concur

[James Clark <jjc@jclark.com>]

How do people feel about the "concur" element?  I'm having some doubts
about whether it's a good idea to include it.

- It's probably the most experimental feature of TREX, and standards
aren't the place for experiments.  Almost everything else in TREX has
some analogue in at least one other spec.

- We're trying to be minimalist: "concur" I think is on the wrong side
of the 80/20 divide.

- It's quite tricky to implement (especially getting coherent error
messages).

- I think people will want to layer type assignment on top of TREX. 
It's hard to make type assignment work for concur.

- Other schema languages don't have concur or anything like it, so TREX
patterns that use concur will be hard to translate into other schema
langauges.

- The most useful application of concur seems to me to be exclusions;
but it would be easy to add explicit support for exclusions (I'll
address this in a subsequent message).

James