See also: IRC log
<Holger> ScribeNick: kelsey
Holger: meeting minutes will rotate via the scribe list
<scribe> ACTION: Holger to edit wiki to reflect scribe list [recorded in http://www.w3.org/2009/08/12-ssn-minutes.html#action01]
Holger: need to have *some* representation of O&M in the ssn ontology
JohnGraybeal: this group should not be focused on describing data sets (though device data will need to be described)
cory: perceived sensor web enablement was the focus, should in part focus on sensors
payam: device vs. phenomena - not interested in device, rather what is being measured
<michael> sorry lost connection
michael: ssn's may be used for phenomena measurement, but also sensor capabilities - hence the need to characterize sensors
payam: Quality of information relative to the device - this is part of observation of measurement or device description?
michael: observation & measurement should reference back to the sensor to avoid redundancy in info reporting
krp: need to have link between the data and the sensor
cory: Quality is described in the SensorML spec (e.g. values generated from sensor should not be trusted if temperature falls outside specific bounds), this *could be* sensor specific and not observation
bermudez: SoS points to link in SensorML doc
payam: How does the observation association relate to the temporal transience of the sensor (e.g.what if the sensor device changes quality over time)
<JohnGraybeal> Can someone point me to our statement of intent to annotate SWE data using a data ontology?
michael: attach sensor historical information in observation vs. time tag in observation used to correlate with device managed history
Payam: what if sensor is no longer available? Accuracy lookup may be lost
Arthur: UID and sensor quality can be sent
<Holger> ack: JohnGraybeal
Payam: this suggest an archiving approach
<krp> Yes, much of this sounds an implementation detail - if the resource representing the sensor doesn't persist I'm not sure there's much we can do about that in the ontology
JohnGraybeal: ontology should focus on device (and data, if desired) description - not system design
<krp> (sorry, meant that if the sensor doesn't persist that doesn't mean the resource representing it doesn't persist)
<michael> agreed we can't 'solve' semantic sensor networks, just build an ontology - but do need some use cases
Holger: target of this discussion is do we need O&M (description of data) in the device ontology
agree with michael - let the use cases drive this
<JohnGraybeal> I would have said do we need an O&M ontology, not does it need to be *in* the device ontology
<michael> we can just import odoe or similar and start there
<krp> I think that a valid point on email was that there isn't much to/in the O&M model
Holger: Linkage of O&M in the
ontology will be discussed in the ontology working group
... what approach should be used for identification/selection of use case drivers?
<michael> what happened to Simon's use cases?
<scribe> ACTION: Holger will follow up with Simon re use cases [recorded in http://www.w3.org/2009/08/12-ssn-minutes.html#action02]
JohnGraybeal: can use cases successfully drive development?
<krp> Are the use cases for ontology validation? Or for analysis to develop the ontology from? Is the question how rigorous or complete the coverage of the use cases needs to be such that they're useful to us?
JohnGraybeal: team could id key ontology requirements (or existing use cases) vs. taking the time required to develop reasonable use cases
<krp> (so yes, as John just said)
<JohnGraybeal> Yes, those are the right questions krp.
<JohnGraybeal> I would be willing to build a "what do you want the ontology for" page to answer the question...
<JohnGraybeal> ... (the question of what the priority uses should be)
<scribe> ACTION: All to send desired ontology capabilities to the mailing list [recorded in http://www.w3.org/2009/08/12-ssn-minutes.html#action03]
<michael> good idea
<scribe> ACTION: JohnGraybeal will create a page on the wiki with brief description of use cases (what the ontology should be used for), send here rather than the mailing list [recorded in http://www.w3.org/2009/08/12-ssn-minutes.html#action04]
put on the wiki directly, send it to the mailing list or send directly to John
<krp> Thanks, bye.
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Found ScribeNick: kelsey Inferring Scribes: kelsey WARNING: No "Topic:" lines found. Default Present: Holger, giorgi, +1.206.662.aaaa, Arthur, krp, michael, +1.202.408.aabb, +1.650.450.aacc, cory, Rodrigo Present: Holger giorgi +1.206.662.aaaa Arthur krp michael +1.202.408.aabb +1.650.450.aacc cory Rodrigo Regrets: AmitS Krzysztof Victor Kerry Got date from IRC log name: 12 Aug 2009 Guessing minutes URL: http://www.w3.org/2009/08/12-ssn-minutes.html People with action items: all holger johngraybeal WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. WARNING: No "Topic: ..." lines found! Resulting HTML may have an empty (invalid) <ol>...</ol>. Explanation: "Topic: ..." lines are used to indicate the start of new discussion topics or agenda items, such as: <dbooth> Topic: Review of Amy's report[End of scribe.perl diagnostic output]