[
Lists Home |
Date Index |
Thread Index
]
Tim Bray <tbray@textuality.com> wrote:
| Arjun Ray wrote:
|
| > Correct. The spec is silent on this, and thus useless for the general
| > problem of vocabulary combination *by syntax*.
|
| Arjun is really getting close to troll territory here,
Speak for yourself, and use a killfile if it helps.
| but this very strong claim should not go unrefuted.
I submit that you haven't thought enough about the problem.
| The syntactic problems of combining markup from multiple vocabularies
| include:
|
| 1. avoiding collisions
A non-problem. The premise underlying this is false. (It's the same
whopping non sequitur as in the "Motivation" section of the spec.)
| #1 is easy,
Yes, but not a easy as it appears.
| and namespaces solve it.
In a silly and myopic sort of way, yes. That's what you get with
non-sloutions to non-problems.
| That's all they've ever done, and all they've ever been claimed to do.
True, as far as the spec is concerned. It's when people try to read more
into the spec (as the "Motivation" section would suck them into) that the
real problems start.
| 2. expressing syntactic restrictions on the markup from each individual
| vocabulary
| #2 is harder, and is addressed, if not completely solved, by DTDs and
| their successors.
Yes, the notion that names from a vocabulary can be used (in conjunction)
in some ways but not in others. But that isn't the concern here, which is
with instance markup only.
| 3. expressing syntactic restrictions on the way the vocabularies
| combine
Not necessary to definitions of vocabularies. Necessary to authors for
individual documents.
| #3 is harder still, and is addressed, if not completely solved, by more
| more modern schema languages like XSD and RNG.
It's an interesting problem from the perspective of derivative schema
construction. But the premise that a document must always be in its
entirety an instantiation of a unitary encompassing schema is false. Or
more accurately, unnecessary.
| 4. (maybe) expressing rules for how to extract vocabulary-specific
| subsets of the combination. But this is maybe getting into
| semantic territory.
| #4... nope, I think that's a semantic ingegration problem.
No, the extraction procedure can be entirely generic, at the parsing
level.
|
