RE: Indexing XML

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

We could do it in IADS as well across 
document collections.  EBT was 
a lot more of a full-up SGML browser and 
had the edge, but doing it wasn't all 
that novel.  Building indexes is an 
established practice.  IADS was not 
web-enabled but a database is a database 
and other IETM systems were database 
driven long before we had the web.

The meaningfulness of any claim like 
that can be questioned apart from the 
marketing value.  But it does little 
harm to the literature as long as we 
are critical about assertions.  Since 
we just did another long round on the 
Semantic Web, authoritative knowledge 
bases, the problems of broadcast systems 
and truth maintenance, we don't have to 
do it again.  Just beware of the langoliers 
of marketing and what they do to systems 
that conflate identity, location, truth 
and falsity.   Don't go to the marketing 
department to verify history.  It isn't 
their job.  They aren't bad; they are 
just there to sell.

Len 
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Bob DuCharme [mailto:bob@udico.com]

Several years before XML became a Recommendation, Dynaweb
offered contextual searching of SGML on the web. XML is a subset of SGML, so
any product that could perform this for SGML could theoretically do it for
XML before XML was invented. I'm not absolutely sure that EBT's Dyna* family
of products supported every feature of SGML declarations, which is the true
test of whether an SGML document can handle XML; if they didn't, addressing
this aspect of Dynaweb directly would be a better way to add weight to your
claim.

xml-dev is not always a friendly place for marketing bombast. That's one
reason I like it!