W3C home > Mailing lists > Public > public-openannotation@w3.org > October 2012

Sanity check Re: F2F Decision: Multiple Resources

From: Bob Morris <morris.bob@gmail.com>
Date: Sun, 28 Oct 2012 17:49:43 -0400
Message-ID: <CADUi7O7duehUy1hWaKa8JoQ5qWhh_SjKr6p6PON3KA9bswm7ug@mail.gmail.com>
To: public-openannotation <public-openannotation@w3.org>
Following Rob's posting of Oct 1, 2012 [1]  explaining  the Chicago
F2F meeting  decision about Multiple Resources, I ask whether the
models I have below are correct and appropriate for the following use
case: We want to annotate several SpecificResources (in our case some
database records) all with the same Source (a certain dataset with a
URI), and to all of which the assertions of a single common Body are
to apply. The oa:Selectors are queries which happen to be the same
except that they differ in some of the bound values in the query.  One
could think of the same problem as one of annotating a spreadsheet to
make simultaneous assertions about all rows that have specific,
different,  values in particular columns.

In the pre-Multiple Resources days, OA permitted multiple targets but
only one body, so for the use case one could write
_:x a oa:Annotation;
  oa:hasBody <theBody>;
  oa:hasTarget <spTarget1>;
  oa:hasTarget <spTarget2>.

<spTarget1> a oa:SpecificResource;
  oa:hasSource <theSource>;
  oa:hasSelector <selector1> .

<spTarget2> a oa:SpecificResource;
  oa:hasSource <theSource>;
  oa:hasSelector <selector2> .

In this case it was (implicit? explicit?) that <theBody> applied to
all the Targets.

Post the F2F decision,  I can see (at least) two solutions, depending
on whether one models with an oa:Set of SpecificTargets each with its
own Selector, or a single SpecificTarget   with an oa:Set of
Selectors.  Are both of the models _:y  and _:z below consistent with
the intent of oa:Set? Are they semantically equivalent?

_:y a oa:Annotation ;
   oa:hasBody <theBody> ;
   oa:hasTarget <theTargets> .

<theTargets> a oa:Set ;
   oa:item <SpTarget1> ;
   oa:item <SpTarget2> .

<SpTarget1> a oa:SpecificResource ;
 oa:hasSelector <Selector1> ;
 oa:hasSource <theSource> .

<SpTarget2> a oa:SpecificResource ;
 oa:hasSelector <Selector2> ;
 oa:hasSource <theSource> .

------------
_:z a oa:Annotation ;
   oa:hasBody <theBody>;
   oa:hasTarget <theTarget> .

<theTarget> a oa:SpecificResource ;
  oa:hasSource <theSource> ;
  oa:hasSelector <theSelectors> .

<theSelectors> a oa:Set ;
  oa:item <Selector1> ;
  oa:item <Selector2> .

------------

One observation: if these are correct and equivalent, they illustrate
something that feels generically true and intuitively obvious: for
serialized OA, but possible not for the RDF graphs, the further down
the Annotation tree you put the container, the smaller the
serialization.



[1] http://lists.w3.org/Archives/Public/public-openannotation/2012Oct/0004.html

Thanks for comments
Bob Morris


-- 
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, 28 October 2012 21:50:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 28 October 2012 21:50:11 GMT