- From: Bob Morris <morris.bob@gmail.com>
- Date: Tue, 21 Aug 2012 16:22:57 -0400
- To: Robert Sanderson <azaroth42@gmail.com>
- Cc: Leyla Jael García Castro <leylajael@gmail.com>, t-cole3@illinois.edu, public-openannotation <public-openannotation@w3.org>
I (at the moment think) I don't have a problem with deprecated structured bodies, because although structured bodies are critical to us, the structure is essentially all in the domain vocabulary and consumers have to be able to make sense of that in the case of actionable Annotations, i.e. those where the producer is annotating a mutable resource that is under the control of a consumer, and the producer is expressing a desire (and justification) for some change to be made on the resource. For, say, googledocs, this would be as though the document supported arbitrary marginal annotations by anyone, but doc change only by the originator. Even if you could do that, the annotation model of gdocs maybe isn't adequate to annotation "conversations", about which more anon. In fact, the way we generate annotations right now, each gets its own UUID-based guid. The "external" stuff lives in a graph in the same namespace and if the annotation moves around in a serialization that includes the structure of the objects in the domain vocabulary, then the consumer and producer can extract the domain content by mutually agreed upon IETF resolutions. There will be issues for those who believe that http URI's must have an IETF resolution that is character-wise the same as the URI :-) That is in practice never the case for us, and of course it leads to complexities for any consumer that needs to extract all the body structure down to a fixed level. But sometimes the extraction is potentially simple if it is based on the (very naughty) resolution based on matching leading strings in the URIs, e.g. CONSTRUCT {?s ?p ?o} WHERE { ?s ?p ?o . FILTER (regex(str(?s), "http://etaxonomy.org/ontologies/ao/ad2db960-c91c-4fd7-bac1-c1ae0ce4bba4")) } Of course, in our environment we have to provide that the consumer can have assurance that the annotation is exactly as produced or at least that the consumer is in a position to further filter out other matches to the regex. We do this with digital signatures included with a wrapper around the serialized Annotation. On Tue, Aug 21, 2012 at 3:30 PM, Robert Sanderson <azaroth42@gmail.com> wrote: > > Hi Leyla, > > I put Stian's name there, regarding his suggestion to deprecate structured > bodies and only have references to external graphs. > > Herbert and I are (of course) very interested in the issue too, and > hopefully it will be informed by the resolution of the multiple-body issue. > > Rob > > > On Tue, Aug 21, 2012 at 12:54 PM, Leyla Jael García Castro > <leylajael@gmail.com> wrote: >> >> Hi there, >> >> @ Stian Soiland-Reyes and Tom Habing: I just saw you are working on >> Semantic/Graph annotations. Do you already have a document to share? I would >> like to know more about the discussion around this particular subject. >> >> Cheers, >> >> Leyla >> >> On Fri, Aug 17, 2012 at 7:02 PM, Tim Cole <t-cole3@illinois.edu> wrote: >>> >>> For those of you not on Tuesday's OA Community Group conference call, I >>> wanted to call your attention to the 'Focus Group' agenda item from the call >>> and to the Open Issues wiki page >>> (http://www.w3.org/community/openannotation/wiki/Open_Issues). >>> >>> >>> >>> The Open Issue page is helping us track issue discussions and concrete >>> proposals for modifications to the Core and Extension Specifications such as >>> are currently being discussed on this listserv. >>> >>> >>> >>> Focus Groups: Most Open Issues list one or more individuals who have been >>> tasked with creating a wiki page summarizing issue discussions and laying >>> out any consensus recommendation(s) for action (or inaction) needed to >>> resolve a particular issue. See for example: >>> http://www.w3.org/community/openannotation/wiki/Style >>> >>> >>> >>> In order to make the upcoming f2f meeting in Chicago as productive as >>> possible, we are asking issue 'Focus Groups' to post pages (and notify using >>> this listserv) by no later than 5 September. This will provide time for >>> Community comment prior to finalizing meeting agenda. Typical issue >>> recommendations / resolutions that we expect to emerge from this process: >>> >>> · discussion concluded, no action required; >>> >>> · no spec change required, but propose adding text to FAQ or >>> examples to Cookbook; >>> >>> · keep issue open while new properties and/or classes are >>> suggested and considered; >>> >>> · make a specific change to the one or both specification >>> documents. >>> >>> This list of kinds of recommendations is not comprehensive. >>> >>> >>> >>> Process is informal by design and please feel free to join in – add your >>> name, add an issue. All OA Community Group members should be able to edit >>> wiki pages. But we do ask that you not get carried away, either by proposing >>> too many new issues or by trying to be part of every single Focus Group. >>> Concentrate on the most important, highest priority – the biggest bang for >>> your time and effort. >>> >>> >>> >>> See also the Conference Call notes that have been posted >>> (http://www.w3.org/community/openannotation/wiki/Telcon_2012-08-14) >>> >>> >>> >>> Thanks, >>> >>> >>> >>> Tim Cole >>> >>> University of Illinois at UC >>> >>> >> >> > -- 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 Tuesday, 21 August 2012 20:23:25 UTC