Teleconference - OGC Quarterly meeting, Washington DC, USA March 2014

Agenda

• Review of work with defining requirements for the conceptual model for GeoSciML v4
  • see XMI package in GeoSciML Subversion at branches/3.2.0/abstractSpecification/Requirementsv3.2.xml
• Discuss bundling of requirements classes to define conformance classes (along the lines of portrayal, basic geo map, and extensions)
  • see GeoSciML v4 trunk uml for proposals
• Discuss implementation spec.
  • the conceptual conformance classes might be implemented in different ways, e.g. simple feature for portrayal, complex features for more content-rich classes.
  • use of abstract classes to decouple namespaces (pattern being used in ISO19115-3)
• Review the future use of workarounds in GeoSciML for:
  • encoding of borehole intervals in WFS/GML
  • FES inability to query number pairs in SWECommon QuantityRange elements
• JSON implementation?

Minutes

• 5(?) attendees in Washington
  • Steve Richard (AzGS USA)
  • Ron Lake
  • Steve, please fill in the other Washington attendees here
• 6 attendees via remote teleconference
  • Ollie Raymond (GA Australia)
  • Eric Boisvert (GSC Canada)
  • Tim Duffy (BGS UK)
  • Mark Rattenbury (GNS NZ)
  • Aaron Sedgmen (GA Australia)
  • Marcus Sen (BGS UK)
• Steve Richard presented a Powerpoint on his progress in documenting requirements (use cases) for GeoSciML v4
  • see Powerpoint presentation here: http://external.opengeospatial.org/twiki_public/pub/GeoSciMLswg/TeleconferenceArlingtonMarch2014/GeoSciMLv4Discussion.ppt
  • using tool in EA to generate MS Word documentation from UML
• Steve mentioned the investigating JSON implementation of GeoSciML
  • Ron Lake commented that there may be a more generic XML -> JSON mapping, rather than specific GeoSciML -> JSON. Needs further investigation
• Discussed methods for documentation of GeoSciML v4 requirements
  • documentation only of v4, not v3.2
  • O&M and SWE specs as examples
  • ACTION: Steve to move any v4 requirements models/documents out of the v3.2 branch and into the v4 trunk in the subversion repository
• Discussed options for modularization of the GeoSciML v4 implementation specification
  • requirements needed at 3 levels - portrayal, basic query, complex query
  • needs to be based on use cases of data delivery and ease-of-implementation, not science discipline use cases like the v3.2 schemas were
  • need to review INSPIRE requirements to inform the definition of basic level of GeoSciML implementation requirements
    • perhaps model like the new ISO19115-3 core & extended pattern, where the core is likely to be very small with implementations of only very basic things like data types (eg, strings) and controlled concepts
    • potentially GeologicTime, Boreholes and LaboratoryAnalysis-Specimen schemas could still stand alone as app schemas. But still need to consider their dependencies on “core” GeoSciML and whether those dependencies mean that there is little benefit in separating them from a single broad “!GeoSciML Extended” app schema.
• Need to review the relation of GeoSciML Boreholes schema to other boreholes standards like WITSML, RESQML, etc
  • DIGGSML geotechnical data model – imports aspects of WITSML in parts
    • eg, http://diggsml.org/2.0a/schemas/Kernel.xsd
    • having funding support problems?
• Need to review potential use of SensorML in, or instead of, LaboratoryAnalysis package
  • note that the use of soft typing in SensorML would require developing vocabularies to support SensorML soft typed elements, and would need to document best practice requirements for use of specific vocabularies.
    • contrary to current “hard typed” elements in GeoSciML
    • Geoscience Australia will be looking at potential implementation of SensorML and will feed back findings to GeoSciML SWG
• Meeting closed after 1 hour.
• Issues not discussed that require future work:
  • GeoSciML Users email list - do we need a separate email list from the current GeoSciML list which has, until now, been used for development of the GeoSciML model
  • No discussion was possible on WFS and FES SWG developments
  • No discussion was possible in the time available about requirements details (ie, what GeoSciML elements need to be in “basic” and “extended”). Follow-up teleconference (at least 2 hours) needed before Tucson face-to-face to progress this - Ollie to organise.

Teleconference details

https://www4.gotomeeting.com/join/329697191

• United States: +1 (646) 982-0002
• Australia: +61 2 9087 3701
• Austria: +43 (0) 7 2088 1399
• Belgium: +32 (0) 38 08 1855
• Canada: +1 (647) 497-9391
• Denmark: +45 (0) 69 91 88 61
• Finland: +358 (0) 942 41 5777
• France: +33 (0) 182 880 455
• Germany: +49 (0) 811 8899 6976
• Ireland: +353 (0) 14 845 975
• Italy: +39 0 699 36 98 80
• Netherlands: +31 (0) 708 912 514
• New Zealand: +64 (0) 9 909 7882
• Norway: +47 21 03 58 95
• Spain: +34 931 81 6668
• Sweden: +46 (0) 852 503 498
• Switzerland: +41 (0) 435 0167 07
• United Kingdom: +44 20 3535 0624

• GeoSciMLv4Discussion.ppt: GeoSciML v4 Modularization - OGC TC Washington DC, March 2014 presentation from SWG meeting