RE: [xml-dev] Marketplace XML Vocabularies

["Jim Tivy" <jimt@bluestream.com>]

Hi Liam

Comments below:

-----Original Message-----
From: Liam R E Quin [mailto:liam@w3.org] 
Sent: Monday, January 04, 2010 6:23 PM
To: Jim Tivy
Cc: 'Costello, Roger L.'; xml-dev@lists.xml.org
Subject: RE: [xml-dev] Marketplace XML Vocabularies

On Sun, 2010-01-03 at 11:29 -0800, Jim Tivy wrote:

> "meaning in XML documents is "extrinsic".  Perhaps it would be safer
> to say that meaning is extrinsic to eXtensible Markup Language - although
> some may argue that even XML constrains expression.

Constrains expression, yes, but that's not the same thing.

> I do not believe, however, that meaning is extrinsic to instances of
> XML that conform to XML vocabularies (XML documents)

I claim that there is no intrinsic meaning in an XML document, but
rather that external knowledge is applied to infer meaning.

To see this, consider any piece of XML out of context:

<i>
  <e>Fiona MacCarthy</e>
  <w>Eric Gill</w>
  <t>506</t>
  <g>b80e87135095c396d5e6adb067a553e1</g>
</i>
[<JT>] 
Many XML documents have meaningful tags or/and a DTD or schema.

As well, my point was based on XSLT as an example:

<?xml version="1.0" encoding="UTF-8"?>

<!-- This file is part of the DITA Open Toolkit project hosted on 
  Sourceforge.net. See the accompanying license.txt file for 
  applicable licenses.-->
<!-- (c) Copyright IBM Corp. 2004, 2005 All Rights Reserved. -->
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform";
version="1.0">
  <xsl:import href="../common/output-message.xsl"/>
  <!-- Define the error message prefix identifier -->
  <xsl:variable name="msgprefix">DOTX</xsl:variable>
  <!--xsl:param name="WORKDIR" select="'./'"/-->
  <xsl:param name="PROJDIR" select="&apos;.&apos;"/>
  <xsl:param name="DBG" select="no"/>
  <!-- Bluestream  7-June-2007
<xsl:param name='FILEREF'>file://</xsl:param> -->
  <xsl:param name="FILEREF"/>
  <xsl:variable name="ORIGINAL-DOMAINS"
select="/*/@domains|/dita/*[@domains][1]/@domains"/>
  <xsl:template match="/">
    <xsl:apply-templates>
      <xsl:with-param name="conref-ids" select="&apos; &apos;"/>
    </xsl:apply-templates>
  </xsl:template>
  <!-- If the target element does not exist, this template will be called to
issue an error -->
  <xsl:template name="missing-target-error">
    <xsl:call-template name="output-message">
      <xsl:with-param name="msgnum">010</xsl:with-param>
      <xsl:with-param name="msgsev">E</xsl:with-param>
      <xsl:with-param name="msgparams">%1=
        <xsl:value-of select="@conref"/>
      </xsl:with-param>
    </xsl:call-template>
  </xsl:template>
  

Now, tell me what those XML elements describe...

Hmm, well, its not possible in general.  We can make a guess and say
that i is for inventory, e is for editor, w is who a biography is
about, t is the thickness of the book in pages, and g could be an
encoded graphic but actually is the guarantee that the record is correct
(it's an md5 hash).

But none of these things are intrinsic to the markup.

This is very different from RDF, where explicit URIs are used, rather
than ungrounded element names. We have ambiguity, which adds richness
to our lives, multiplicity without (I hope) duplicity.

> I would argue that the use and meaning of XSLT Xml is intrinsic to the
> XSLT XML itself.  This is not to say I could not use XQuery on some XSLT
> documents to count how many time the <xsl:if> tag was used.
I think you contradict yourself there.  Or that we are somewhat at
cross-purposes.  The whole point is that the XQuery expression you use
doesn't need to know anything at all about the meaning, behavioural or
otherwise, of the XSLT markup.
[<JT>] Not a contradiction - this is not a mutually exclusive matter where
either you are right or I am right, rather I think we both are, but my
argument is I am more right :) - if you'll excuse the humour...

> This discussion also touches on controversies such as should there be
> an XML representation for languages like XQuery or Java...
There already is for XQuery at least, W3C XQueryX.
[<JT>] Yes, I am aware of XQueryX.  My question is this:
Does the same XQuery expression in XQueryX have less intrinsic meaning than
XQuery compact non Xml Syntax.  Does XQuery compact have no intrinsic
meaning?

Best,

Liam

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org www.advogato.org