W3C home > Mailing lists > Public > www-talk@w3.org > November to December 2000

Re: Semantic Document Framework(s)

From: Sean B. Palmer <sean@mysterylights.com>
Date: Tue, 7 Nov 2000 23:34:21 -0000
Message-ID: <001c01c04913$42d167c0$e672073e@z5n9x1>
To: "Aaron Swartz" <aswartz@swartzfam.com>
Cc: <www-rdf-interest@w3.org>, <www-talk@w3.org>
> Formation, perhaps, but the content has to come from somewhere. People are
> notoriously bad about writing metadata.

Hopefully the task will be made a lot simpler with WYSIWYG editors. Really,
the stupidity of the system arises from the fact that HTML is a presentation
language, not a semantic one (if there is such a thing). People are only
concerned about how their document looks: but that can be done with CSS. The
document itself should be marked up, not for display, but for storage: as an
information resource.

> I don't see how this is possible with your example, or, in truth, with
> really any system of markup (at least not reliably).

On the contrary, if you extract all of the Dublin Core attributes from the
example I gave, it gives you a very useful overview of the document. I'm
sure I could have added more.
Also, it isn't hard to write an XSLT script to extract the attributes:

<xsl:transform
    xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"
    xmlns:util="http://purl.org/dc/elements/1.1/"
    >
<xsl:template match="*[@dc:title]">
  <xsl:value-of select="@dc:title"/>
</xsl:template>
<xsl:template match="*[@dc:description]">
  <xsl:value-of select="@dc:description"/>
</xsl:template>
<!-- don't pass text thru -->
<xsl:template match="text()|@*"></xsl:template>
</xsl:transform>


> Excuse me if I'm wrong, but I thought this was a goal for XHTML and XSLT.
> I've also heard things about the next generation of WML being an XHTML
> profile/subset.  I don't see how metadata description helps such files
> creation.

How do you transform HTML into WML 1.0/1.1? You can't because they are so
different, HTML is so much more complex than WML.

> Better semantic tagging helps, though.

Precisely: if you had a semantic XML source, you could go to either XHTML or
WML easily.

> > 3. Use the foot section to generate an RDF description of the Further
> > Examples DL.
> I don't quite follow what you're doing here.

Transforming the DL list into RDF: Dan Connolly's idea.

> > 4. One day, hopefully, validate the dang thing.
> And this would help, how?

Well, why validate any document?

[...]
> > 6. Use it as a good example of Semantic markup, far removed from HTML.
> I don't see the semantics here...
> Sorry, I feel like I'm missing something.

The main fact is that HTML is a fairly presentation based markup language
that says very little about its content; the second point is that the SW
will require some kind of (semantic) output medium. Why not do both in one
go?
Tim Berners-Lee's vision describes a searchable web of logic:
"We note that logical systems provide provably good answers, but don't scale
to large problems. We see that search engines, remarkably, do scale - but at
the moment produce very unreliable answers. Now, on a semantic web we can
imagine a combination of the two. For example, a search engine could
retrieves all the documents which reference the terms used in the query, and
then a logical system act on that closed finite world of information to
determine a reliable solution if one exists." -
http://www.w3.org/DesignIssues/Evolution.html
As this obviously requires semantic document formats, SDF type structures
have to be implemented, and work, to ensure the SW comes about. If not, why
on earth are we bothering for?
The whole point of RDF is as metadata to XML: if you believe that people
aren't going to bother writing metadata for their documents, then why should
any of us bother with RDF? The answer is that HTML and the current WWW are
reaching the limits of what they can do, and the Semantic Web and associated
techologies are the future...
We just need to do more *doing*, and that's all I'm trying to achieve with
SDF.

I hope that clears everything up a bit!

Kindest Regards,
Sean B. Palmer
----------------------------------------------------
The Semantic Web: A Resource - http://xhtml.waptechinfo.com/swr/
WAP Tech Info - http://www.waptechinfo.com/
Mysterylights.com - http://www.mysterylights.com/
----------------------------------------------------
"The Internet; is that thing still around?" - Homer J. Simpson
Received on Tuesday, 7 November 2000 18:34:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 October 2010 18:14:24 GMT