Re: XFO Mapping...

Subject: Re: XFO Mapping...
From: Guy_Murphy@xxxxxxxxxx
Date: Thu, 6 May 1999 10:40:19 +0100

Personaly I don't really see this as a problem for search engines. Unless
the Web adopts one single semantically rich data model, the search engines
have to resort to tactics other than relying on the data model. It is
unlikely that by this point you're going to be able to cage XML
distribution into a single data model, and certainly not without killing
XML+CSS, which allows people to throw any data model they want onto a Web
page (XML+CSS considered harmful? :)

I can only really speak from my own experience, and having worked on an
intranet app utilising a categorisation engine that can determine
"aboutness" for anything from Web pages to Word files, I see no reason for
search engines not to adapt to new Web content.

If we want to give search engines a helping hand I feel this best addressed
by a common meta data standard for use by search engines for indexing,
which would likely recieve wide adoptence due to the large amount of people
wanting indexed by search engines. Then we hit the problem of meta data
being abuse by porn and warez sites, so we are back to square one whereby
the engine cannot rely on a data model, but must infer meaning from the

Personaly I don't feel search engines are any more the concern of XSL than
they are the concern of CSS. XSL is a styling language concerned with
presentation, not meta data and indexing. I'm not saying other areas
shouldn't be considered and facilitated by XSL, but only if doing so can be
bought cheaply.



xsl-list@xxxxxxxxxxxxxxxx on 05/06/99 07:54:55 AM

To:   xsl-list@xxxxxxxxxxxxxxxx
cc:    (bcc: Guy Murphy/UK/MAID)
Subject:  Re: XFO Mapping...

There is one group of "blind" for which aural constructs won't be of help:
indexing agents.  I will hazard the guess that search engine queries at
equal if not far exceed the number of blind *people* viewing web documents.

 XSL-List info and archive:

Current Thread