Fw: Is there anyone working on a binary version of XML?

[Oren Ben-Kiki]

Stephen D. Williams <sdw at lig.net> wrote:
>I don't understand how to use XSL in a general way to acheive a 'delta
tree' architecture.  I
>have a vague idea, but nothing that I could see being automated
sufficiently.  Can you
>elaborate?


The following (from section 2.7.12 of the current XSL draft):

<xsl:template match="*|@*|comment()|pi()|text()">
    <xsl:copy>
        <xsl:apply-templates select="*|@*|comment()|pi()|text()"/>
    </xsl:copy>
</xsl:template>

Will copy all input to the output without modification. You can then add
templates to do specific modifications. For example:

<xsl:template match="TAG/@ATTR[.='OldValue'">
    <xsl:attribute name="ATTR">
        NewValue
    </xsl:attribute>
</xsl:template>

Will take all 'TAG' elements in the input document which have an 'ATTR'
attribute whose value is 'OldValue' and change its value to 'NewValue'.
Given the power of XSL match patterns and the power of the construction
elements, I think you can express any reasonable 'delta' on the input XML
tree.

Of course, this is outside the scope of the XSL intent as it stands today.

<Rant-and-Rave>
The transformation part of XSL is just what we need for:

- An XML query language. Think about it - an XML query language should (i)
be XML; (ii) allow selecting arbitrary parts of the input XML document(s);
(iii) allow constructing result XML document(s). The transformational part
of XSL already does 80% of that. Does anyone consider making XQL a proper
superset of XSL? Not a chance. Everyone is intent on creating a new
language. XQL at least reuses the match pattern syntax, while inventing a
new incompatible way of creating the results tree; XML-QL goes for broke and
reinvents the whole thing.

- A standard way to convert XML documents to legacy non-XML languages. Oops,
I just said non-XML languages. Excuse me.

- New and unexpected uses, such as the one above: expressing differences
between XML trees (which by itself has a lot of interesting applications).

But no, due to historical reasons XSL was created as part of a style
language, so we'll just have to use a different language for each of the
above uses and any new one which comes along (making sure they are
incompatible, of course).

Never mind that CSS is alive and kicking and supported by the very same W3C
is another way of specifying style. Never mind that CSS is staying away from
anything which might look like XML syntax, and is well along the way of
inventing a new match pattern language of its own, whose only advantage over
the XSL one is that it is incompatible with it.

I'm sure it all makes sense for _someone_. Whatever the reasons are, what I
see is "Job security for XML professionals for the next millennium".
</Rant-and-Rave>

Sorry, I just had to get it off my chest :-)

Have fun,
    Oren Ben-Kiki



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)