variants and ordering

Dear All,

Herewith my humble attempt to bring some order to a small area of chaos:
variants.

There are 3 major components to a language tag: language, script,
region, each of which may take variants. In addition there may be
variants to combinations of components. Regional information is used for
locale information, and while I can see there may possibly be a need for
regional variants, there is no evidence of their existence yet and
certainly not of their interaction with the other components. So I will
ignore them for the moment and perhaps return to them once a basic model
is established. This leaves 2 components and therefore 3 types of
variants: language variants, script variants and language+script
variants which I will call orthography variants. The key concept I am
proposing here is that variants are categorised, based on their prefix,
as to what kind they are and then may be ordered in the tag based on
that categorisation.

A language variant takes a language only prefix. A script variant
currently takes no prefix, but I propose that they take a script prefix.
An orthography variant would take a language-script prefix (requiring both).

Mark Davis has proposed a canonical ordering of variants within a tag:

Any sequence of variant tags MUST be reordered such that for each variant subtags A and B, A comes before B if

   1. A has no prefix and B has a prefix
   2. or A and B are not ordered by #1, and A is part of a prefix (given
      the previous non-variant subtags) for B

Re: variants and ordering

Martin Hosken scripsit:

> But this has been challenged with the example of
> en-scouse-fonipa. I.e. what people are saying is that language takes
> precedence over script, which makes perfect sense to me. I think the
> argument that one can say wider over narrower is unhelpful here since
> the two variants are from different categories.

Well, maybe it does and maybe it doesn't.  A language variant may be
less of a barrier than a script variant: as an American, I can understand
Scouse fairly well either in spoken form or in more or less conventional
orthography, but deciphering it in IPA (even though I know IPA) would
be much more difficult.

For that matter, the language-before-script rule doesn't even always apply
correctly when non-variants are involved.  For example, the standard
language line in the Scandinavian Peninsula runs north and south, but
the dialect isoglosses mostly run east and west.  So a Norwegian can
probably understand the speech of a Swede who lives at about the same
latitude; but in written form, Danish is for historical reasons much
more accessible than Swedish.

> Based on this preamble, I propose the following:
> 
> Any sequence of variant tags MUST be reordered such that for each
> variant subtag A and B, B comes before A unless any of the following
> are true:

[detailed rules snipped]

Re: Re: Proposals for changes to latest draft

[chair hat off]

At 16:44 07/02/28, Doug Ewell wrote:
>John Cowan <cowan at ccil dot org> wrote:
>
>> 2.2.3 bullet 6 adds the notion that the hitherto private tag Qabx is to be usurped to mean "no written
content".  I agree that the "unwritten languages" script tag does not do the job here, but I deny that it is
proper to steal a private-use tag, and I don't see any need to have such an entity at all.
>
>Using Qabx was my idea, to replace Mark's original proposal to use Zxxx for this purpose.  Zxxx is the ISO
15924 code element meaning "code for unwritten languages."  I take that quite literally to mean
"languages that do not have any commonly accepted writing system," which admittedly would be a rather odd
piece of information to put in a language tag, and which is very different IMHO from "content that happens
not to be in a writing system," such as signed or spoken content.  This was my interpretation, but I
confirmed it with Michael, the ISO 15924 Registrar.

If the ISO 15924 Registrar is fine with a tag "code for unwritten languages",
then he should be fine with another tag "code for material not in a particular script". It would be weird if it
were possible to tag audio tapes in a
language that doesn't have any way to be written yet, but audio tapes e.g.
in plain English wouldn't be taggable (assuming an old-style tagging system
where you have to fill all fields with some data).

Has such a proposal been made? I don't think it's urgently needed, but it would
be way much better than a private tag.

>When an ISO standard that establishes a coding structure provides for private-use code elements, as ISO
639 and 3166 and 10646 and 15924 do, applications that use those standards are generally permitted to
allocate code elements from the private space for their own needs.  For example, CLDR uses the ISO 3166
private-use code element QO for "Outlying Oceania."  We are permitted to do similarly for the Language

Next draft

Addison, Mark, Doug,

The deadline for submission of drafts before the upcomming IETF is
approaching rapidly (early next Monday). Addison, would it be possible
for you to submit an updated draft including the points from Mark's
recent proposal that seem to be without controversy so far?
Doug, your draft is more recent than Addison's, so I leave it to you
to decide whether you want to submit a new draft now or not.

Regards,    Martin.

#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#  http://www.sw.it.aoyama.ac.jp       mailto:duerst <at> it.aoyama.ac.jp     

Re: Next draft

Any changes to 4645bis would be based on:

1.  the new ISO 639-3 data files
2.  changes to 4646bis

I could generate a new draft based on (1), but there is always the 
possibility that (2) might render that new draft instantly obsolete. 
I'd rather see a revision to 4646bis first, and then update 4645bis so 
it covers both (1) and (2).

--
Doug Ewell  *  Fullerton, California, USA  *  RFC 4645  *  UTN #14
http://users.adelphia.net/~dewell/
http://www1.ietf.org/html.charters/ltru-charter.html
http://www.alvestrand.no/mailman/listinfo/ietf-languages

Re: Re: Proposals for changes to latest draft

Is it on to propose a tag simply labelled "audio"?
mg

On 1 Mar 2007, at 04:52, Martin Duerst:

> ...
> If the ISO 15924 Registrar is fine with a tag "code for unwritten  
> languages",
> then he should be fine with another tag "code for material not in a  
> particular script". It would be weird if it were possible to tag  
> audio tapes in a
> language that doesn't have any way to be written yet, but audio  
> tapes e.g.
> in plain English wouldn't be taggable (assuming an old-style  
> tagging system
> where you have to fill all fields with some data).
>
> Has such a proposal been made? I don't think it's urgently needed,  
> but it would
> be way much better than a private tag...

- -
Marion Gunn * EGTeo (Estab.1991)
27 Páirc an Fhéithlinn, Baile an
Bhóthair, Co. Átha Cliath, Éire.
* mgunn <at> egt.ie * eamonn <at> egt.ie *

Re: Next draft

I'll see if I can get to it today or tomorrow.

I may even include some things that are still controversial. It is 
merely a draft, after all, and getting wording place will allow us to 
debate and close the issues. I'd like to see the next draft plus one be 
the working group last call draft if possible.

Addison

-- 
Addison Phillips
Globalization Architect -- Yahoo! Inc.
Chair -- W3C Internationalization Core WG

Internationalization is an architecture.
It is not a feature.

Martin Duerst wrote:
> Addison, Mark, Doug,
> 
> The deadline for submission of drafts before the upcomming IETF is
> approaching rapidly (early next Monday). Addison, would it be possible
> for you to submit an updated draft including the points from Mark's
> recent proposal that seem to be without controversy so far?
> Doug, your draft is more recent than Addison's, so I leave it to you
> to decide whether you want to submit a new draft now or not.
> 
> Regards,    Martin.
> 
> 

Re: Next draft

Fw: Revised Internet-Drafts Submission Cutoff Dates for the 68th IETF Meeting in Prague, Czech Republic

Hi -

fwd fyi

> From: <ietf-secretariat <at> ietf.org>
> To: <ietf-announce <at> ietf.org>
> Cc: <wgchairs <at> ietf.org>
> Sent: Thursday, March 01, 2007 9:00 PM
> Subject: Revised Internet-Drafts Submission Cutoff Dates for the 68th IETF Meeting in Prague, Czech
Republic 
>
> All revised Internet-Drafts (version -01 and higher) must be submitted 
> by Monday, March 5th at 9:00 AM ET.
> 
> Revised Internet-Drafts received after the cutoff date will not be made
> available in the Internet-Drafts directory or announced until on or 
> after Monday, March 19th at 9:00 AM ET, when Internet-Draft posting 
> resumes.  Please do not wait until the last minute to submit.
> 
> Thank you for your understanding and cooperation. If you have any 
> questions or concerns, then please send a message to 
> internet-drafts <at> ietf.org.
> 
> The IETF Secretariat
> 
> FYI: The Internet-Draft cutoff dates as well as other significant dates
> for the 68th IETF Meeting can be found at http://www.ietf.org/meetings/cutoff_dates_68.html.
> 

Re: Proposals for changes to latest draft

Marion Gunn <mgunn at egt dot ie> wrote:

> Is it on to propose a tag simply labelled "audio"?

Not a bad idea.  We've already established that "unwritten" is not a 
script, and of course it's not a language or a region either.

--
Doug Ewell  *  Fullerton, California, USA  *  RFC 4645  *  UTN #14
http://users.adelphia.net/~dewell/
http://www1.ietf.org/html.charters/ltru-charter.html
http://www.alvestrand.no/mailman/listinfo/ietf-languages