W3C home > Mailing lists > Public > public-openannotation@w3.org > August 2012

Re: Selection Filtering

From: Robert Sanderson <azaroth42@gmail.com>
Date: Tue, 14 Aug 2012 11:48:02 -0600
Message-ID: <CABevsUE+LGLHkV7LnYai0SdUMtcSbJCWqENavrr9wxHevMaT6g@mail.gmail.com>
To: Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>
Cc: t-cole3@illinois.edu, public-openannotation <public-openannotation@w3.org>
Apologies! Yes, I meant only Selectors as opposed to State/Context/etc,
rather than only a single Selector.  Indeed, multiple Selectors on a
specific resource are alternates and hence wouldn't affect this particular
issue.

However... given the fidelity of different selectors, you would *not* want
to use a fragment URI unless it precisely expresses the area.

For example, if the area is a circle described with an SVG Selector, it
would be very bad to have a #xywh= media fragment that identified the
bounding box around the circle as the identifier for the Specific
Resource.  In this case, all of the different paths within the bounding box
would be conflated across Annotations.

So ... I don't see it as a show-stopper, but we don't say that you can't do
it in the specification. The text from the spec is:

The Specific Target is typically identified by a URN, as an HTTP URI would
imply that the exact nature of the Specific Target was available to
retrieve by dereferencing the HTTP URI.

Which seems at most a weak SHOULD recommendation.

Thanks!

Rob

On Tue, Aug 14, 2012 at 7:07 AM, Stian Soiland-Reyes <
soiland-reyes@cs.manchester.ac.uk> wrote:

> On Mon, Aug 13, 2012 at 10:27 PM, Robert Sanderson <azaroth42@gmail.com>
> wrote:
> > And, to clarify, it is to be used as the identifier for the specific
> > resource *with* a Selector, rather than in place of it.  The reasoning:
>  the
> > URI is not obfuscated as it's in hasSource, conflation isn't really an
> issue
> > as the description is also available, and that it's explicitly only for
> when
> > there's a single selector.
>
> Why the requirement for a single selector? Did we not agree that the
> selectors are 'OR'-ed rather than 'AND'-ed? (Ie. they are alternative
> ways to select the same thing).  The spec does not say anything about
> the interpretation of multiple states or multiple selectors. I think
> it should.
>
> And so there might be many ways to select a resource appearing in
> another (say image in web page), but I don't see that as a stopper for
> giving it a resolvable URI  - ie. a URL.
>
> (remember: URI = URL union URN )
>
>
>
>
> --
> Stian Soiland-Reyes, myGrid team
> School of Computer Science
> The University of Manchester
>
Received on Tuesday, 14 August 2012 17:48:31 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 14 August 2012 17:48:32 GMT