[tei-council] Looking ahead

[Syd Bauman]

Having been one of the major proponents of co-occurrence constraints
over the years, I can't believe I'm going to say this.

My gut instinct is that this is a P6 feature. The implications are
pretty significant. First off, it's hard to imagine doing this in a
backward-compatible way (although I suppose it's possible, it would
almost assuredly be ugly). But moreover, we don't even support
interleave because there is no equivalent in DTDs -- what on earth
are we going to do with co-occurrence constraints? And lastly, if we
had co-occurrence constraints available, we should be using them.

Lastly, this would be a significant, if not major effort, distracting
us from other endeavors, and requiring resources that perhaps should
be applied elsewhere.


> Does the idea horrify or excite you all?

The ideas of re-working ODD and getting co-occurrence constraints are
both very exciting. The idea that we might again disregard the
membership's explicit desire that we focus on training materials,
documentation, outreach, etc. in order to do this now is horrifying.


If you want to work on ODD implementation stuff, though, Sebastian,
supporting alternation between groups of attributes (i.e.
      <attList org="choice">
        <attList org="group">
          <attDef/>
          <attDef/>
        </attList>
        <attList org="group">
          <attDef/>
          <attDef/>
        </attList>
        <attDef/>
      </attList>
or such.) is still on the docket.