SSN-XG Meeting Minutes 22 June 2010

Hi,

Thanks to those who attended the telephone conference last week.
And a special thanks to Kerry for scribing. 
The minutes of the meeting are here: http://www.w3.org/2010/06/22-ssn-minutes.html  

1 Follow-up from last week actions 
----------------------------------

1.1 Zakim status
----------------

The European Zakim bridge is still down - the advice from W3C is to access the teleconf via the US access point using a VoIP solution like Skype.


1.2 Status of ACTION-30 and ISSUE-9
-----------------------------------

- ACTION-30 Define the final place where the XG ontology will be delivered http://www.w3.org/2005/Incubator/ssn/track/actions/30 
- ISSUE-9 Ontology status after the closure of the XG 
http://www.w3.org/2005/Incubator/ssn/track/issues/9

Laurent commented on recent exchanges with W3C Team (Coralie Mercier and Thomas Roessler). Basically, there is a request for us to state what we want to do after XG close. 

General agreement between the teleconf participants that it is important:
1) to have W3C hosting the ontology in conditions maximising its impact(W3C-ish URI and physical location "equal" to namespace)
2) to plan now what the option to maintain the ontology after the end of the XG. Several projects will continue after the end of the XG.

First part of the discussion was about on the XG work should be extended in general. The normal rule is that at the end of an XG, the output of the XG is frozen. Then, the two usual continuation options are:
- start a new working group (in W3C jargon, this is called a (Recommendation Track) Working Group http://www.w3.org/2005/Incubator/REC-track.html
- do a member submission (there has been Member submissions for ontologies e.g. SIOC but none with a W3C-ish URI)

The third continuation option is if the XG participants make a strong commitment as a "community" to have it maintained at its W3C-ish URI with an arrangement which limits the workload of the W3C Team to a minimum. This approach is related to what W3C is doing for the organisation ontology submitted by Epimorphics because it is backed by the buoyant eGov and Linked Data community. 

So, this is definitively a "to be continued" discussion at several levels:
- technical details for the period to the closure of the XG and after it
- W3C-related follow-up action plan for the XG: in what forum?
- XG participants and community engagement behind the XG ontology in the long run


2. Ontology deliverable
-----------------------

Discussion in two parts: 
- progress of operating model sub-part (Michael) 
- Phenonet example (Laurent)
Plus a pass on the list of examples as a whole

2.1 Progress on operating model
-------------------------------

Michael commented very quickly what he has done (see http://www.w3.org/2005/Incubator/ssn/wiki/Operating_Model and  http://lists.w3.org/Archives/Public/public-xg-ssn/2010Jun/0024.html ). 

Payam was only present via IRC so this discussion will have to be carried to next meeting because. 

Let me highlight Kevin's remarks about re-factoring some aspects of the proposed ontology (somewhat echoing the one we had on measurement Capabilities)

kevin: are any of the constraints used over different ranges or only in a particular context? 
... are has-constraints intended to be used only some ranges. Why not subclasses of those things?
... e.g enviromental survival range and env operating range seem to be specific constraints. Is there a contraint that can be used in all places?

Please contribute to this discussion if possible with better examples or references of similar work. 

2.2 Phenonet example
--------------------

I have explained the scope of the Phenonet example using Kerry's diagram of the architecture
http://www.w3.org/2005/Incubator/ssn/wiki/Architecture

My work until know has focused on System --> Sensor --> Parameter aspect. 
I intend to extend what I've done to cover the deployment aspects of the ontology too.

In parallel, I have assembled some external ontologies to populate this example:
- units of measurement based on the SysML-QUDV ontology design pattern 
- physical quantities (Parameters, variables) based on CF 
I am also evaluating the NEESGRID ontology as the source of definitions for the product metadata and for spatial coordinates.

Finally, I have explored the ontology delivery packaging options 
- distribute the examples in independent parts 
- or bundle everything together in a single OWL file 
I want to have documented conventions which allow to use the ontology modules, separately or put together in a single file,  with Protégé, Top Braid, a triple store (e.g. SESAME) or even to load it directly to a DL reasoner like Pellet.

The wiki workspace for the Phenonet example is: http://www.w3.org/2005/Incubator/ssn/wiki/Agriculture_Meteorology_Sensor_Network


2.3 Review of the examples list
-------------------------------

My priority is to identify the "most valuable examples" which should be delivered through the XG-managed workspace. 

So I have updated the list of examples 
http://www.w3.org/2005/Incubator/ssn/wiki/Semantic_Sensor_Net_Ontology#Examples
Thanks to David and Kevin for their inputs. 

3. Any Other Business
---------------------

Kerry has been in contact with Open University (UK) and hopes that this W3C member will delegate some participants to the XG. 

Kevin requested at the teleconf to have a more detailed timeline up to end of the XG, especially to help the XG to "survive" (not his words) the summer holiday impact on the rate of progress. 

I agree this is an important issue so I gave a brief outline of the work plan I have in mind. Here are some complements to what I said at the teleconf. 

During the last month,my planning priority has been the ontology deliverable. For the forecoming months, I want to have two release steps for it (mid-July and mid-August). I'll continue to create targeted issues and actions for what I think is on the critical path and also to help everyone aware of our progress (closure of specific work items).

For the other deliverables, one release step should be enough (dates to be defined):
- The Semantic Markup deliverable is roughly under control - some of the missing bits were discussed last week especially the examples.
- For the XG report, the part which is on the critical path is the "What happens after the XG" bit. We need to start this part as soon as possible. 


Cheers
Laurent

Received on Thursday, 24 June 2010 05:31:43 UTC