand, doesn't have
> the will or inclination to personally develop a tag standard.  It's
> clear that the questions are not always obvious and the answers are
> never clear cut - check out the archives, man!  The user will want
> boxes to fill in and a button labelled "Save".

Let them type anything they want -- because they want just that.

> 5) In the end, without a definitive standard, ogg tags will likely be
> left in the hands of GUI developers.  Why is this so bad?  Well,
> imagine MusicMatch decides to implement a certain set of tags, and
> lacking any good external documentation to reference, they develop
> there own "Ogg Standard" which is published to their web site.
> Immediately they become a large, influential force in the Ogg Vorbis
> community.  Again, lacking good documentation to reference, P2P
> developers latch on to MusicMatch's standard as *the* way to tag oggs.
> :P

Any program that doesn't show *all* tags is not compliant.  If a program
restricts the user in what tags he can input, users will not like it.

> 6) Developers, being the sticklers that they are, will probably not be
> comfortable with the Standard.  Discerning developers, anyway, who
> recognize the short-comings of the id3 system.  They will want
> something concrete on which to build code.  Not an obscure, idealistic
> dream. :)

Erm, are you bashing the proposal now, or the standard?

> 7) The existing Standard, even to the hard-core user, is only useful as
> a framework developing a personal standard.  Those who will make oggs
> and keep them for themselves might be perfectly satisfied by this
> situation.  But it does nothing to build and support an Ogg Vorbis
> community.

Please explain.

> 8) I, as a user, am not interested in an audio format that does not
> provide a reasonably robust system for identifying music.  I have
> avoided encoding *any classical music at all* to the MP3 format.
> Instead, I have been holding for Ogg Vorbis.  I'm sure there are many
> others that feel the same.

If you have a Vorbis file, looking at the comment field will show you
what it is.
If, on the other hand, you want to search for a certain file, you should
use some library program.  A database like that restricts the user in
what data (what tags) it can hold, but for a good reason: to make precise
searches possible.  The comment field does not need any restrictions
like that.

> 9) Any concerns about the Proposal as a whole should have have been
> brought forward *long* ago.  Long before so many people have
> contributed so much time to develop the Proposal with the understanding
> that it would eventually become the "New Standard".  This work has been
> conducted in public on the vorbis list, and always with the overriding
> goal of developing something which benefits the Ogg Vorbis community as
> a whole.

a) We didn't invite any of that discussion
b) Concerns _have_ been brought forward, they just have been ignored.

http://www.xiph.org/archives/vorbis/200112/0083.html (me)
http://www.xiph.org/archives/vorbis/200112/0401.html (Rillian)
http://www.xiph.org/archives/vorbis/200112/0162.html (Monty)
http://www.xiph.org/archives/vorbis/200112/0138.html (Monty)

and many many more.

> 10) Overall, I feel the Proposal is nothing but beneficial to users,
> developers, xiph, the entire Ogg Vorbis community, the Open Source
> community, humanity, etc.  I haven't seen any concrete contrary
> arguments.  In all honesty, I'd love to see them.  I'm always up for a
> little debate. ;)

I feel it harms users and developers.  Simpler and more general is
easier and more powerful.  Standardizing "ARTIST", "TRACKNUMBER" etc.
is good for easy adoption of Vorbis in older software that is id3 or
CDDB-centric.  There is no need to standardize any other tag.

<p>And now, back to your regular scheduled program...

<p><p>Segher

<p>--- >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.