Agenda

1. Logistics: Welcome, Minutes Review, Agenda Review, Chat Room
2. New entries in related ontologies page
3. W3C SSN XG progress report
4. Status of Action Items
5. Updates to Ontology
   1. Modifications: Changes proposed based on work to date
   2. Additions: Terms from the facets that have been added
6. Exercising the ontology in practical applications
7. End Logistics: Next meeting, Action items, Attendees

Minutes

Logistics: Welcome, Minutes Review, Agenda Review, Chat Room

Minutes review (note, the May 12/09 meeting was postponed; Last minutes are from the April 28/09 meeting).

Agenda approval

New related ontologies

The Device related ontologies page has been updated with entries for Semantic Sensor Network Ontology and  CESN Ontology.

W3C SSN XG progress

• Starting point seems to be based on CSIRO's Semantic Sensor Network Ontology
• They are developing use cases, ontology
• Their process may go faster than ours
• Need to think how to proceed / align with them?
  
  

Status of Action Items

Updates to Ontology

+ class Capabilities
+ object property: System hasCapabilities Capabilities (0..n)

+ class MeasurementCapabilities
+ object property: Sensor hasMeasurementCapabilities MeasurementCapabilities (0..n)
    hasMeasurementCapabilities is sub-property of hasCapabilities

+ class Frequency subclass of MeasurementCapabilities
+ class Accuracy subclass of MeasurementCapabilities
+ class Precision subclass of MeasurementCapabilities
+ class ResponseTime subclass of MeasurementCapabilities

This is the updated ontology: http://mmisw.org/ont/mmi/20090512T011137/device
and this the corresponding diagram:

Discussion:

1. Note: not all the contributions received so far have been included in the ontology, basically because i) we are going through an incremental, community process still, and ii) time constraints.
2. Some options from facets list have been added as subclasses of MeasurementCapabilities, but this may be changed later (they may be better captured with data properties for example).
3. Could Accuracy, Frequency, Precision be different from a Sensor to its Parent System?
4. What happens to the "Precision" under MesurementCapabilities when that data is passed on to another (compositional) system?
5. We could make the MeasurementCapabilities concept something that can be associated not only with sensors (or systems if we decide to generalize it) but also with other entities like data sets, measurements, etc.
6. Suggestion to associate MeasurementCapabilities to System, rather than Sensor, and exclude them, via restrictions, from some subclasses of System like Sampler. This could be simpler.
7. Once more: we need to make the concepts "atomic system" and "composite system" explicit in the ontology. The are properties that potential apply to all kinds of systems, but there are others that can be more appropriately associated with atomic systems in some cases, and others with composite systems in other cases.
8. What do you do with multiple-process devices? One can model this compositionally, according to one's needs.
9. Take a simpler example, where there is a single-sensor device. Should MeasurementCapabilities be associated with the whole unit, or just the sensor on it? Often the configuration happens at the level of the device, not the single sensor (transducer) on it. The current definition of sensor is ambiguous about whether it means to refer to just the transducer component, or the whole device.
10. Note that CSIRO model is an example of the model where sensor is more compositional, and the simple thing is called a transducer. But there is no compositional element to their model (except at the process level), so we can't tell whether sensor is compositional or not.
11. Carlos: Producing a few instances is welcome, so we can compare those concrete illustrations and thus enhance the process of creating the ontology.
12. Could we work through the SeaBird catalog and describe it? What is the process to produce an instance?  Protege would be used, yes.  Do the catalogs provide XML descriptions?
13. Goal of using the catalog is to review the abstractions and see if we have them right, at the highest level.
14. John agrees that all 3 of the modes: instantiation, high level abstraction, and continuing to model the facts, are potentially useful.
15. Nan: Do a simple example and a really complex one and one in the middle, but don't do the SBE catalog.
16. All: continue with facets, or do an instantiation if you can. Will review products at next meeting.
17. Keep the materials flowing to Carlos that you have produced.
18. Roy Lowry: I will try and look at a couple of examples and expose the results so we can decide if I'm helping or sending us on a wild goose chase.

Exercising the ontology in practical applications

1. See discussion about instantiations in the previous topic. Luis reiterated the need to have test cases. 

End Logistics: Next meeting, Action items, Attendees

Action Items

Next Telecon Time

Next telecon is Tuesday 2009-06-09 at 1500 UTC.

Attendees:

• Bruce Andrews
• Bob Arko
• Luis Bermudez
• Nan Galbraith
• John Graybeal
• Roy Lowry
• Carlos Rueda