- From: Bob Morris <morris.bob@gmail.com>
- Date: Sun, 5 Aug 2012 14:58:40 -0400
- To: shannon.bradshaw@gmail.com
- Cc: Robert Sanderson <azaroth42@gmail.com>, public-openannotation <public-openannotation@w3.org>
But IMO it doesn't resonate with the needs of data annotation at the record level. In our case this is potentially the digital records describing each the 2.5 billion specimens estimated to be held by the world's natural history collections. The current some several hundred million such digital records (the rest are on paper) are littered with errors for many reasons, and we are using machine- and human-generated annotations to make assertions about how erroneous, incomplete, or conflicting records can be transformed into those more fit for one or another purpose (as well as to help accelerate the 10-year worldwide project to get the rest of the specimen descriptions digitized). In the community doing this, there are several important XML-Schemas and RDFS or OWL DL ontologies that have been in play for quite a few years and there is \typically/ a need for our annotations to express rather arbitrary transformations, or at the very least, for it to be machine-deducible that the annotations are germane to a transformation of the oa:Target. <StrongOpinion> Basically, our current need is for data annotation to address fitness-for-purpose, and my guess is that most people annotating documents also have that motivation. But it's hard to see how to model fitness-for-purpose without reference to knowledge representation in the domain of the Body and Topic. From this perspective, I continue to believe that Style doesn't belong in an annotation knowledge representation---I see it as just a tool based on thousands of years of document production, by which an ao:Annotator is hiding some fitness-for-use concept that is potentially integrable with someone else's were it only clearer why the Annotator designated, or cared about, that style. But, if you are able to, e.g. express that your red stuff is meant to denote that this part of the document signals something the consuming agent should somehow care about, why shouldn't that concern be expressed with something less context sensitive than "text has red background color". </StrongOpinion> To the extent that my StrongOpinion analysis is shared, it is perhaps an argument that Style belongs in oax. Bob Morris On Tue, Jul 31, 2012 at 4:19 PM, Shannon Bradshaw <shannon.bradshaw@gmail.com> wrote: > On Mon, Jul 30, 2012 at 12:19 PM, Robert Sanderson <azaroth42@gmail.com> > wrote: >> >> >> Instead of attaching to the Specific Resource, oa:hasStyle would >> attach to the Annotation. > > > +1 from me. > >> >> The current requirements from the use cases and stakeholders are only >> for describing stylistic or rendering features, rather than arbitrary >> transformations. For example, red strike-through and yellow >> background is required, but we don't have a strong case for >> transformation of arbitrary XML into HTML or JPEG into PNG. > > > This description resonates with the needs of our user community. > > -Shannon > > -- Robert A. Morris Emeritus Professor of Computer Science UMASS-Boston 100 Morrissey Blvd Boston, MA 02125-3390 IT Staff Filtered Push Project Harvard University Herbaria Harvard University email: morris.bob@gmail.com web: http://efg.cs.umb.edu/ web: http://etaxonomy.org/mw/FilteredPush http://www.cs.umb.edu/~ram === The content of this communication is made entirely on my own behalf and in no way should be deemed to express official positions of The University of Massachusetts at Boston or Harvard University.
Received on Sunday, 5 August 2012 18:59:08 UTC