Agenda

• Logistics: Welcome, Minutes Review, Agenda Review
• Status of last action items
• Review new version of the ontology
• Review Terms page
• Discuss proposed classifiers
• Tracking strategy: 1) Reactivate TRAC? 2) Activate SF tracking (since we're already using the SVN there)?

Minutes

Logistics: Welcome, Minutes Review, Agenda Review

Comments on Last Minutes/Last Meeting

Status of last action items

• [Carlos] Update SVN repository with versions discussed. 
  
• Done: the two ontologies (the OWL file and the diagram in both cases) were tagged 20080819_draft_2a and 20080819_draft_2b, respectively.
• deviceontterms.html (with the contents of the Terms page) was put under version control in SVN.
  
[Carlos] Update diagram per 2008-08-19 conversation 
• Done. Diagram attached below. Tag: 20080902_draft.
  

• [Carlos] Look at paper: Done. Extracted part of the OntoSensor ontology for comparison.

• [Bob] Look at paper

• [all] Contribute list of your known classifiers for sensors and samplers
• Roy Lowry sent this email with attached spreadsheet sampler_classification.xls. He says: "In summary, the most important categorisation is the nature of the sample collected, which needs to be described by a multi-level hierarchy." -- more details in the email.
  
• Jesper Zedlitz sent this email. : "I asked our Physical Oceanography research unit how they categorize their instruments." -- more details in the email.
• John Graybeal sent this email: "The classifiers for sensors on the Shore Side Data System mostly resolve around functionality.  We have two types of equipment, sensors and platforms." -- More details in the email.

Review new version of ontology

The changes incorporated in the new version are:

• Deployment no longer a subclass of Process.
  
• System is now a subclass of Process.
• Identification properties captured with a new Identifier class as a possible MetadataGroup associated with a process. (MetadataGroup comes from SensorML.)
  

• This is in general a bit closer to the SensorML model [1].
• The distiction between "atomic" and "composite" processes is explicit in SensorML.
  
• Also, contrast with OntoSensor [2].
  

References:

• [1] The diagram ontosensor3.png shows part of the OntoSensor model from here.

• [2] Carlos created a roughly corresponding class diagram from one of Mike Bott's slides (from MMI workshop last week) describing SensorML processes.
  

Comments during telecon

Review Terms page

• Need to update it to be consistent with current version of ontology.
  

Discuss proposed classifiers

• See referred emails above.  Need to specify a top-n classifiers.
  

Tracking strategy

• Not discussed.
  

End Logistics: Next Meeting, Action Items, Attendees

Next Telecon Time

Action Items

Attendees

• Bruce Andrews
• Luis Bermudez
• John Graybeal
• Carlos Rueda
• Jesper Zedlitz