home | books | articles | gleanings | case studies | hire
other sites: widgetopia | blueprints for the web | metafooder


 


 


« not quite | main | read this »

meta-kidding yourself

from Metacrap

"A world of exhaustive, reliable metadata would be a utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. "

The reasons why decent metadata is near-impossible to come by are funny and true--- and point toward why the semantic web is going to be a hard row to hoe.

Posted at December 20, 2002 08:28 AM


Comments

 

See Christopher Alexander, A City Is Not A Tree. (At least, if I remember it correctly.)

Posted by Randolph Fritz at December 20, 2002 09:53 AM


~~~

Interesting stuff, and very likely true for the web as a whole, but not the whole picture. Within limited domains where you can exercise a degree of control, meta-utopia may indeed be possible. At my last job, the big ongoing project was a move of all the documentation for all the company's projects to be moved to SGML and described with metadata, which would then be used to aide searches through the corpus of documents. Since the people creating the documents were under the thumb of the people dictating that this would be the way things work, they actually had a pretty good chance of succeeding.

The problem, of course, is that even in this relatively small circumstance, it takes a hell of a lot of work to get to this point, and when I was laid off a year ago, they still had a way to go to get to meta-Nirvana. From what I hear from my friends still there, they're still plugging away.

I'm sure you're aware of this, Christina, since your current employer's main gig is generating metadata for a limited subset of the web. It would be hard to call Yahoo! a failure.

Posted by ralph at December 21, 2002 02:35 PM


~~~

"And for metadata to be useful, it has to be honest."

You see, that's where I disagree. A lot of the more useful metadata (the ambigious type) is never without bias and therefore can never be "honest". The best you can do is recognize the bias, and the bias may actually be a large part of the value you get from the metadata: without bias the value of the metadata would be low. Bias ads value. I'm gonna write a blog post on this now :)

Posted by PeterV at December 22, 2002 06:49 AM


~~~

So here we go:
http://poorbuthappy.com/ease/000732.html

Posted by PeterV at December 22, 2002 06:59 AM


~~~

I just called Yahoo's classification system a failure in an article I'm writing :) I know it's borderline...

Posted by PeterV at December 22, 2002 07:10 AM


~~~



Post a comment
*Name:


*Email Address:


URL:


Remember me?

Comments:

bold italic underline link


posting can be slow; please wait a few seconds before hitting the button again.

The extra-fine print
wording stolen by the more-eloquent-than-I kottke
The bold, italics, and link buttons (and associated shortcut keys) only work in IE 5+ on the PC.
Hearty discussion and unpopular viewpoints are welcome, but please keep comments on-topic and *civil*. Flaming, trolling, and ass-kissing comments are discouraged and may be deleted.
All comments, suggestions, bug reports, etc. related to the comments system should be directed to me.


mail entry to a friend

Email this entry to:


Your email address:


Message (optional):




« not quite | main | read this »

 

 

 

home | books | articles | gleanings | case studies | hire
other sites: widgetopia | blueprints for the web | metafooder