Re: suggest validator prefer URI to FPI

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dan Connolly <connolly@w3.org> wrote:

>In the validator catalog […] we have
>
>PUBLIC  "-//W3C//DTD XHTML 1.0 Strict//EN"
>        "REC-xhtml1-20020801/xhtml1-strict.dtd" […]
>
>I suggest adding entries that key on the URI before those entries,

To what purpose?

The number of document instances that reference an external subset by a system
identifier without also including a public identifier is insignificant; so the
dubious benefit of the miniscule amount of bandwidth conserved is more than
offset by the sharply increased maintenance cost of doubling the size of our
(already quite bloated) catalog files.


>Case in point... this page <http://www.w3.org/2001/sw/BestPractices/>
>$Revision: 1.78 $  […] has a somewhat curious document type delcaration:
>
><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
>  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

Yes, the author of that page is clearly in error. The result of an FPI and SI
referencing different external subsets is at best undefined. I suggest you
notify the page's author of this issue so the erroneous SYSTEM identifier can
be corrected.


>The validator seems to recognize the given FPI and ignore the URI:

The Markup Validator will prefer the PUBLIC Identifier to any SYSTEM
Identifier, yes.


>I expected it to check the document agains the transitional DTD;

Then your expectation was clearly unwarranted.


>that is, I expect it to use the URI that was provided, since FPIs don't
>give the same network effects

I'm quite certain I disagree with the argument put forth there and in the
referenced document, but in any case I think any postulated relevance of
economic theories applied to technology are entirely irrelevant to the issue
at hand.


>Hmm... I wonder if it's wise to override relative URIs with entries
>like...
>
>SYSTEM "spec.dtd" "http://www.w3.org/XML/1998/06/xmlspec-v20.dtd"

No, probably not. It was however necessary in order to prosess the "spec-prod"
XML versions of W3C Recommendations as these typically (at the time that entry
was added) did not choose to make use of a PUBLIC Identfier and included a
relative SYSTEM Identfier referencing a location that did not exist.

Hmmm. Is this the "Network Effect" you were thinking of? That noisome negative
part of the whole "network effect" hype that so many dotcoms failed to take
into account? If that's the case, then sure, prefering a SYSTEM identfier to a
PUBLIC identfier does have several negative effects such as the above.

This is clearly a widespread problem judging by the W3C Recommendatations that
encourage authors to reference a DTD with a relative SYSTEM identifier in
documents not destined for publication in the www.w3.org URL space.


Thanks for your feedback on this.

- -- 
When I decide that the situation is unacceptable for me, I'll simply fork
the tree.   I do _not_ appreciate being enlisted into anyone's holy wars,
so unless you _really_ want to go _way_ up in my  personal shitlist don't
play politics in my vicinity.                   -- Alexander Viro on lkml

-----BEGIN PGP SIGNATURE-----
Version: PGP SDK 3.0.3

iQA/AwUBQREOVqPyPrIkdfXsEQIfqQCgrlKXuhRrDc4aZi4GC4fUSCKSElsAn37v
OandBHYdNXkJ/q2fWzTHAFzT
=4e6s
-----END PGP SIGNATURE-----

Received on Wednesday, 4 August 2004 12:27:31 UTC