Notes for September 26th meeting

[FranckCo]

Agenda for September 26th meeting

Update on issues

• Issue Link between StatisticalProgram and StatisticalProgramCycle #88: pull request made with additional text, reviewed and merged (see ca120aa for details)

• Issue Clarify text regarding data and metadata #89: still some discussion on the text, see pull request

• Issue Use case: Statistical compendium #84 assigned to Florian: write a "COOS use case" document

• Issue Comments on datasets in relation to DCAT #81 assigned to Florian and tagged for version 2. Issue Link between datasets with same content but different structures #97 created for one of the questions, subject should be clarified for the other issue to create ("link between statistical data set and product based on GSIM revision")

• Issue Use Case: link information resources (methodologies) and business processes #78

A first version of the ReferenceDocument class is provided in commit c9ded05. Note the following:
. The singular is used instead of the proposed ReferenceDocuments. Each document is a specific resource, so it is preferable to link the documents individually.
. coos:ReferenceDocument is defined as a subclass of foaf:Document
. the Spanish translations are missing

A first version of the informs property is provided in commit 1a49994. Note the following:
. The domain of the property is coos:ReferenceDocument and its range is coos:Activity
. Multiple occurrences of the property can be used to link multiple instances of coos:ReferenceDocument (see above)
. The Spanish translations are missing

Pull request created and submitted to Flavio and Zoltán

Comment by Flavio: How about "To provide supporting information for performing an activity"? -> OK, commit added to the PR

Propose to add text at the end of 3.3: an example would be useful

Expecting Zoltan's review

• Issue Promotion of COOS #76: do we submit for NTTS? Who?

• Issue Feedback from Rosemay McGrath #72: assigned to Christine and tagged for version 2

• Issue Feedback from Jay Greenfield on the COOS main document #69: assigned to Franck (to set up dedicated meeting) and tagged for version 2

• Issue coos:StatisticalOrganization and regional statistical offices #65: assigned to Edgardo

• Issue Classify GAMSO activities #63: assigned to InKyung

• Issue Inclusion of Statistical Support Program #60: Flavio contributed, to discuss

• Issue Statistical or not Statistical #59: assigned to Franck, waiting for other issues

• Issue Model links between GSBPM and GSIM #19: to discuss

• Issue Mapping to GSIM Business Group #4 closed

Ontology expert feedback

• Documenting the source standard for the different items (ref): still to do.

• OWL equivalence between class and ConceptScheme members for product content and product presentation (ref.): pull request created, reviewed and merged, see commit 1fad04f for details.

• Application profile (ref.): issue to create at tag for version 2.

• Activity/Activity category (ref.): work in progress by Florian, greater impact than initially thought

Public review

We might have to postpone by a week (see previous point)

AOB

[InKyungChoi]

Regarding public review: I went through email list used for GSIM revision and GSBPM revision, not sure if the profile is right. Any suggestions (on top of those already made: contact@sdmx.org, twg@sdmx.org and swg@sdmx.org, gonzalezmorales@un.org) would be appreciated!

[FlavioRizzolo]

On the DDI community, we might want to include Arofan Gregory (ilg21@yahoo.com), chair of the DDI CDI working group, and who is also with CODATA leading the development of the cross-domain interoperability framework (CDIF). Other people unrelated to the official stats world but that might have something interesting to say are Wendy Thomas (wlt@umn.edu), chair of the DDI Technical Committee, and Jon Johnson (spuddybike@gmail.com), who is a member of the DDI Technical Committee, metadata practitioner and organizer of EDDI, the European DDI Conference. Wendy and Jon could provide a different take on COOS, I think.

[InKyungChoi]

List so far

Expert reviewer (first round) + recommendations
• Jay Greenfield, nightcleaner@gmail.com
• Matjaz, m.jug@cbs.nl
• Robert McLellan, robert.mclellan@canada.ca
• ? Monica Scannapieco, scannapi@istat.it
• Rosemary McGrath, rosemary.mcgrath@stats.govt.nz
• contact@sdmx.org, twg@sdmx.org and swg@sdmx.org
• gonzalezmorales@un.org
• Arofan Gregory, ilg21@yahoo.com
• Wendy Thomas, wlt@umn.edu
• Jon Johnson, spuddybike@gmail.com

SSG members
• Zoltán Vereczkei - Chair Hungary
• Anna Dlugosz Poland
• Cory Chobanik Canada
• David Barraclough OECD
• Essi Kaukonen Finland
• Flavio Rizzolo Canada / DDI
• Florian Vucko France
• Dan Gillman United States
• Edgardo Greising ILO
• Juan Muñoz Mexico
• Kevin McCormack Ireland
• Franck Cotton France
• Martina Hahn Eurostat
• Mauro Bruno Italy
• Omurbek Ibraev Kyrgyzstan
• Waleed Mohamed Egypt
• Matjaz Jug Netherlands
• Carlo Vaccari Italy
• José de Jesús Luján Salazar Mexico
• Juan Eduardo Rioja Romo Mexico
• Manuel Cuellar Rio Mexico
• Andrea Petres Hungary
• Janusz Dygaszewicz Poland
• Federico Segui Uruguay
• Daniel Blanc Uruguay
• Helda Mitre Albania

[InKyungChoi]

Meeting notes (September 26th)

Participants: Franck Cotton, Jenny Linnerud, Flavio Rizzolo, Michele Riccio, Zoltan Vereczkei, Juan Munoz, Edgardo Greising, Florian Vucko, Anna Długosz, Paolo Francescangeli, Dan Gillman and InKyung Choi

Issue #81 Comments on datasets in relation to DCAT

• Regarding issue I: although underlying data values might be the same, the contents in different formats (e.g., wide format vs. long format) are not necessarily same semantically. It is not just that we move content from data to metadata (or vice versa). The difference is also semantic not just structure between different formats. For example, qualifying attribute in the wide format applies to the whole entry (record), while in the long format, we can say at more granular level. Action: to document somewhere in issue about these differences and subtleties.
• Regarding issue II: In GSIM, Product "consumes" Data Set. In COOS, we say coos:dataSet is a sub-class of coos:product. We need to make sure if there is no inconsistency. Action: to open an issue with clarification

Issue #78 Use Case: link information resources (methodologies) and business process: Action: @Vereczkei Zoltán to provide examples of reference document

Issue #76 Promotion of COOS: ISI session is approved. Idea to organise a session during NTTS - to be further discussed in the Supporting Standards Group meeting

Issue #60 Inclusion of Statistical Support Program

• Flavio made a proposal where we have "production phase" and "overaching activity" instead of GSBPM phases.
• It would be better to separate "change phase" (GSBPM 1-3) and other GSBPM overarching processes in “overarching activity”. Action: @Inkyung Choi to make a comment on github issue

Issue #65 coos:StatisticalOrganization and regional statistical offices

• It is important to widen the scope to include those in national statistical system, not just national statistical organisations
• Edgardo made a proposal Action: to continue discussion on github issue

Issue #63 Classify GAMSO activities: Action: @Gillman, Daniel - BLS to make a comment on the issue about "Manage Statistical Methodology" being statistical activity

Ontology expert review

• Action: @Cotton Franck to open an issue about application profile and tag for version 2
• The impact of differentiating activity and activity category is bigger than expected. Would need to postpone the public review by one week.