[vorbis] TAG Standard - ENSEMBLE/PERFORMER tags

Glenn Maynard g_ogg at zewt.org
Fri Jan 4 00:25:39 PST 2002



On Thu, Jan 03, 2002 at 01:57:51PM -0800, Craig Dickson wrote:
> My mistake; I meant not in the tags. Given that the size is not really
> much of an issue after all, I agree with Jonathan that the metadata
> would be the place for this.

OK, we're in agreement here.

> I wasn't assuming that we weren't using UTF-8; the question was whether
> you wanted the Chinese title in actual Chinese glyphs or in a
> Romanization. I didn't make that clear enough. The advantage of
> Romanization is that it is at least minimally readable by most computer
> users around the world. (Of course, then you have the question of which
> system of Romanization to use; for Chinese, at least, there are
> several.)

I agree that a romanization tag is at least worth consideration.  I
don't think it's worth bothering with which system--people are going to
use what system they want (there are multiple romanization systems for
Japanese, and probably all non-Roman alphabets, and if a user is
romanizing it, he's going to use his preferred system, not what a
standard they probably didn't read wants.)  

It's possible to do romanization automatically (for Japanese and
probably Chinese and Korean; I don't know about others like Hebrew.)
For CJK, it requires a non-trivial kanji database and some work, so it's
not something a player should be expected to do.

> Of course, you started talking about all this stuff as "tags", and only
> accepted Jonathan's suggestion that it should be metadata after my reply
> to you had been posted, so obviously I could not at that time have been
> aware that you were going to accept the idea.

Because I read anything that is storing this type of data as a "tag",
whether it's in a simple comment block or a metadata stream.  I only
began differentiating in here when it became important.

> As a metadata issue, this can be deferred until someone comes up with a
> real proposal for the format and usage of the metadata (currently,
> AFAIK, there is none). If we agree that it's not in the tags, it's not
> an issue that needs to be settled today.
 
I think discussion of what the metadata stream should handle, what its
format should be, and so on should take place before anyone spends time
actually putting anything together.  That's what I'm interested in
starting.

> If you think I haven't been reading it, then I'd say YOU haven't been
> reading it. I've been an active participant in threads on this topic for
> months. In contrast, your first message to this mailing list appears to
> be dated 17 December 2001, according to my archives, unless you were
> posting under another name previously. I don't see anything prior to
> that from you this year under the name "Glenn Maynard" (or, for that
> matter, anyone with "Glenn" or "Maynard" in their "From:" header).

I had already agreed to the distinction between the trivial tags header
and the metadata stream well before I received this message; that's what
it seemed you had missed.  (Of course, you may not have received those
messages yet; SMTP can do that.)


-- 
Glenn Maynard

--- >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.




More information about the Vorbis mailing list