SAX and delayed entity loading

[John Cowan]

Simon St.Laurent wrote:

> How is [a notation] any better than a MIME
> type, which does about the same thing?  It is slightly more flexible - like
> a gun with a barrel that can rotate 90 degrees, right at your foot.

It also allows you to escape the dilemma "either private use only,
or published in the IANA registry".  See below.

> So you'd like to use the Web as one giant registry?  I suppose I could
> ponder this, but again, I have serious doubts as to its usefulness.  If
> only on grounds of longevity - /X10/ could go dim in ten years, with no way
> to figure out what happened.

Very true.  Hence the need for URNs, or URLs managed as URNs.
No formal property whatever *guarantees* persistence: persistence
is a property of how humans behave (in the case of MIME, how
Jon Postel behaved and how his successors are expected to behave).

> External X10 data could, however, change its
> identification without some lucky person or program tracking down every
> reference to it in documents to reflect the change.

Again, I believe notations (whether used directly, or mediated through
schemas/DTDs) are far more useful when governing elements than when
describing whole remote documents.

> Putting the notations
> in documents creates a maintenance headache that shouldn't be
> underestimated.  Of course, it depends on whether its easier to change the
> documents or the server providing the extra resources.  Right now, it's a
> lot easier to change the server.

Okay, I'm convinced now that we really need the MIME-types-as-notations
DTD fragment.  Stand by.....
 
> >> Public registries and self-identifying documents seem like the only out of
> >> this mess to me, but so far that hasn't gone nearly far enough.
> >
> >The WWW *is* a public registry, or at least can encapsulate one
> >using PURLs and such.
> 
> See above for longevity issues.  Something safe and centralized, complete
> with version control, seems like the only way out of this nightmare.

Safe central registries can be *created* on top of the WWW (as
the PURL folks are doing) without sacrificing the utility of
ad-hoc-but-globally-accessible universal names for things.
 
-- 
John Cowan	http://www.ccil.org/~cowan		cowan at ccil.org
	You tollerday donsk?  N.  You tolkatiff scowegian?  Nn.
	You spigotty anglease?  Nnn.  You phonio saxo?  Nnnn.
		Clear all so!  'Tis a Jute.... (Finnegans Wake 16.5)

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)