Re: Protocol question: PreferMinimalContainer

Right, the preference is sent to the container but unless the first page is
embedded in the response (see Example 39:
https://www.w3.org/TR/annotation-model/#annotation-page ), it only applies
when the page is retrieved.  However in order to keep the URIs separate
(for caching reasons) between embedded and not embedded annotations, the
container needs to know which the client wants, so it can link to the right
first page.

R

On Wed, Jun 15, 2016 at 9:11 PM, Benjamin Young <byoung@bigbluehat.com>
wrote:

> Related to my confusion is this line from 4.1.1:
>
>
>
> “If no preference is given by the client, the server should return the
> full annotation descriptions. The server may ignore the client's
> preference.”
>
> http://w3c.github.io/web-annotation/protocol/wd/#client-preferences
>
>
>
> Preference is expressed on the AnnotationCollection, but pagination URLs
> are used to link to the contents of the container, so…”SHOULD return the
> full annotation descriptions” is really talking about what happens after a
> request is made for the first or last page linked from the
> AnnotationCollection?
>
>
>
> If that’s indeed the case, I’d love to tweak that sentence to express that.
>
>
>
> Cheers!
>
> Benjamin
>
> --
>
> http://bigbluehat.com/
>
> http://linkedin.com/in/benjaminyoung
>
>
>
> *From:* Benjamin Young [mailto:byoung@bigbluehat.com]
> *Sent:* Wednesday, June 15, 2016 11:48 AM
> *To:* public-annotation@w3.org; Robert Sanderson <azaroth42@gmail.com>
> *Subject:* Protocol question: PreferMinimalContainer
>
>
>
> Given that an AnnotationContainer document doesn’t every include `items`
> direction—only links to `first` and/or `last` pages—what is the value add
> of `PreferMinimalContainer`?
>
>
>
> http://w3c.github.io/web-annotation/protocol/wd/#client-preferences
>
>
>
> It seems that the response to that request would be identical to either
> the default (PreferContainedDescriptions) or, given the preference for
> minimalism, it would be identical to PreferContainedIRIs—in which case one
> or the other of these isn’t needed.
>
>
>
> Each of these preferences really only dictates the IRIs used for `first`
> and `last` afaict.
>
>
>
> Or am I missing something obvious? :)
>
>
>
> Cheers!
>
> Benjamin
>
>
>
> --
>
> http://bigbluehat.com/
>
> http://linkedin.com/in/benjaminyoung
>
>
>



-- 
Rob Sanderson
Semantic Architect
The Getty Trust
Los Angeles, CA 90049