- From: Mary Morris <marym@finesse.com>
- Date: Tue, 28 Nov 1995 11:38:02 -0800
- To: jtauber@cyllene.uwa.edu.au
- Cc: www-html@w3.org
> > <META HTTP-EQUIV="Subject.DDS.scheme" CONTENT="Dewey Decimal System"> > > <META HTTP-EQUIV="Subject.DDS" CONTENT="004.64"> > > <META HTTP-EQUIV="Subject.SOS.scheme" CONTENT="Some other system"> > > <META HTTP-EQUIV="Subject.SOS" CONTENT="Model Trains"> > > Why not just: > > <META HTTP-EQUIV="DDS" CONTENT="004.64"> > <META HTTP-EQUIV="SOS" CONTENT="Model Trains"> > > ? > > Is there something significant about starting with "Subject" that I don't > know about? (quite likely :-) Actually yes. I am working on a book called Web Page Design: A Different Multimedia. Since I am horrified at the state of current search engines, I am proposing some META tag additions to most web pages to make them easier to index. In cases where the robot-gatherer is intellegent enough to understand specific classification systems, I would like it to encourage the use of them. In the case where the robot doesn't have the rights to use the classification system (such as Dewey Decimal is a copyrighted system) or isn't aware of the system, I would like them to be able to identify the general nature of the information element from the name or http-equiv definition. Parsing "search" off of the begining of a longer string appears to be pretty simple. I would expect to use classification systems in both the subject and keywords defintions with a slight variation in each, thus using a simple classification definition won't define the difference between keywords and subject. Since I am writing this book for publication, I can't pass out the material for general review, but I would be interested in feedback from a few people that are interested in reviewing the material. This chapter should be available on Friday. Mary
Received on Tuesday, 28 November 1995 14:44:18 UTC