Re: Efficient XML - request for input

[Stefan Decker <stefan@db.stanford.edu>]

Hi,


>>Interesting - why is RDF considered to be to complex?
>>E.g. a namespace-mechanism ist necessary if multiple vocabularies (or 
>>ontologies)
>>are used. I see now way arround them.
>
>In RDF's case, I just feel that sometimes its usage over complicates 
>things.  For instance,
>http://rpmfind.net/linux/RDF/beowulf/i386/bert-examples-1.04l-1.i386.rdf

Lets see:
the first namespace (rdf) is the usual rdf namespace, that distinguishes
RDF-code from other XML-code. Otherwise there would be no way to identify 
rdf-code,
The second namespace identifies the used vocabulary (here for identifying 
RPM-modules).
Theres indeed a possibility so simplify it a bit: the rpm-namespace could be
the default one.
Then the example can be rewritten as:

<rdf:RDF
   xmlns:rdf="http://www.w3.org/TR/WD-rdf-syntax#"
   xmlns="http://www.rpm.org/">

    <rdf:Description 
about="ftp://rpmfind.net/linux/Beowulf/i386/bert-examples-1.04l-1.i386.rpm">
      <Name>bert-examples</Name>
      <Version>1.04l</Version>
      <Release>1</Release>
      <Arch>i386</Arch>
      <Os>Linux</Os>
    </rdf:Description>
<rdf:RDF>

I don't see a way how this could be expressed more elegant in another way 
without
introducing ambiguities?


>This data does not _really_ need the two namespaces.  The format could 
>much simpler and easy to
Which one would you drop?

>  read than it is.
How?

>Simplicity is elegant.

Sure - it should as simple as possible, but not simpler.

Ciao,

         Stefan


>Cheers,
>
>Matthew
>--
>Matthew MacKenzie
>CTO/VP R&D
>XML Global Technologies, Inc.