[tei-council] facsimile diagram

[James Cummings]

Yup, I agree with what Conal says below.  I think it a shame that a set of
xpointer schemes to point into all different sorts of media hasn't been
done.  The ISO standard in question is 21000-17, I believe and only covers
MPEG.  I think it would be great if the TEI had a single similar xpointer
scheme method for pointing into multiple types of non-XML media.  However,
I don't think this is realistic or really advisable for P5 1.0.  If ISO/W3C
do more in this area, it would be changes which are backwards compatible
and no problem to include in P5 1.x sometime in the future.

-James

Conal Tuohy wrote:
> Dot wrote:
>>> James reminded me that there is work in the MPEG/ISO/W3C communities
>>> on defining new Xpointer schemes to point into assorted kinds of
>>> multimedia which we should at least acknowledge somewhere.
>>> 
>> Conal had looked into this several months ago and had I believe 
>> included it in a very early draft of the plan, but at that time the 
>> multimedia Xpointer scheme was only for pointing into audio/video 
>> files and not into still images (which would have made our work much 
>> easier). This may have changed.
> 
> I think that's right, and I still believe that using URI fragment
> identifiers for pointing into images is not an useful approach.
> 
> A particular URI fragment identifier syntax is defined only for a
> particular set of internet media types.
> 
> So HTML has a fragment identifier syntax in which the identifier is a
> pointer to either an @id attribute, or the @name attribute of an <a>
> element. XML has a complex syntax in which the identifier refers to an
> element with a matching @xml:id attribute, or it can use XPointer. SVG
> has that, too, and also has an "svgview" alternative.
> 
> But these different syntaxes apply only to HTML, XML, and SVG
> respectively; other media types need to have registered their own
> fragment identifier syntax, and neither JPEG, TIFF, GIF nor PNG images
> (I am pretty sure) have such a syntax defined. The MPEG group's pointer
> syntax applies only to MPEG video (or did, when I looked into it).
> 
> I believe that the facsimile markup should therefore not rely on
> fragment identifiers, since at best they will be too varied to be
> helpful, and at worst, they will simply not be defined for many useful
> image types. _______________________________________________ tei-council
> mailing list tei-council at lists.village.Virginia.EDU 
> http://lists.village.Virginia.EDU/mailman/listinfo/tei-council
> 


-- 
Dr James Cummings, Oxford Text Archive, University of Oxford
James dot Cummings at oucs dot ox dot ac dot uk