W3C home > Mailing lists > Public > www-rdf-interest@w3.org > July 2004

Re: Real data

From: Sandro Hawke <sandro@w3.org>
Date: Thu, 22 Jul 2004 07:45:57 -0400
Message-Id: <200407221145.i6MBjvGC006423@roke.hawke.org>
To: Eric Jain <Eric.Jain@isb-sib.ch>
Cc: rdf-interest <www-rdf-interest@w3.org>


> The problem is that I can't guarantee that anything will ever be 
> directly resolvable in a stable way, especially as I have so far not 
> been able to convince the people in charge of the web site to support a 
> more logical naming scheme

I understand; I think there's a chicken/egg problem here, since there
is no good demonstration yet of how useful/important it is.

Some options:
   - use purl.org URLs instead of uniprot.org ones
   - make a new domain  (data.uniprot.org, uniprotsw.org, ...)
   - ask someone else with a stable domain to redirect for you

In general, I would be interested to hear what the people in charge
of the web site are saying, because this is must be a very common
problem.   

> And if your identifiers can't be resolved directly, I feel 
> it is better to be clear about that and use URNs, right?

It might still be better to use HTTP URLs in the hope that you'll be
able to serve for them someday.  I'm not sure.

> Another issue is that we reference a lot of data in other databases. 
> Virtually none of these databases have stable URLs for accessing their 
> resources (/cgi-bin/ etc). We therefore keep a list that allows us to 
> map identifiers for a given database to the address-of-the-day.

Can you use HTTP redirects to do that mapping, too (assuming the above
problem is solved)?

     -- sandro
Received on Thursday, 22 July 2004 07:44:35 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:58 UTC