November 8, 2005: The Memetic Web


Peter Morville already beat me to it, so I won't re-hash Bob Doyle's concept
of the Memetic Web here. I do think there's promise in tagging a document
with metadata that's both unique and meaningful, especially in closed sites
like intranets, where abuses are likely to be minimized, and using
vocabularies made up of unique IDs, like zip codes and ISBNs (more
descriptive metadata will be a hassle).

I'm not sure that users will be comfortable or, for that matter, able to
express their queries in a syntax (here's one for zip codes:
"MEMOZIP-02138-6707") that is intended to be more computer-readable than
human-readable. There may be exceptions of course; if, for example, I'm
looking for information on a specific book, it'd be pretty handy to learn
how to plug a ISBN into Bob's memetic syntax. But all in all, it'll be
difficult to achieve a critical mass of memetically-motivated users to
justify content producers' investments in tagging their content using Bob's
scheme.

Where I think this would be more useful, as I've suggested to Bob, is to
introduce to concept to search engine vendors. They can teach their systems
to recognize memetic tags ("oh, the user searched on the zip code
'02138-6707'; let's add meme tag 'MEMOZIP-02138-6707' to their query before
we execute it"). If a few vendors support the Memetic Web as Bob is
proposing, there might be enough incentive for content producers to start
tagging their content this way. And motivated users will follow...


BLOUG PERMALINK & COMMENTS
http://louisrosenfeld.com/home/bloug_archive/000399.html

LINKS MENTIONED
Peter Morville on the Memetic Web ::
http://www.findability.org/archives/000068.php
Memetic Web site :: http://www.memeticweb.org/