At 03:59 PM 11/1/2002 -0800, Tim Bray wrote:
>Ann Navarro wrote:
>
>>What I'm most interested in is a means of using such a list (whomever
>>defines it) in a way that doesn't require gymnastics like internal DTD
>>subsets. I don't think that's asking for anything outlandish.
>
>(a) first of all, why does it have to be internal? If it's external &
>identified by a well-known URI I assume that processors could have cached
>versions and/or only retrieve when referenced.
OK, the objection really isn't about internal, it's about having to author
them.
>(b) the amount of gymnastics required is going to be about the same
>whether it's DTD or some other syntax, I could devise a bunch of syntaxes
>for saying that &foobar; means ს but none of them are really going
>to be dramatically simpler than the <!ENTITY syntax that DTDs use.
That assumes that you're not just a consumer of a DTD, but an author of one
(and further assumes that one is present). I don't believe those
assumptions are necessarily warranted.
Ann
