[
Lists Home |
Date Index |
Thread Index
]
Ann Navarro wrote:
> My problem has always been that to use named character entities in an
> XML Schema-based language, you have to use a DTD *as well* to make this
> happen.
>
> This makes XML Schema mind-bogglingly annoying as an incomplete solution
> for a language definition, that we have no choice but to use based on
> political pressure.
First, in background I would say that HTML is not an XML Schema *based*
language, the semantics of HTML are described in written prose and
implemented in compiled code, the schema provides error checking and
redundant expression of some of the constraints.
Second, I agree that it's annoying that there's not one place you can go
to write down everything about your language.
However, the argument on the other side is reasonable. The constraints
that govern which elements and attributes can go where and what kind of
values they can have are like fish, and the macros for giving names to
hard-to-write data are like bicycles. Just because DTDs did both fish &
bicycles we all kind of started to think that these kinds of things
belonged together. But really, they don't. You wouldn't reasonably ask
that your schema also functioned as your stylesheet, or your compression
engine, or any number of other things; it's just far from obvious that
funny-character-naming is a thing that a schema should do.
I wouldn't be surprised if a dozen years from now, when you do a
language, you end up doing a schema, and some stylesheets, and some
other things that we don't know about yet, and a DTD fragment for naming
characters. Sigh. -Tim
|