[tei-council] <content> vs <mixedContent>

[Sebastian Rahtz]

On 4 Oct 2014, at 23:26, Lou Burnard <lou.burnard at retired.ox.ac.uk> wrote:

>> <textNode> is digging its own grave, IMHO.
> 
> It's about to hatch and fly away, you mean.

yes please.  if you want text nodes, specify them using macros like a christian,
and change these macros in the normal way if you need.

but this leaves the same problem with <mixedContent>. what does the
implicit “text" in there actually expand to, and can the user override it in any way?
if it expands to use “#PCDATA", then it breaks our convention about 
using macro.xText wherever possible, and would be distinctly
disrecommended in a TEI context.. If it expands to macro.xText, then:

 a) is it literally an expansion, i.e. do changes to macro.xText affect it?
 b) how does the user say they really just text? (and how can they choose string or token? cf Eric)
 c) macro.xText is a purely TEI artefact, so does it expand to something different
     in a simple new schema?

Again, I don’t like the idea of some kind of magic rules embedded in the ODD processor.
So it has to be that <mixedContent> means rng:text only, and that people must
explicitly add in <classRef key=“model.gLike”/> to be conform ant.

I am puzzled about why we didn’t consider all this about the lack of clarity
over what “text” is in the summer in Oxford
when we invented <textNode>; or, if we did, why we don’t seem to have
written it down…


--
Sebastian Rahtz      
Director (Research) of Academic IT
University of Oxford IT Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

Não sou nada.
Nunca serei nada.
Não posso querer ser nada.
À parte isso, tenho em mim todos os sonhos do mundo.