RE: ungetable http URIs

Hi folks,

Actually I think there are two questions here:

First, within the scope of the demo: I think we have agreement that we
should not be using ungetable URLs right? So there are two possible
solutions: 
a) Should we modify the XSLT stylesheet and re-run it to create the Artstor
data so that it uses URNs rather than URLs avoiding ungetable URIs? 
b) Or should we use Joseki infrastructure to ensure that it is possible to
get those URLs? 

Andy, perhaps you can clarify - if we switch to using URNs, can we use
Joseki are does this create problems?

If we agree on a), then I can create the data fairly quickly. If this is
holding up progress, perhaps we need to go with a) for now - particularly as
we are not necessarily looking for the best solution, rather the one that
helps the demo move along? If we agree on b), then we need to revisit the
problem of creating a Joseki instance that only serves data to select IP
addresses, and set up two Joseki instances, one within the HP firewall and
the other at MIT - I think we had a proposal about how to solve the security
problem?

Within the longer term scope of SIMILE, I think there are 4 possibilities:
c) use URLs that currently ungetable, but might be getable in the future
d) use URLs and use Joseki infrastructure to ensure they are getable
e) use URLs and some other type of infrastructure e.g. Tomcat to return "no
further data"
f) use URNs

have I missed anything?

Dr Mark H. Butler
Research Scientist                HP Labs Bristol
mark-h_butler@hp.com
Internet: http://www-uk.hpl.hp.com/people/marbut/


> -----Original Message-----
> From: Nick Matsakis [mailto:matsakis@mit.edu]
> Sent: 01 December 2003 17:17
> To: Stefano Mazzocchi
> Cc: SIMILE public list
> Subject: Re: ungetable http URIs
> 
> 
> 
> 
> On Fri, 21 Nov 2003, Stefano Mazzocchi wrote:
> 
> > If I got to vote, I would vote +1 for "getable" URIs because I think
> > that they don't add complexity, they are consistent with the general
> > XML movement, and they are potentially more valuable in the future.
> 
> If I got to vote, I would say 'getable' URIs should be 
> assigned to things
> that can be expressed as bits, or things that are elements of 
> RDF schemas
> (e.g. if you come across something with an RDF type you don't 
> recognize,
> it would be nice if there was a schema saying something that type at a
> conveniently retrivable URL).
> 
> I think things that don't meet those requirements should get ungetable
> URIs.
> 
> Nick
>

Received on Monday, 1 December 2003 12:33:51 UTC