[tei-council] Quoting a TEI object

[James Cummings]

Laurent Romary wrote:
> In many cases, one has to map schemas and describe the mappings by  
> reference to actual objects in the respective encoding schemes. In  
> this context do (should) we have a mechanism to refer uniquely to TEI  
> objects, beyond the reference to the TEI namespace (<author> in http://www.tei-c.org/ns/1.0) 
> , sort of TEI unique identifiers?

As you suggest with your next message I'd think the proper way to cite 
an element, class, or datatype generally is to point to its reference 
page. http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-tree.html

However, we might want to re-examine that... while the hierarchy should 
remain the same, deprecated elements (etc.) disappear.  I have no way 
easy way to point to something at a particular date in time...unless 
maybe I point to it at a particular revision in the SVN source tree?

Pointing to a location in the SVN source tree isn't necessarily very 
pretty but does work and is accurate:

See for example an earlier version of <event>

http://tei.svn.sourceforge.net/viewvc/tei/trunk/P5/Source/Specs/event.xml?revision=231&view=markup

As a cognate example, I can point to any revision of a page in the TEI 
wiki at any point in its history. We should be able to do this, somehow, 
in the official way we are meant to cite TEI sources as changing objects 
in time.

However, maybe an easier solution is to cite:
http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-event.html
but on the generated reference pages like this contain a link to the 
appropriate place in the SVN repository? i.e.
http://tei.svn.sourceforge.net/viewvc/tei/trunk/P5/Source/Specs/event.xml?view=log 

which lists all the revisions.

-James

-- 
Dr James Cummings, Research Technologies Service, University of Oxford
James dot Cummings at oucs dot ox dot ac dot uk