Re: Can everyone be happy?

John Cowan wrote:

> "Paul W. Abrahams" wrote:
>
> > The possibility remains,  as I and many others insist,
> > that the namespace name identifies no useful resource at all but is
> > merely a convenient and presumably unique identifier.
>
> How can something be an identifier if it doesn't identify anything?
> To use TimBL's analogy, that is a handle without a pot.

My words: "identifies no useful resource at all".  Your words: "doesn't identify
anything".   I accept the possibility that a namespace name always identifies
something, but the "something" may be entirely useless.  Sure, it's a point in Web
space, but it's like the identifier of the person Squinsonia McMurgelstein; it
provides no useful information that can't be gotten by examining the identifier
itself.

> > So suppose we introduce another attribute, which I'll call `nstype'.
> > Something like
> >
> >     <elt xmlns:a="http://www.sushi.org/squid.schema" nstype:a="schema">
> >
> > would indicate (obviously) that a schema describing the namespace whose
> > namespace prefix is "a" is to be found at
> > http://www.sushi.org/squid.schema.  For a namespace whose name identifies
> > nothing useful, the nstype attribute would be omitted -- and we'd have
> > precisely the current situation, augmented perhaps by the fixed-base
> > convention.
>
> Been there, done that.  The trouble is that a single namespace may have
> many associated schemas

Perhaps a better example would have been:

    <elt xmlns:a="http://www.sushi.org/squid-doc. html" nstype:a="htmldoc">

which indicates that the URI points at documentation in HTML form.

> In the terminology I introduced yesterday, a
> namespace is a vocabulary rather than a language.  TimBL prefers to say
> that it is a super-language, but a language without a syntax seems to be
> a self-contradictory notion.

If the namespace name identifies metadata about the namespace, then nstype
describes what is located, which might be a vocabulary.   If it doesn't identify
such metadata, then nstype should be omitted.   My example of a schema was
probably ill-chosen, since there's also "schemaLocation" to locate a schema.   In
the presence of "schemaLocation", it's unclear (to me at least) what significance
the namespace name might have beyond its power to distinguish this namespace from
all other namespaces.

Paul Abrahams

Received on Thursday, 22 June 2000 19:22:21 UTC