xml:id RE: [sc34wg3] Compact syntax requirement question

Robert Barta sc34wg3@isotopicmaps.org
Wed, 20 Jul 2005 15:52:46 +1000

On Tue, Jul 19, 2005 at 02:56:44PM +0200, Bernard Vatant wrote:
> >| 	CTM should be easy to validate
> >
> > That's an interesting requirement, but I'm not sure exactly what you
> > mean by it. Validate on what level? Syntactically? Or against a schema?
> I'm amazed by those questions. If you specify a language, I guess you provide ways to
> check if the files you produce are conformant to the specification. Call it well-formed,
> valid, whatever....

Hold on, hold on.

The background is:

  - every (formalized) language has a structure (syntax). Every string
    (one-dimensional languages) follows this syntax.

    There are languages to express this syntax. EBNF (argh) is one of them.

  - every (formalized) language has a meaning (semantics).

    Also here there are languages, depending on the complexity or computational
    model underlying the language.

So 'well-formed' is merely a syntactical conformance where a string
conforms to a syntax, but not every syntax-conforming string might be
a correct string (e.g. 'type correctness' if the language understands
types could be violated).

So we should be now on the same page.

> ...................., in any case : when I get a file "foo.ctm", how do I make sure it's
> conformant to the CTM specification? What kind of tool do I use? When I have an XML file,
> I know that I have two possible levels of validation (at least), and the ways to check it
> in my XML editor.

Could be a pretty bogus argument, because I always wondered:

Conjecture [A]: Is it possible to construct an XTM file which
                validates against the XTM DTD and which 'violates'

What about

<association id="a1">
   <topicRef ....href="#a1"/>

In this case, for instance,




   Contains the IRI reference that is the topic reference. This IRI
   reference shall conform to the requirements of XLink and have a
   fragment identifier which shall be what [W3C XPointer] calls a
   shorthand pointer (formerly barename).

[ Ah, I found accidentally a typo!! ]

So far, so good. And TMDM



   [player]: A topic item. The topic that plays this role in the association.

And if the XTM serialization rules



   ..... If no such topic item exists, a topic item is created, and l
   added to its [item identifiers] property.

then I wonder whether this all works.


If conjecture [A] is true, my argument, then, would be "if XTM (+
DTD/Relax/Schema) cannot guarantee that my content which I put in with
this nifty-hefty-trendy XML development thingy is TMDM-sane, how does
XTM editing make sense at all?"

Hope I am wrong with my conjecture [A]. ;-)