Main Navigation

User login

Devices Ontology WG Mtg 2010.05.18

Agenda/minutes for Device Ontology Working Group meeting 2010.05.11

Note: 2010.05.11 telecon was postponed to 2010.05.18

iEtherpad- Google Docs

(http://ietherpad.com/devont20100511 -- not working well today; perhaps there're other clones more stable but...)

NOTE: We started using/evaluating Google Docs for minutes notes. This is the document for today: https://docs.google.com/document/edit?id=1jo1Go7x3rwdDOYr3uA_eRaWPInhZ-_UIms48FAJfJCY#

which is in the "MMI DevOnt" folder at Google Docs. This folder will faciliate the sharing for future documents. Carlos will be sending additional invites to share the folder as necessary.

Agenda

Logistics: Welcome, Minutes Review, Agenda Review
Misc
1. EtherPad to go away
2. ont2dot
MMI Device Ontology - consolidated documentation
Modeling specific instrument models
VaisalaWM30 instrument - Inputs/Outputs, MeasurementCapability
Instantiations
Other topics
End Logistics: Next meeting, Action items, Attendees

Minutes

Logistics: Welcome, iEtherpad, Minutes Review, Agenda Review

iEtherpad: see link above
Minutes review
Agenda approval

Misc

EtherPad to go away

We’ve been using ietherpad, one of the many clones. However, it hasn’t been working well today; so we decided to try out the new version of google docs today (http://docs.google.com/support/bin/answer.py?answer=176442&hl=en) which has real-time collaboration features including embedded chat... Unfortunately, seems like there is not track of who-wrote-what, but we will continue evaluating it.

New ont2dot tool

Graph representation of ontologies: New ont2dot module. See README. Example usage:
$ mvn exec:java -Dexec.args="http://mmisw.org/ont/mmi/device"
$ dot -Tpng device.dot > device.png

Note: very preliminary version.

Any intent to make it a Protege plugin? No for the moment, but if anybody is willing, please go ahead.

MMI Device Ontology - consolidated documentation

See http://marinemetadata.org/community/teams/ontdevices/devont

Entry point (for humans) for the documentation of the ontology including notes and observations and related information.

License?
CC Data License?
John commends this license because it is designed particularly for data, rather than source

Why is this data???
this is ‘data’ because it is not executable in and of itself, but instead is reasoned over and reused as an entity in its own right. (The difference between ‘data’ and ‘source’ is kind of like the difference between ‘data’ and ‘metadata’ -- an arbitrary distinction in some ways.)

Who are the ‘owners’ or contributors of this ontology?
There are 2 lists, the “everyone” list (anyone who has every provided input) and the “principals” list (criteria to be defined -- at least 6 months participation or at least 10 relations, say). There is also a 3rd list, the people currently involved, which may be a useful list but may not be worth the time to track it.

Who is giving the license?
MMI is not a legal entity, so it can’t really declare what license applies. It is a problem that many universities have their own legal restrictions on the products of their members.
For discussion purposes, assume Carlos is the author. Then who is the copyright owner depends on his terms of employment. Ditto if there are other authors.

Related aspect: Definitions (where and how to link to them)
Two main kinds of definitions: 1) Definitions for the concepts introduced in the DevOnt itself, eg., MeasurementCapability; and 2) Definitions needed for instantiations of the ontology and related concepts, eg., for measurement capabilities or inputs/outputs, need to refer to definitions of associated variables/parameters, eg., wind speed, temperature, etc. In the latter case, use rdfs:isDefinedBy.

About rdfs:seeAlso
“rdfs:comment, rdfs:seeAlso, rdfs:isDefinedBy and rdfs:label are included here because some constraints which apply to their use can be stated using rdfs:domain, rdfs:range and rdfs:subPropertyOf. Other than this, the formal semantics does not assign them any particular meanings” from http://www.w3.org/TR/rdf-mt/#RDFSINTERP This is what I thought and hoped. Italics mine.---Bob M.

Modeling specific instrument models

Modeling a certain instrument model via subclassing the Sensor (or Sampler) class and adding appropriate restrictions as in the VaisalaWM30 example.

Applying this to the Multibeam instruments (several models)
Hierarchy of models? no, at least initially
Luis to help Bob in creation of instances.
Bob will provide relevant materials to move forward

VaisalaWM30 - Inputs/Outputs, MeasurementCapability

See http://marinemetadata.org/community/teams/ontdevices/devontacc.

Nan: Please note that the VaisalaWM30 example may need to show how wind direction accuracy will depend on measured wind speed. That may be a much more complex example.

Not dicussed

Instantiations

Mechanics to create ontology instances. Is the issue about how to use standard tools, or about automating the generation of instances?

...

Discussed in part in the Multibeam topic above.

End Logistics: Next meeting, Action items, Attendees

Action Items

DevOnt tracker: http://code.google.com/p/devont/issues/list

[Carlos] provide MMI links about URIs
[Carlos] determine approach for capturing controlled definitions --> rdfs:isDefinedBy
[Bob A., Carlos] look into ProcessOutput vs. MeasurementCapability related issues, especially in the context of the Multibeam ontology.
[John and Reyna] Follow up on scale references