Re: Application of OpenAnnotation to Earth Observation and Climate Science use case

Hi Rob,

Thanks for your help with this.

I think for reasons of simplicity option 1) works best.  In my original e-mail I'd meant to suggest using tagging.  This seems to fit to me.  - We would set oa:motivatedBy to oa:tagging and use an oa:SemanticTag as the body pointing to a skos concept from a given collection.  Does that make sense?  (bear with me, a lot of this is still new to me :)

Cheers,
Phil

From: Robert Sanderson <azaroth42@gmail.com<mailto:azaroth42@gmail.com>>
Date: Thu, 10 Oct 2013 11:56:06 -0600
To: Philip Kershaw <philip.kershaw@stfc.ac.uk<mailto:philip.kershaw@stfc.ac.uk>>
Cc: public-openannotation <public-openannotation@w3.org<mailto:public-openannotation@w3.org>>
Subject: Re: Application of OpenAnnotation to Earth Observation and Climate Science use case


Hi Phil,

That's a great question, and thanks for letting us know about your project :)  Please keep us informed as it progresses!

A couple of things to think about with the modeling question of where to attach the information ...

1.  If the body were used in another annotation, would it always be true that it has the same research domain?

I can imagine situations where in one annotation the same body is linked to a dataset because of its relevance in one domain, but to another dataset because of some other reason.  Then in a triple store it would be impossible to separate them back again to their original annotations.  If this is a real situation, then I think it has to go on the annotation.  I agree about the relationship to Motivation, but that it's not quite the same thing.  (For others, it's also clearly not an Expectation)

2.  If there were multiple bodies, would they all have the same research domain, or would it be permissible in your context to have (for example) three bodies of a single annotation, each with a different domain?

This is more about workflow and processing, IMO.  You may consider it important that three associations all remain together in a single Annotation.  If this is the case, then it wouldn't be possible to distinguish which body had which domain if the domain is on the Annotation.  If the situation in 1 above is also true, then the rather verbose solution would be to associate the information with a SpecificResource.

So the three options:

On the Body, because it's always true:
{
  "@id" : "http://annotations.stfc.ac.uk/datasets/1/annotation1",
  "hasTarget" : "http://datasets.stfc.ac.uk/datasets/1",
  "hasBody" : {
    "chars" : "... sea surface temperature has risen ...",
    "dcterms:subject": "http://ontology.org/terms/seaSurfaceTemperature"
  }
}

On the Annotation, because it's not always true, but only ever one:
{
  "@id" : "http://annotations.stfc.ac.uk/datasets/1/annotation1",
  "dcterms:subject": "http://ontology.org/terms/seaSurfaceTemperature",
  "hasTarget" : "http://datasets.stfc.ac.uk/datasets/1",
  "hasBody" : {
    "chars" : "... sea surface temperature has risen ..."
  }
}

Or on a Specific Resource, if neither:

{
  "@id" : "http://annotations.stfc.ac.uk/datasets/1/annotation1",
  "hasTarget" : "http://datasets.stfc.ac.uk/datasets/1",
  "hasBody" : {

    "chars" : "... sea surface temperature has risen ...",
    "dcterms:subject": "http://ontology.org/terms/seaSurfaceTemperature"
  }
}












On Wed, Oct 9, 2013 at 11:06 AM, <philip.kershaw@stfc.ac.uk<mailto:philip.kershaw@stfc.ac.uk>> wrote:
Hi all,

It would be great to get your thoughts on this.  - I'm working on a project called CHARMe, where we are using the Open Annotation data model to make annotations of climate science and earth observation datasets.  A number of the use cases we have can be implemented by applying OA without any extension.  However,  we have a requirement to associate a research domain with any given annotation.  This item would be a restricted list of terms.  It is analogous but not the same as the concept of an oa:Motivation.   It could for example include terms such as like land cover, sea surface temperature or sea ice extent.   There a number of different vocabularies in the Earth Sciences that we could apply here.  My question is, would this be best expressed as an explicit new property of an annotation or as an annotation body?

Cheers,
Phil


--
Scanned by iCritical.



-- 
Scanned by iCritical.

Received on Wednesday, 23 October 2013 10:59:38 UTC