RE: CFC: Basic Roles Proposal

This discussion, and in particular the “sematic conundrum” that Doug describes and that Jacob cites, takes me back to a basic question:  What is (or was) the original business case for multiple bodies?
Was it:

(a)    To be able to bundle many independent and otherwise separate annotation into a single annotation, for efficiency; or

(b)   To be able to express relationships among several bodies working together (i.e. playing different roles) for a single annotation.

I honestly don’t recall whether or not we actually singled out one or the other and I know both were discussed.

I think we have to bite the bullet and say we can’t have both.

So Rob, you said that the example I cited is “purely hypothetical”.  Do you mean by that that it is invalid?  (In which case it should be removed.)  That would mean you think that the answer is (a) and (b) is disallowed.

Ray

From: Jacob Jett [mailto:jgjett@gmail.com]
Sent: Tuesday, September 01, 2015 1:38 PM
To: Doug Schepers
Cc: Robert Sanderson; Benjamin Young; Denenberg, Ray; Web Annotation
Subject: Re: CFC: Basic Roles Proposal

Hi Doug,

Perhaps Rob can correct me if I'm misremembering but, having the motivation at the level of annotation was how we avoided this semantic conundrum.

Now we simply have nothing to tell them. That we can motivations on targets is the price we pay for the functionality that they want (essentially using annotations as bags of annotations). I expect it be such a seldomly exercised use case that collapses very easily into a multiplicity structure that we won't have to remark upon it at all.

Regards,

Jacob


On Tue, Sep 1, 2015 at 12:30 PM, Doug Schepers <schepers@w3.org<mailto:schepers@w3.org>> wrote:
Hi, Jacob–

On 9/1/15 12:52 PM, Jacob Jett wrote:
On Tue, Sep 1, 2015 at 10:06 AM, Robert Sanderson wrote:

    Except that as per 3.2.5, we might want to remove motivation from
    annotation completely.  Hence I left them off the examples.  Also
    the motivation on the Annotation would just be the set of
    motivations on the specific resources.

+1 for leaving motivation off of annotation. I think it makes more sense
to capture the motivations for why each of the bodies is present (i.e.,
how does it relate to the target).

I like how you put this. I think this is really a key to understanding the model, and a great way to explain it to developers and implementers (and power users, and extenders).

* A motive on a body is the way that each body relates to the targets.


Does the reverse hold true? I'm not sure which of these is more true:

* A motive on a target is the way that each target relates to the bodies.

* A motive on a target is the way that each target relates to the annotation.

* A motive on a target is the way that each target relates to the bodies and the annotation.

Would it depend on the motivation? I'd like to think not. I'm less curious about it a semantic level (though I suspect that's a question you'd like answered), but more at the level of clearly explaining it to the casual developer in a way that sticks.

Regards–
–Doug