> From [log in to unmask] Tue Jun 3 16:34 MET 2003
> X-RAL-MFrom: <[log in to unmask]>
> X-RAL-Connect: <[log in to unmask] [130.238.162.227]>
> Mime-Version: 1.0
> X-MIME-Autoconverted: from 8bit to quoted-printable by ori.rl.ac.uk id
> h53EY8S03404
> Date: Tue, 3 Jun 2003 16:33:06 +0200
> From: Mikael Nilsson <[log in to unmask]>
> Subject: Re: URI fragment IDs, MIME types and DCMI PURL redirects
> To: [log in to unmask]
>
> On Tue, 2003-06-03 at 16:02, Pete Johnston wrote:
>
> > I think I was really wondering about the second URI
> >
> > http://dublincore.org/2003/03/24/dces#title
> >
> > and whether it serves any purpose to include a fragment identifier in
> > that URI at all.
>
> Good observation! I hadn't thought about it in that way...
>
> I guess you're right; it really makes *no* sense at all. There is no
> "#title" object in there to reference...
>
> So, does it matter? I guess it at least creates semantic confusion :-).
Does it? There are no triples - as far as i know - concerning these gadgets.
>
> I'd rather that we try to be very careful about the semantics of what we
> do. So why not propose the removal of the fragment id in the redirect?
Would not hurt, but also would not help much: elements/1.1/title then would
redirect to elements/1.1/ - I would prefer not to have a redirect at all.
rs
>
> /Mikael
> --
> Plus ça change, plus c'est la même chose
>
>
|