Re> More on Classifying Web Sites

[Peter Kumaschow]

This was posted to some newsgroups in the comp.infosystems.www 
heirarchy. Looks like a method for self-classification of web sites is 
already developing. That's what I love about the internet, as soon as I 
think of something it usually is already happening somewhere out there. 
Now shall chaos or order or a fluctuation between the two ensue?

On 30 May 1996 13:09:41 -0400, floydb1 at lib109.its.rpi.edu (Barry B
Floyd) wrote:

>
>Alta Vista (search engine) uses the <META> tag,
>where they want label='keywords' and label='description'.
>
>Other search engines may be using keywords/description META tags, if
>found in HTML pages.
>

The real benefit of Identify is that content is classified under a
consistent hierarchy (they call it an ontology) so once you get a
result set, you can filter it by subject area.  This may sound like
Yahoo, and it is, sort of.  You may have heard, though, that Yahoo
has recently put a prioritization scheme in place, because they
can't get to everything fast enough.  This way everyone classifies
their own stuff.

The best thing about Identify, though, is the product stuff.  They
only have CDs and Movies right now, and only one vendor for each
so far, but just search for Pink Floyd The Wall, and you'll get an
example of how cool this could be.  You get CDs and movies in the
same search, with links right to where you can buy them.  Imagine
in the future getting several results for the same CD, so you can
comparison shop.  Better yet, imagine getting CDs, movies, books,
T-shirts, posters, even concert tickets from a dozen different
vendors, all in the same place.  How cool!



-- 
-----------------------------------
Peter Kumaschow                        
Email:  peterk at openweb.net.au
Phone:  02 267 1222 ext 110
Mobile: 0419297948                     
URL:    http://www.wr.com.au/webtext