W3C home > Mailing lists > Public > w3c-sgml-wg@w3.org > April 1997

Re: xmllink-970406 various

From: Peter Murray-Rust <Peter@ursus.demon.co.uk>
Date: Thu, 24 Apr 1997 11:27:54 GMT
Message-Id: <5953@ursus.demon.co.uk>
To: w3c-sgml-wg@w3.org
I have a need to carry out Boolean operations on the results of TEI searches
and would like to explore language support for this.  (It arises out of
my 'epiphany' (I gather from RL contact that this is the accepted term in
SGML circles) that SGML/XML documents can be used as databases.) 

An XML document holds a list of XML-LINKs (perhaps manually crafted, perhaps
retrived from a search engine).  The surfer wishes to avoid certain sites
and is used to the NOT operator.  At present XML-LINK has no mechanism
for supporting this, and it would have to be application-dependent.

As non-textual applications become common I suspect this will become a very 
desirable feature.  (It was important enough that I built it into costwish.)

One way forward may be via regular expressions on the components of a TEI
pointer.  I haven't thought this through, but I suspect it would give me
most of the power that I need, and equivalent to retrieving a number
of fragment sets and operating on them.  It highlights the need for a 
regexp at this stage, and I know Tim has been fishing for one (we'd need
one that could be translated into the most likely languages that will be
used to support XML, rather than simply a set of compiled C libraries).


BTW I have had the opportunity to meet a number of the WG IRL at the London
XML course - these are the first SGML subspecies that I have met and it was 
a real pleasure :-)

Peter Murray-Rust, domestic net connection
Virtual School of Molecular Sciences
Received on Thursday, 24 April 1997 06:45:24 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:25 UTC