- From: Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>
- Date: Mon, 28 Jan 2013 15:22:30 +0000
- To: Leyla Jael García Castro <leylajael@gmail.com>
- Cc: public-openannotation <public-openannotation@w3.org>
On Mon, Jan 28, 2013 at 11:48 AM, Leyla Jael García Castro <leylajael@gmail.com> wrote: > About the provenance, I understand that a complete mapping to W·C PROV is > provided in Appendix A. But some of it is also mentionen in the Description > column in the table with the properties. Are oa:annotatedAt and > oa:serializedAt subclasses of prov:generatedAtTime? If that is the case, > either you explicitely include the subclassing in the description for all > properties or none of them. Otherwise it could be confusing. No, in Chicago we concluded that oa:annotatedAt and oa:serializedAt can't both be subproperties of prov:generatedAtTime, as that would imply that annotation and serialization is the same event. Doing it formally, we are talking about two different entities, one that is the concept of the annotation (the oa:Annotation, say <annotation1.rdf#ann1>), which is oa:annotatedAt whenever it was structurally formed. Then there's the serialization, say <annotation1.rdf>, which might have been made much later. The oa:serializedAt from oa:Annotation is a shortcut to describe this entity from an oa:Annotation. To avoid confusion, neither was made subproperty of prov:generatedAtTime. It is OK in PROV for an entity to be generated by multiple activities, but the generation event is meant to be the same. (For instance pushing a file to an SVN server and the more abstract "Publishing file" can be two different activities, but they both must agree on when that happened. So if we have two different timestamps this would get a bit inconsistent. Here the two activities can be thought of as the narrow "Making this serialization" and a longer lived "Publishing the annotation". > Nice to hear that! I will probably contact you guys as I have some > questions/comments about PAV. Is there any draft that you can share? A while > ago I was looking for a link to an updated version with no luck, but I have > not tried lately. I'm sorry, PAV did not have a proper HTML representation until this upcoming 2.1. http://pav-ontology.googlecode.com/svn/branches/2.1/pav.html is the current draft in HTML form, based on http://pav-ontology.googlecode.com/svn/branches/2.1/pav.owl is the OWL. Feedback welcome - but sooner rather than later! ;-) I'm hoping to put this all on http://purl.org/pav after finalizing some of the descriptions. Any hints on how to do an XSLT transform or similar magic from the RDF/XML to do some kind of redirect to an HTML page are welcome; as purl.org does not do content negotiation. -- Stian Soiland-Reyes, myGrid team School of Computer Science The University of Manchester
Received on Monday, 28 January 2013 15:23:22 UTC