Tuesday, February 13, 2007
http://wiki.esipfed.org/index.php/Semantic_Web (for call in details).
Rahul Ramachandran, Sunil Movva, Kathy Fontaine, Howard Burrows, Robert Raskin, Peter Fox
Outreach activities (to other clusters)
Cluster activities - ESIP use cases and interactions with TIWG - http://wiki.esipfed.org/index.php/Application_UseCase_Template
We discussed details of the use case template based on a comparison table developed by Howard which contrasted the Application Use Case template (left column) from the Technology Use Case template (right column), the latter was developed for EIE based on a template from Michael Burnett. The Application template was built on a modified W3 format with modifications adopted in the VSTO project. Howard Burrows posted some notes: see http://wiki.esipfed.org/index.php/Use_Case_Process_Comments which identified the similarities and differences between the two formats. We proceeded to work through the table, led by Howard's discussion.
The agreed intent was to determine if the application template covered the appropriate material, in the correct order, and potentially to merge the application and technical use case formats.
First item - we do need to change the name of the page from TIWG_ We also noted that the application use cases will be domain specific whereas the technical ones are likely to span domains.
Several examples of completed templates would be required.
Under 1.1 Purpose - general agreement that this was an important part of the use case (and not present in the technical use case format) there was a suggestion for the person filling in the Purpose to make links to the FEA diagram (see ) as a reference model covering performance and business reference (currently these are covered in the non- functional requirements section). http://wiki.esipfed.org/index.php/Use_Case_Discussion
Actors: A longer description is required to indicate and distinguish between primary and secondary actors.
Functional and non-functional requirements - a description is needed to define the terms and the differences and also to relate these to the Purpose (see above).
Technology choices (which are filled in by the technology providers not the application use case developers): there was not a consensus on whether this section was appropriate in the document. The two sides were: full specification for a particular version of the use case, and not linking a specific technical implementation to the application need to encourage alternate and evolving technical solutions. The decision will will partly depend on whether the use cases are versioned.
Use of section 1.7?
We agreed that a section to define what defines success for the use case (beyond successful outcomes) and then elaborate - in essence, the definition of a metric and a way to quantify it.
Peter will make some updates and volunteered to provide some examples.
We then moved on to other topics.
Demonstrations for ESIP summer meeting
- status (see wiki for current submissions)
Rahul - NOESIS
Planning for breakout workshop/ plenary/ demos (what extra time do we need for cluster activities).
We discussed various needs: business meeting, plenary and demonstration time, and work group time as well as the need to meet with some of the other cluster activities, especially outreach to the application clusters.
Peter to prepare a priority list and participate in the meeting planning telecons.
Main work items
Data type ontology - review of existing efforts
http://www.w3.org/TR/owl-ref/#Datatype MMI work on conversion of GCMD datatypes and SWEET current representation.
Service ontology - review of existing implementations, detail of what is represented and what is needed Check this link for recent docs: http://www.ai.sri.com/daml/services/ owl-s/1.2/
We agreed to write a short outline/white paper to define the terms (of reference), needs, current work. Rahul, Rob, Peter, Liping (and Luis) agreed to work on this.
We discussed how to broaden participation - groups to invite: MMI, LDEO/IRI, SCOOP, Penn State, others (please send these names). Howard suggested Mark Gahegan (Penn State).