W3C home > Mailing lists > Public > www-rdf-interest@w3.org > June 2001

Re: [topicmapmail] RE: Topic Maps Graph in XML, with API

From: Murray Altheim <altheim@eng.sun.com>
Date: Wed, 27 Jun 2001 10:44:04 -0700
Message-ID: <3B3A1B64.C547F07B@eng.sun.com>
To: Graham Moore <gdm@empolis.co.uk>
CC: www-rdf-interest@w3.org, "'Topicmapmail@Infoloom. Com'" <topicmapmail@infoloom.com>, xtm-wg@yahoogroups.com
Graham Moore wrote:
> I think the DOM API as a accessor to topic maps is useful up to a point but
> ultimately will not convey the true functional semantics that are in
> TopicMaps. It will still be necessary to have Semantic Topic Map API even if
> under the hood it is implemented in terms of the DOM - for many of the
> multidimensional aspects of TopicMap query and access this will not scale.
> Simply things like topic.getNameInScope(ScopeSet) will not be supported by
> the DOM. And it is this and many other methods that will actually make the
> model useful. A topic is a topic a DOM Node is a dom node it is not a topic.
> That may sound obvious but there is a difference.

I'll have to read over Michel and Steve's recent posting, but I must agree
with Graham on the DOM. It doesn't scale well. When dealing with the 3000
topics and 11,000 associations in Cycorp, it bogged down horribly. I was
forced to write a lot of machinery that supplemented DOM transactions in
order to do even fairly trivial things. 

We talked last year about having an attribute on an XTM document indicating
its merged status. I was told this wouldn't be enough because there was 
still various kinds of TM processes that the DOM wouldn't support. I didn't
see that then, but certainly do now (after trying to code the thing). The
biggest problem I've seen though is performance. Manipulating a large 
DOM tree is really not very efficient. Large topic maps would be impossible,
at least so far as I can tell. I've for the past six months had a subproject
investigating using an XML database as a backend replacement for the DOM.

Like I said, I haven't read over the recent graph materials (which I will
likely do next week). But I tend to think that Steve Newcomb was originally
correct in saying that standard XML tools won't suffice.


Murray Altheim                            <mailto:altheim&#x40;eng.sun.com>
XML Technology Center
Sun Microsystems, Inc., MS MPK17-102, 1601 Willow Rd., Menlo Park, CA 94025

      In the evening
      The rice leaves in the garden
      Rustle in the autumn wind
      That blows through my reed hut.  -- Minamoto no Tsunenobu
Received on Wednesday, 27 June 2001 13:45:32 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:07:36 UTC