[
Lists Home |
Date Index |
Thread Index
]
On Thu, 27 Feb 2003, Bullard, Claude L (Len) wrote:
> If this were done as part of site metadata, wouldn't
> that get rid of the need for a centralized database?
especially since the whole framework for registration and ownership of an
identifier, all the legalities that come with it as well as propagation of
that information is already in place with DNS.
which makes it a simple (since it's merely technical) issue of figuring
out how to share the metadata.
apart from that ->
the google model relies on users being able to click on a resource and get
_the_real_thing_ as opposed to a google copy. it means you don't have to
take googles or any other search engines word for the authenticity of the
information. similarly, you could have a namespace-uri search engine,
which merely tells you where to find what you're looking for. quite
useful, if properly done.
then the problem to solve is how the search engine gets hold of that
information. hmm, maybe i'd better go read up on this REST business...
/m
Martin Klang
http://www.o-xml.org - the object-oriented XML programming language
|
