[sc34wg3] Almost arbitrary markup in resourceData

[Lars Marius Garshol]

* Jason Cupp
|
| I'm +1 too, but I can't vote...

You can speak out, at least, and that's worth something, too. This is
in any case not the formal ISO vote.
 
| If you can put anything at the end of an <resourceRef>, then it
| doesn't make sense to restrict a <resourceData> to PCDATA -- even
| without use-cases, this seems straightforward.

That's part of the rationale, yes.
 
| What makes sense for me then, is to allow XTM to play well with the
| current state-of-the-art, which is any XML + namespaces. 

Yep. That's what we are planning.

| Why do so many people dislike namespaces?

Because although they do a useful job they do so in a fantastically
ugly way. For more detail, read the archives of XML-DEV, where you'll
find years of arguments about this.
 
| Instead of new XTM tags to indicate content <resourceXML>,
| <resourceBase64>, you could establish PSIs to use as occurrence
| types:
| http://www.topicmaps.org/xtm/1.0/#xml,
| http://www.topicmaps.org/xtm/1.0/#base64, other wise a processor would
| assume PCDATA -- wasn't there someone doing PSIs for mime-types?

This is one possibility, though you don't need any flags to know
whether it is XML or not. If there are no elements (or PIs, I guess)
in the <resourceData/> element then it's not XML, and all is well.

Base64 is different, but we haven't made a decision on whether to
support that. It's just been suggested by Dmitry. A PSI-based solution
for indicating base64-interpretation of the content might work.

Anyway, thank you for speaking out!

-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >