On Sunday 15 December 2002 16:39, Lars Marius Garshol wrote:
> * Kal Ahmed
>
> | I always thought that XSLT wouldn't work particularly well for topic
> | map presentation, but recently I tried a slightly different
> | approach, using an XSLT stylesheet to work on a mini-topicmap
> | extracted from a larger topic map.
> |
> | By doing this extraction, I can force the extracted "fragment" to be
> | a consistent topic map (as defined by XTM) and furthermore ensure
> | that references are "normalized" to topicRefs. Then I have a small
> | topic map that can be far more easily processed by XSLT.
>
> Would be interesting to see the code examples. Do you have it lying
> around anywhere?
>
Its in CVS for TM4J
(http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/tm4j/tm4j/src/org/tm4j/topicmap/utils/TopicMapFragment.java)
> | And it was all done with TM4J & Saxon...so cost is not an issue ;-)
>
> Right. :-)
>
> * Lars Marius Garshol
>
> | One of the open issues with the TMQL requirements is whether an output
> | language should be part of TMQL (in much the same way that XSLT and
> | XPath fit together).
>
> * Kal Ahmed
>
> | Are you suggesting something like XSLT or something more like the
> | XQuery model ?
>
> I was merely suggesting a requirement, not proposing a particular
> solution to it. :-)
>
> I was thinking of something like XSLT, but I'm not opposed to
> something XQuery-like. Probably what we end up with will be influenced
> by what execution model we give TMQL.
>
Agreed.
> I submitted to Barcelona (or was it Baltimore?) a proposal for a paper
> that would discuss the construction of TMQL, such as how to refer to
> topics, possible execution models, interaction of execution model with
> integration of query language into larger context, and so on. They
> wouldn't have it, though.
>
Thats a shame. I guess that it is something which would be useful for the TMQL
group though.
> | I can see that an XQuery model could be really useful - combined
> | with a topic-type based templating system it would be quite
> | flexible. In fact, I am pretty sure that using a combination of
> | tolog, the topic map extraction algorithm I have for my XSLT
> | transforms and a bit of XSLT magic you could produce something
> | really cool.
>
> Robert Barta has (or has proposed, I'm not sure which) this in AsTMa?,
> so you may want to look at it. I thought it interesting, but haven't
> looked very closely at it.
>
> | Until this conversation, I hadn't really thought about the XQuery
> | style of doing things, but I think that it could have alot of
> | mileage. A pipeline of: XML template with embedded TMQL -> TMQL
> | processor -> XSLT processor -> Output format would be pretty
> | flexible.
>
> It's an interesting idea, I agree. Someone should look at it. :)
>
> Geir Ove and I have been thinking of another approach as well, which
> is closer to the XMLC that Enhydra uses. Whether we'll have time to
> write it up remains to be seen. My queue of good ideas that need to be
> written up grows faster than I can find the time to empty it. :-(
>
A feeling I am only too familiar with ;-)
> | Perhaps the application of TMQL in the manner described above would
> | actually eliminate the need for a templating language. One less
> | thing to learn is a Good Thing.
>
> Yeah, we shouldn't add more capabilities than we actually need. The
> good thing is that we'll do this one step at a time, so people can do
> experiments with different approaches before we choose one. I think
> the more brainstorming and experimentation we can do, the better.
Yes, and now is the time to start doing those experiments and reporting on our
findings I guess. Even without a standard query language, working with the
languages that exist should provide valuable input to the standardisation
process.
Cheers,
Kal
--
Kal Ahmed, techquila.com
XML and Topic Map Consultancy
e: kal-5x+ggncFKvmB+jHODAdFcQ@xxxxxxxxxxxxxxxx
p: +44 7968 529531
w: www.techquila.com
|
