[tei-council] multiple namespaces in Tite documents

[Kevin Hawkins]

Thanks for straightening me out on terminology.

On 9/26/2010 12:37 PM, Lou Burnard wrote:
 > Mandating a specific namespace prefix seems like a rather bad idea to
 > me. Using the same one consistently in examples however is perfectly
 > reasonably

(Right now examples do not use any namespace.)

On 9/26/2010 12:39 PM, Sebastian Rahtz wrote:
 > sure, if you are using a RELAX NG schema, all bets are off and you
 > can do it how you like, with or without prefixes.

Huh.  <oXygen/> XML Editor 10.0 (build 2008120916) wouldn't let me!

 >>   I believe
 >> Tite is the first time since the introduction of the namespace mechanism
 >> in P5 the TEI has declared elements outside of the TEI namespace, so I
 >> want to think through the question of how to best document these things.
 >
 > is it different from the MathML or SVG or RELAX inclusions. tho?

I think so, because we're including from a namespace we manage.

 >> Still, is there any sense in prescribing a namespace -- the string
 >> before the colon, not the URI where the namespace is documented
 >
 > Well, no. Sorry. them's the rules. The namespace prefix is an arbitrary
 > convenience-only  binding of a namespace
 > (which is definitely NOT, by the way, "the URI where the namespace is 
documented"!)

I realize the prefix can be arbitrary, but if we prescribe use of a 
particular one, we reduce barriers to interchange of Tite documents.

However, this may all be moot ...

On 9/26/2010 12:40 PM, Sebastian Rahtz wrote:
>> Personally I'd be much happier if we were able to roll Tite into the
>> main TEI
>
> amen to that......

Lou said rolling Tite into the TEI is for future work, but I'm trying to 
document such future work in SourceForge right now.  So no need to wait! 
  This is now a feature request:

https://sourceforge.net/tracker/?func=detail&aid=3075996&group_id=106328&atid=644065

In light of this request, let's hold off on supplying any namespace in 
the examples.

--Kevin