DTD for the files produced by the RST-tool?

[John Bateman]

> I don't use Marcu's tool, I use Mick's one and I have a dtd for that.
> I downloaded Marcu's one but I didn't have time to use it, so I don't know
> which are the differences between outputs. Send me an email if you need
> the dtd that I have.

I would be interested in the DTD. We have a DTD for a somewhat more strictly
interpreted RST than Mick's coder supports as part of our corpus of analysed
multimodal documents at http://purl.org/net/gem . Go to the corpus
page and look for annotation formats. I would like to compare,
as I think ours is simpler and easier to read than Mick's but extensions
and changes are always welcome.

> 
> ps. I know of some xml editor which has some buit in routine to generate
> dtd or xschema definitions from xml files. I am pretty sure that xmlspy
> provides this functionality.

Yes it does. But you have to go over the resulting schema/dtd to check it
for sillinesses. It generally does a good job though.

Olga:

For Daniel Marcu's system, the output always used to be more like Lisp
(I have not looked to see if this has changed recently). We also have
various tools and tricks for going from Lisp to something more like XML
but this is not always the prettiest solution. There is also an XML Schema
for Lisp as such.... which would be a fairly heavyhanded solution but
possible in theory!

Best,
John.


-- 
John Bateman
FB10, Sprach- und Literaturwissenschaften
Universität Bremen
28334 Bremen, Germany.

Tel: +49/421-218-9483
Fax: +49/421-218-4283 (or 218-7801)
http://www.uni-bremen.de/~bateman