You are here: Foswiki>GEO_SIF Web>SifMeetings>SifMeeting-2013-08-23 (06 Sep 2013, DavidArctur)EditAttach

SIF Meeting, 23-August-2013

Convened: 13:30 - 15:00 UTC

Attending: Steve Browdy (SB), David Arctur (DA), Joan Maso (JM), Andy Mitchell (AM), Paul Eglitis (PE), Siri Jodha Khalsa (SJ), Bart De Lathouwer (BD)

Also referenced: Wim Hugo (WH), Doug Nebert (DN), Osamu Ochiai (OO), William Sonntag (WS), Barb Ryan (BR)

Recording: David Arctur (DA)

Agenda and Discussion

  • Standards and SAs for quality (to be delivered to DSWG DDQ)
    • SB: sent out email for SBA POCs' feedback on how they report QA (standards & special arrangements used; best practices for assigning quality indicators)
    • Sent to 30-40 folks, have received 4 responses: little response about their standards, mainly about quality control practices. Will post these on SIF twiki. Will report back to DDQ in few weeks.
    • SJ: RDA Metadata Standards Directory https://www.rd-alliance.org/filedepot_download/418/100 This report outlines a work plan by Research Data Alliance to document current metadata practices, and plans to improve communications and uptake of metadata standards in geosciences. At meeting where these were discussed, Erin Robinson suggested GEOSS SIR to be mentioned in context.
    • SB: Eye on Earth initiative created a "standards array" a couple years ago, for Global Network of Networks (GNON), which will make use of SIR.
    • Options for followup: (1) respond to RDA directly, suggesting using SIR; (2) send email to SIF to review the RDA doc first; (3) SIF folks joining the RDA interest group.
    • SJ: RDA is broader scope than SIF/SIR.
    • DA: let's combine (2) & (3) [ ACTION ] DA will forward doc to SIF, and recommend SIF members to join RDA interest group if interested. Go to https://rd-alliance.org/working-groups/metadata-standards-directory-masdir-working-group.html
    • JM: there's a glossary in this doc, which we should compare with SIF's glossary, [ ACTION ] JM will compare the glossaries and report back to SIF. http://www.dlib.indiana.edu/~jenlrile/metadatamap/seeingstandards_glossary_pamphlet.pdf
  • 2013 interoperability topics (results in whitepapers)
    • JM: GEOLabel update: idea has been around for about 5 years; graphical representation of data in simple way to help choose among different datasets. GeoViQua working on this, surveying users, now have a basic graphic concept: circle with 8 facets, with colors & coverage representing different metadata completeness. Changes based on user feedback comments. Have API and web service to generate GEOLabel based on specified inputs. www.geolabel.info
    • What can SIF do to help GeoViQua in general? (1) SIF promotion; (2) used in GEOSS
    • SB: (not taking sides) this idea of GEOLabel is supported by many folks, and also criticized by some folks (in terms of how label is derived and used). SIF should take a hard look at GEOLabel purpose, goals, how constructed, and write an opinion doc on interoperability implications, pros, cons of GEOLabel.
    • SJ: how does this have bearing on interoperability?
    • SB: would not have impact on technical interop, but in terms of semantic interop it could have some impact; also in terms of institutional interop.
    • SJ: just knowing it's automated, and concern with lack of completeness of metadata, something with high quality but poor completeness of metadata could be poorly represented.
    • JM: 5 of the 8 facets come from metadata user's point of view; 3 other facets come from existence of user feedback and publications generated by users, relying on user feedback.
    • SB: if GEOLabel included user feedback, could address some of SJ's concerns.
    • PE: (1) Can development of the GEO Label help identify the source of the data and support provider accreditation; (2) is the generation of a GEO label tracked?
      Providers having their data used in GEOSS being identified as the provider is important concern for data contributors. SO I feel this is important.
    • JM: (1) yes; (2) no, but interesting idea
    • SB: how does provenance
    • JM: if it's possible to find provenance in metadata, then you click on the facet for that info to see it
    • JM: started with conceptual idea, have evolved to a concrete concept, now should go back to see what this may be improving. If SIF could help with this review, would help with development. Could use tutorial as starting point to learn more about GEOLabel.
    • PE: SIF should avoid opinions but identify how the label can be used and also identify limitations and make suggestions how those limitations can be overcome.
    • DA: perhaps this could be brought into a structured discussion group like at OGC or ESIP?
    • AM: share others' interests and concerns expressed so far. For NASA, we're building a metadata rubric, where metadata providers can reveal how complete they feel their metadata is. Like the idea of expressing a graphic for this, but basis for the graphic presents hurdles (ie, if automated and based on total completeness measures)
    • SB: agrees with PE that we shouldn't express opinions. [ ACTION ] SB will create a page on the SIF twiki for this issue, and ask for members' comments, and send to GeoViQua or other groups.
    • Legal interoperability:
      • SB: Legal Interop subgroup of DQ subgroup is working on a white paper; need to wait for that before making comments.
    • Discuss work and timeline for resource casting
      • SB: Review draft paper regarding registration alternatives; this is paper about future registration, but we should also consider current working registration approaches. Three ideas: (1) use 'resource casting', a way of registering resources without manual input to CSR; (2) batch processing input from data provider; (3) interop between broker and CSR, or remove CSR.
      • Resource casting is most innovative. SB working on draft doc about this, for review by Sept 20 meeting on this topic.
      • DA: should data providers now be registering through both DAB and CSR?
      • SB: yes
      • SJ: implications for GEOSS Data CORE; many resources being brokered but not registered in CSR. GCMD for example.
      • SB: GCMD is registered in CSR. All resources should be registered in CSR, either directly or via a catalog registered in CSR (eg, GCMD).
      • SJ: people are voting with their feet, and just working through DAB.
      • SB: resource casting could help address this, so that providers don't have to manually register with CSR, and could also fetch info DAB needs. [ ACTION ] SB will create SIF twiki page for further discussion on this.
    • Discuss work and timeline for linked data
      • SB: were going to defer this till after resource casting, but would like to discuss sooner. LD is important for Eye on Earth, and in other situations.
      • DA: not sure what SIF needs to do on this; should follow rather than try to lead. RDF, object identifier schemes, etc..?
      • SB: could identify pros & cons, and how could affect GEOSS. [ ACTION ] SB will start a SIF twiki page on this, ask for comments.
  • SIF Request for Assistance (deferred to next meeting)
    • Discuss inclusion on GEOSec site vs. GEO Web Portal
  • GEOSS Tutorials (deferred)
    • Review those that need attention
    • Decide on work schedule
  • Any other business
    • SEINet vs GEOSS for content search (deferred)
    • EarthCube? update: crawling to find XML docs that describe catalogs and other resources that could be harvested by broker, is part of an EarthCube Building Blocks project to be awarded
    • Eye on Earth update (UNEP-Live)

Next meeting date: 6 September 2013 Main focus will be to review input to Sept 20 working meeting.

-- DavidArctur - 23 Aug 2013
Topic revision: r3 - 06 Sep 2013, DavidArctur
 

This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback