- From: Paolo Ciccarese <paolo.ciccarese@gmail.com>
- Date: Fri, 5 Jun 2015 14:52:50 -0400
- To: Robert Sanderson <azaroth42@gmail.com>
- Cc: Susanna Netseven <martinelli@netseven.it>, Annotation WG <public-annotation@w3.org>
- Message-ID: <CAFPX2kBvqFfXg0KOWh5eeZj+1fWF-4Lgv8_uQ0Cn1bjhsYw6ZA@mail.gmail.com>
Or even better a specialization of the oa:linking, which is by definition 'untyped'? I believe references/citations are a very good widespread example we should provide guidance on. On Fri, Jun 5, 2015 at 2:49 PM, Paolo Ciccarese <paolo.ciccarese@gmail.com> wrote: > I am wondering if oa:linking is another option. > > If I select a reference I can either say: > - this text is described by this set of triples representing the citation > (oa:describing) > - this text represents a link of this document to an external document > (oa:linking?). This document cites another document through the citation > text that works as a proxy > > The second is expressing more explicitly the linking nature of the > annotation. > Thoughts? > > > > On Fri, Jun 5, 2015 at 2:35 PM, Robert Sanderson <azaroth42@gmail.com> > wrote: > >> >> Hi Susanna, all, >> >> I don't think that tagging is appropriate for that particular use case, >> as it would imply that the body tags the target. Personally, I would say >> that it describes the target, using a machine readable language rather than >> a human language. >> >> For example, an equivalent annotation with the body "This paper cites >> another paper" and the annotation with the body "<paper1> cito:cites >> <paper2> ." have the same actual intent, just different intended audiences. >> >> Does that make sense? >> >> Rob >> >> >> >> On Fri, May 29, 2015 at 1:38 AM, Susanna Netseven <martinelli@netseven.it >> > wrote: >> >>> Hi again, >>> we are discussing about the correct interpretation about motivatedBy >>> predicate. >>> >>> Our Web Annotation model tries to fit both implementation and model >>> correctness needs. >>> At the moment we use body as a graph, cause we provide a client that >>> permits a sort of semantic tagging about the target (that we used to >>> identify as *oa:SpecificResource*) >>> Our body graph contains triple as: >>> >>> <target1> <predicate> <object> >>> >>> where <predicate> could be any of those predicates:* cito:cites*, >>> *cito:describes*, *cito:repliesTo,* *rdfs:comment*, >>> *cito:includesQuotationFrom *and predicates used for >>> other purposes for example temporal tagging, spatial tagging, >>> identifying target as LOD resource (<target> <identifyAs> <dbpedia:Paris>), >>> identifying target as a Class (<target> <dcterms:type> <foaf:Person>). >>> >>> We are discussing about a particular motivation that could cover all of >>> these (and other) >>> cases, and from our point of view the correct one could be *oa:tagging*. >>> Because actually we are doing semantic tag about a resource. >>> Is that correct? >>> >>> Susanna >>> >>> >>> >>> >> >> >> -- >> Rob Sanderson >> Information Standards Advocate >> Digital Library Systems and Services >> Stanford, CA 94305 >> > > > > -- > Dr. Paolo Ciccarese > ORCID: http://orcid.org/0000-0002-5156-2703 > -- Dr. Paolo Ciccarese ORCID: http://orcid.org/0000-0002-5156-2703
Received on Friday, 5 June 2015 18:53:19 UTC