On Aug 23, 2007, at 11:18 AM, Eric Jain wrote:
Almost no database, data center, or publisher uses HTTP URIs for
identifying their digital objects, and stable HTTP URIs at present
aren't adopted or the common denominator for identifying digital
objects in the life science domain either. And that's not b/c it
is technically difficult to do so, or because nobody knew that
stable URIs are desirable.
There are over a hundred databases I have to deal with in the
context of UniProt [see http://beta.uniprot.org/docs/dbxref], and
the one common thing they all have are URLs for their resources.
True, most URLs are not as stable and nice as we'd like them to be
Yeah, because they are URLs. But don't almost all of the databases
you have listed there use identifiers (accession numbers, etc) for
identifying their objects? Obviously, each one has their own (and
volatile) way of translating their identifiers into a URL. That's
(part of) the problem. So, yeah, they share a common thing, namely
having a website and serving their holdings through HTTP, but they're
not using URLs to identify their digital objects, are they?
-hilmar
--
===========================================================
: Hilmar Lapp -:- Durham, NC -:- hlapp at duke dot edu :
===========================================================