Re: [w3photo] General update

Hi Libby,

>location
>entire image depicts something
>part of image depicts person
>entire image depicts event
>creative commons license for image and metadata
>thumbnail (maybe?)
>
>Some of these would be optional, some not (license must be compulsory I
>think).

Sure ... CC may be compulsory for www2004, but just recommended for common
use ?

Do you think to use 'creationEvent' for location ? I hope this also will be
determined in this discussion (not this week).


>I'd prefer annotator to be a person so we can provide a better
>identifier for them than name, but maybe that's not acceptable in the
>annotations schema?

Property for annotator (an:author) is taken from Annotea schema, and it is
defined just as subPropertyOf dc:creator, no range is specified. So,
foaf:People can be allowed as the object of an:author, but may have the
same confusion as dc:creator.

Also, I've noticed this property is rarely used in annotation, even in
Amaya output. I'm not sure how we should treat this one.


>I think something like this:
>
> <rdf:Description rdf:about=''>
>  <an:annotates
>rdf:resource='http://kanzaki.com/works/2004/imgdsc/040122_1340.jpg'/>
>  <an:author>Masahide Kanzaki</an:author>
>  <an:created>2004-01-22T16:05:44Z</an:created>
> </rdf:Description>
>
>is definitely required so that we can identify (and remove if necessary)
>annotations from a particular person or at a particular url.

Yes, possibly

 <an:author>
  <foaf:Person>
   <foaf:mbox_sha1sum>e8198...</foaf:mbox_sha1sum>
   ...
  </foaf:Person>
 </an:author>

although annotators may not want to give their mail addresses. It'll become
too complicated if mbox be optional to input annotation ?

>It does leave room for multiple pictures annotated in the same file (a
>good thing I think), although it won't allow annotations in the same
>file made at different dates. This could be avoided by allowing
>something like
>
><rdf:Description rdf:ID='anno1'>
>
>as well as
>
><rdf:Description rdf:about=''>
>
>This doesn't affect the rdf structures but might affect some
>RDF aggregator implementations which use a url as a way of identifying a
>selection of RDF. This is because the latter version links the document
>itself with the annotation, and the former does not. So the latter
>version (which is slightly restrictive) might be encouraged by us in
>case we have to swiftly remove data.

Hmm... the latter is simpler, and easier to implement.

If we follow Annotea syntax, I think the latter will become something:

<an:Annotation>
 <an:annotates rdf:resource="imageuri"/>
 <an:body rdf:resource=""/>
 ...
</an:Annotation>

In this sense, the former style will be:

<an:Annotation rdf:ID="annot1">
 <an:annotates rdf:resource="imageuri"/>
 <an:body rdf:resource="#annot1"/>
 ...
</an:Annotation>

But is this too tricky ... ?

cheers,

-- 
KANZAKI, Masahide a.k.a. masaka
http://kanzaki.com/info/webwho.rdf
mailto:webmaster(at)kanzaki.com
#Please use above address for a personal mail
#instead of post@kanzaki.com, which is for list only.
==================================
This is the TEMPORARY discussion list for the W3 Semantic-Photo History
Project. For questions, contact greg@fotonotes.net.

Subscribe Instructions
To:   semantic-photolist-request@unitboy.com
Body: subscribe

Unsubscribe Instructions
To:   semantic-photolist-request@unitboy.com
Body: unsubscribe

Help
To:   semantic-photolist-request@unitboy.com
Body: help

Received on Friday, 23 January 2004 08:59:22 UTC