[tei-council] gaiji in PCDATA, or <g> in text

[Syd Bauman]

Christian Wittern writes:

> > <localName>        allow <g>
> 
> As I said elsewhere, this is "text only" in my book. Since it is a
> modern, locally defined name made up by the project, there is no
> reason to insist on using <g>, which would only complicate matters
> in this "meta" situation.

OK. I've just reverted this from 'macro.xtext' to 'rng:text'.


> > <eg>               text only
> > <egXML>            text only
> 
> Why would these be text only?  I would "allow <g>" here

Because these are examples of XML encoding or other computer
documents. At lest for <egXML>, where the contents are defined to be
well-formed XML, it definitionally cannot have a character outside of
Unicode in it, so cannot need to have such a character encoded with
<g>. (You may want to give an example of <g>, but this is handled,
like any other example element, by using an element from a different
namespace, e.g. "http://www.tei-c.org/ns/Examples")

I suppose if someone wants to use <eg> to show an example of RTF or
some even more bizarre language, there might be some character in use
that is not actually a Unicode character?