[tei-council] TEI ODD Meeting

[James Cummings]

Hi Council,

This is my initial report to Council on the the TEI ODD meeting 
after DH2012, emailed before the teleconference to save time in 
the meeting.

As you know there was a meeting of those interested in TEI ODD 
who happened to be going to DH2012 in Hamburg immediately after 
the conference. Partly this was because there was a panel on the 
future of ODD and it made sense having such a group of people in 
one place, to ask others to join us, and discuss the future and 
possible developments for ODD.

You can see the abstract for the panel at:

http://www.dh2012.uni-hamburg.de/conference/programme/abstracts/future-developments-for-tei-odd/

And even see the panel itself as it was recorded at:
http://lecture2go.uni-hamburg.de/konferenzen/-/k/13922

Then on the Saturday and Sunday after the conference instead of 
going on excursions a hearty band of people including:

Brett Barney, Syd Bauman, Piotr Banski, Lou Burnard, James 
Cummings, Bertrand Gaiffe, Stefan Majewski, Trevor Muñoz, Brian 
Pytlik Zillig, Doug Reside, Raffaele Viglianti, Ron Van den 
Branden, Peter Stadler, Sebastian Rahtz, Laurent Romary (skype)

sat in a room together and discussed a variety of ODD-related 
topics. I will be coming up with a fuller report and converting 
any proposals to SF tickets in time for the face2face meeting. 
Sebastian's comments on the take-home messages for him were:
    * ODD2 has life in it. we don't need to consider an 
incompatible ODD3 at this stage
    * we _badly_ need that new Roma implementation
    * it probably _is_ possible to replace the dependency on 
RELAX NG with a new set of elements, needs proof of concept
    * the use of <specGrp> and <specGrpRef> is not well-enough 
understood, but can be a powerful technique
    * locally scoped changes are in theory possible, but it needs 
much more thought
    * the Gaiffe Manoeuvre should be formalized, possibly with a 
new attribute or element, as it is clever. I think it justifies 
an element cloning method
    * chaining of ODDS is an interesting technique which needs 
some examples and experiments

You can see these and many more interesting partly-formed 
thoughts and experiments in a set of haphazard notes that we took 
at: http://tinyurl.com/odd-dev-dh2012 which is open to commenting 
by the world if you wish.

All in all I think the meeting was a success and has created some 
suggestions for where ODD might go in the future.

Best,
James

-- 
Dr James Cummings, researchsupport at it.ox.ac.uk
Research Support, IT Services, University of Oxford