Re: XML Example and DTD Archive?

Peter Murray-Rust (peter@ursus.demon.co.uk)
Thu, 27 Nov 1997 01:00:02


At 15:17 26/11/97 -0800, Don Park wrote:
>Fellow XML Developers,
>
>I have searched for but could not find an extensive archive of XML examples
>and DTD. If there is such an archive, please let me know. If not, I would
>like to build one so we can all benefit.

Don,
This is a most exciting offer!

You are right that there is no *extensive* archive of XML material and we
are suffering because of that lack. Certain people have contributed things
which may (or may not) be consistent with the latest draft :-) - that's one
of the problems. The places where these are reported are:
- XML-DEV , and I try to extract things like this into XML-JEWELS at
http://www.vsms.notingham.ac.uk/vsms/xml/jewels.html
- http://www.sil.org/sgml/xml.html - Robin Cover keeps an eagle eye for
anything of value.

Jon Bosak's Shakespeare, and religion are pre-eminent and are a good test
for whether a system can cope with 'real documents'. I haven't looked at
religion, but Shakespeare has a clean and natural markup without
attributes. So it's not a torture test. (I don't think there are DTDs - I
think I hacked my own). I don't think there is any mixed content in
Shakespeare

Michael Sperberg-McQueen wrote a torture-test for XML parsers early this
year. We seriously need this up-to-date - maybe Michael is reading this :-)

I have written a lot of Chemical markup language (CML) at
http://www.vsms.nottingham.ac.uk/vsms/java/jumbo
and it uses attributes heavily. However there is NO mixed content in CML,
and the output is disappointing without a chemical browser :-)

There are snippets of XML in the XSL spec, an the RDF spec and in the
MathML spec. None of these have (I think) DTDs [MathML has one in principle].

I have now tested 3.5 parsers under JUMBO and have found that there is
sufficient variation between them that we really need some test documents.
(Some of the variation is behavioural - i.e. should a browser fail if it reads
<!DOCTYPE FOO SYSTEM "foo.dtd"> and foo.dtd doesn't exist.)

In my view, collaborative *action* is worth many kilowords of discussion,
and if you can help put together such a resource it would be extremely useful.

Best Wishes

P.

Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg