discussion of the Creative Commons Metadata work

[Matthew Haughey <matt AT haughey.com>]

On Dec 3, 2003, at 10:21 AM, Cory Nelson wrote:
> What do you guys think?

We've looked at that option before and here's what we learned while 
researching it:

- most major search engines (including Google) ignore meta tags that 
are self-descriptors. Due to years of rampant keyword abuse, they can't 
be used as a trusted index point. The only meta tags that search 
engines look at are the directives that say "don't index this". A  
Creative Commons meta description would kind of fall into the middle, 
but still be something that would have trust issues.

- There is a lack of detail with a single meta tag in the head of a 
document. Does the license apply to the page itself? The contents of 
the page? Which contents? All contents? Should more specific 
limitations be in new meta tags or in attributes on the CC meta tag?

- Somewhat related to the previous one is an issue of proximity. A tag 
in the head of a document could be pretty far from a downloadable mp3 
on a page carrying the license meta tag. With the HTML/RDF option, you 
can put "this song is released under CC" and put the full descriptive 
mp3 RDF directly in the page, right next to the mp3 file it describes 
(it can also include the URL to it).

- Related to the first point, meta tags represent the way things have 
been done in the past and most companies are moving away from it. I 
know there isn't a lot of embedded RDF on the web, but a few 
applications have been created (like mozCC) to scrape pages, look for 
CC data, and display information appropriately, and many more are on 
the way. Embedded metadata in RDF format looks like a possible future 
for search engines, following the much-debated Semantic Web, and this 
could be one real-world application to foster building an RDF-aware 
search engine.

Matt