W3C home > Mailing lists > Public > public-openannotation@w3.org > March 2013

Re: May a Specific Resource be object of the oa:hasScope relationship ?

From: Lutz Suhrbier <l.suhrbier@bgbm.org>
Date: Thu, 21 Mar 2013 17:43:54 +0100
Message-ID: <514B38CA.6000109@bgbm.org>
To: public-openannotation@w3.org
Hi Bob,

regarding your use case, you are right that there is no further 
advantage of aggregating T1,T2;B1,B2 in a single annotation.

But, if we exchanged the manuscripts by two dataset elements describing 
a geographical position by longitude and latitude values, and the 
annotator observes that they are not matching the location information 
also given in that dataset, then he has to propose new values for both 
(long and lat) values in order to define the new (correct) position.
And, you can not subdivide it in two individual annotations, because 
they have no relationship to each other. If a curator examines and 
verifies only one of them(e.g. longitude), then he will conclude that 
the corrected value will not make the position pointing at the correct 
location, and thus, probably not oa:editing the wrong original data.

We have lots of those data element interrelationships in our data. So, I 
don't see a way to omit usage of Composites and hasScope in the use 
case(s) of both our projects.

best regards
Lutz

> I imagine Lutz's underlying question (at least, it would be mine, and
> his and our project have similar goals) is something like this:  When
> an annotation is actionable, is it reasonable for a community of
> practice to act as though the actions derived from B1 are applicable
> only to T1, those for B2 only to T2, etc.
>
> One thing that is a little unclear to me is what Lutz means to model
> by the bundling that would make it different from modeling each
> Target/Body as a separate annotation.  Presumably, that is what must
> be preserved by an interpretation of an action to be taken.
>
> I wonder if the following is a simple scenario, motivated by
> oa:editing,  about which the Context model can be applied for
> examination:
>
> T1 a manuscript meant to be with UK English spelling
> T2 a manuscript meant to be the same work  differing only in  U.S.
> English spelling
>
> B1 carrying an assertion that a specific occurrence of "color" should
> be "colour"
> B2 carrying assertion that a specific occurrence, \\\corresponding to
> the above one/// of "colour" should be "color"
>
> For this scenario, I can't exactly see why the motivation oa:editing
> wouldn't just as reasonably be met with two separate annotations, so
> it needs more(?) to fit Lutz' problem.
>
>
>
> On Thu, Mar 21, 2013 at 8:15 AM, Robert Sanderson <azaroth42@gmail.com> wrote:
>> Hi Lutz,
>>
>> Sorry, travel got in the way!
>>
>> Yes, you're correct with what you describe -- it's a hint that the
>> object of hasScope is some container or context in which the subject
>> appears, and not a validity constraint on the annotation.
>>
>>
>> So if you had a composite body, of B1 and B2, and a composite target
>> of T1 and T2, with hasScope between B1/T1 and B2/T2, the semantics are
>> the following statements:
>>
>> The set of (B1,B2) annotates the set of (T1, T2), as potentially
>> further clarified by the motivation.
>>
>> Plus:
>> B1 is in the context of T1.
>> B2 is in the context of T2.
>>
>> But these two don't change the first statement.
>>
>> Thanks for your patience!
>>
>> Rob
>>
>> On Thu, Mar 21, 2013 at 12:08 PM, Lutz Suhrbier <l.suhrbier@bgbm.org> wrote:
>>> Hi Rob,
>>>
>>> sorry, did I raise some internal discussions about the right interpretation
>>> of the oa:hasScope to SpTarget construction within a SpBody ?
>>>
>>> Otherwise, I would appreciate If you could make a statement if my
>>> interpretation of it's meaning is correct ("when the annotator created that
>>> SpBody, he was examining the SpTarget the oa:hasScope property connected
>>> with SpBody refers to") ?
>>>
>>> Sorry, I don't want to push you, but we would need that information for our
>>> project internal discussion on how to finally realise our use case with OA.
>>>
>>> best regards
>>> Lutz
>>>
>>>
>>>
>>>> Hi Rob,
>>>>
>>>> hmm, you're right. It does not really solve my problem. But may be, I can
>>>> live with a certain level of weakness.
>>>>
>>>> So, if I want to keep may construction using oa:hasScope to express my use
>>>> case, what I get is only a weak relationship between the scoped body and the
>>>> sptarget.
>>>> It says something like, "when the annotator created that body, he was
>>>> examining the part of the document which was selected within the scoped
>>>> sptarget."
>>>> Then, it is up to the application how it will interpret that kind of
>>>> "hint". It is up to the application to decide, if it wants to respect the
>>>> potential intentions of the annotator creating that body while examining a
>>>> specific part of the target. Or, to not respect them and then possibly
>>>> visualize some nonsense to the client user.
>>>>
>>>> Is that a correct interpretation of what I would get if i kept my
>>>> construction like it is ?
>>>>
>>>> Lutz
>>>>
>>>>
>>>>
>>>>> Hi Lutz :)
>>>>>
>>>>> On Mon, Mar 18, 2013 at 1:27 PM, Lutz Suhrbier <l.suhrbier@bgbm.org>
>>>>> wrote:
>>>>>> Only regarding choices, I understood that at maximum one of the specific
>>>>>> resources can be selected by the client application. Your definition
>>>>>> below
>>>>>> sounds like it may be more than one ?
>>>>> On a semantic level they all apply, but the importance is that you
>>>>> only have to display one of them.  For example, it's true that the
>>>>> English version and the French version of a comment both apply to the
>>>>> target, but you only need to show one of them to the user as they have
>>>>> equivalent content.
>>>>>
>>>>>
>>>>>> As in my use case, all SpTargets must be applied, and also all SpBodies
>>>>>> must
>>>>>> be applied as well, the Composite appears to be the right concept to
>>>>>> express
>>>>>> this.
>>>>> Yes, but semantically the body composite applies to the target
>>>>> composite. You need to have all of the resources in each composite in
>>>>> order for the annotation to make sense.
>>>>>   From the spec:  "A Composite is a set of resources that are all
>>>>> required for an Annotation to be correctly interpreted."
>>>>>
>>>>> So the body composite is a set of resources that, as a set, annotates
>>>>> the target.
>>>>>
>>>>>
>>>>>> What's about the interpretation of oa:hasContext related to a SpTarget
>>>>>> within SpBodies in general, and within SpBodies in a Composite in
>>>>>> particular.
>>>>> There isn't any special interpretation of hasScope for when the
>>>>> subject and/or object also have roles in particular annotations.
>>>>>
>>>>> The, admittedly quite fuzzy and intentionally so, description from the
>>>>> spec is that hasScope is "... to capture the context in which it was
>>>>> made, in terms of the resources that the annotator was viewing or
>>>>> using at the time" and that it does not make the relationship
>>>>> expressed by the annotation valid only with the resources that were
>>>>> being viewed.
>>>>>
>>>>> So if you have an image with hasScope to a web page as the target.
>>>>> And the comment is "I like this image", that just says that the image
>>>>> appears within the page, and that you like the image, not that you
>>>>> like (the image in the page).
>>>>>
>>>>>> the image specified in the SpTarget must be seen by the client
>>>>>> application
>>>>>> in the context of the web page.
>>>>> The spec says:
>>>>> "This does NOT imply an assertion that the annotation is only valid
>>>>> for the image in the context of that page, it just records that the
>>>>> page was being viewed."
>>>>> (caps added for emphasis)
>>>>>
>>>>> Which definitely could be clearer, of course.
>>>>>
>>>>>> If I related now a Body to a SpTarget, than it would mean that
>>>>>> the content of that body must be seen within the context of that
>>>>>> SpTarget,
>>>>>> or not ?
>>>>> Thus, no :(
>>>>>
>>>>> Hope that ... sort of ... helps, even if it doesn't solve the problem?
>>>>>
>>>>> Rob
>>>>>
>>>>
>>>>
>>>
>
>
Received on Thursday, 21 March 2013 16:44:25 GMT

This archive was generated by hypermail 2.3.1 : Thursday, 21 March 2013 16:44:25 GMT