XSL Debate, Leventhal responds to Stephen Deach

[Simon St.Laurent]

At 01:06 PM 6/22/99 -0400, you wrote:
>"Simon St.Laurent" wrote:
>> 
>> At 07:38 AM 6/22/99 -0400, Paul Prescod wrote:
>> >But if you are going to pursue this "disabling the semantic web" stuff
>> >you'll have to show that XSL is somehow better at doing so than the DOM,
>> >which you seem to think is just fine.
>> 
>> Sure thing.  XSL provides a semantic-free vocabulary for representing
>> content.  
>
>One of the very first deliverables of the World Wide Web Consortium was a
>semantic-free vocabulary called "HTML." It has been the target of terabyes
>of dumbed-down information.

HTML has been a tremendous success in that regard, letting people get
started without requiring enormous understanding.

However, HTML was an opening step that made the Web possible (SGML sure
didn't do it), giving the world exposure to the use of marked-up text for
human-readable hypertext creation.

HTML's limitations became very clear fairly quickly.  The agent developers
smashed into it early on, having to give up their hopes because every site
looked totally different and it was too hard to figure out what information
was what.  Regular Web developers found themselves ensnared in browser
incompatibilities as competitive pressures created incompatible dialects.
Turning to plug-ins, incompatibilities between browsers and plug-ins and
plug-ins and information sources became enormous issues.

XML was supposed to remove these obstacles and _create_ the possibility of
a semantic Web.  Agent developers would at least get reasonable hints to
figure out what information was what.  Web developers could escape the
conflicts of browser-developer-mandated vocabularies and turn to formatting
standards that would let them present information without having to
incinerate the underlying meaning of the documents.  Plug-ins might still
have issues, but at least data formats could be unified under an open
format that could avoid the ugliest problems of their predecessors.

This dream was just getting started when it became clear that a tool that
was supposed to be a key part of XML's march forward could, in fact, be
used to go backward, into formatting domains that make HTML look
meaningful.  Not only that, but the entire architecture of the tool made
such backward transformations easy and recommended practice.  By putting
half the spec's support on one side of a transaction and the other half on
the other, one side could get all the intelligence and the other side some
pretty-looking babble.  

XML is now facing a U-turn, before it had even really marched out on the
Web.  If you disagree, that's fine.  Just don't try to claim that "XSL is
no worse than HTML and we should all love it because of that."  XML was an
attempt to do better than HTML, and these claims ring hollow.

>I hereby mourn the passing of the semantic web. We hardly
>new ye! As part of my mourning period I now begin a three minute boycott
>of the above technologies and a permanent boycott of threads discussing
>XSL's impact on the "semantic web."

Then I suppose the semantic Web must be stillborn, as it hasn't arrived
yet.  As for your permanent boycott of discussion, that may be sensible for
an issue where both sides believe their arguments too strongly to give ground.

Simon St.Laurent
XML: A Primer / Building XML Applications
Inside XML DTDs: Scientific and Technical (July)
Sharing Bandwidth / Cookies
http://www.simonstl.com

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/ and on CD-ROM/ISBN 981-02-3594-1
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)