Page tree

Friday, 13 April 2016, 15:00-16:30 CEST

Connection details

Recording: MP4AdobeConnect viewer

Agenda

[15:00-15:10] Welcome, approval of the agenda and minutes of the previous meeting (for discussion and agreement) (Michael Lutz)

[15:10-15:45] Status registry federation testbed

  • Review of the changes in the RoR conformance classes
  • Feedback from participants
  • Next steps

[15:45-16:15] Demo of the RoR prototype

  • Demo of the "harvesting" use case
  • Demo of the "federated search" use case

[16:15-16:30] AOB

  • Collecting FAQs

Minutes

Attendees

Alejandra Sanchez, Jesus Barrera, Paloma Abad (ES), Zoé Petty (FR), Esa Tiainen (FI), Christian Ansorge (EEA), Lorena Hernandez, Daniele Francioli, Emanuela Epure, Michael Lutz (JRC)

Welcome, approval of the agenda and minutes of the previous meeting

Michael welcomes Zoé Petty who is a new member of the group. She will be replacing Etienne Taffoureau.

The minutes of the last meeting were approved without comments.

The agenda was approved as proposed.

Status registry federation testbed

Daniele explains the new points of the ROR requirements documentation 

  • Download URL
    • Michael asks what happens if you decide to use the content negotiation approach.
    • Daniele explains that the system will check first if the content is available through content negotiation otherwise, it will check that the download URL is working.
  • Added minimum frequency for harvesting (set to daily)
    • Any of the publications office frequency values will be available to set the harvesting frequency except the “twice a day” which will be not supported.

Daniele shows the new features implemented in the RoR prototype.

  • Login section
    • The online demo is considered an alpha version of the RoR that can be used by anyone interested by requesting an account as indicated in the help section of the application.
  • Private area
    • Adding the URL to a registry descriptor file
    • Launching the harvesting
    • Overview of the harvesting jobs
      • List of registers
        • Date last/next jobs
        • Success
    • Validation features / validation reports
    • Possibility to start manually the harvesting at a register level.
    • Available history of the different harvesting jobs that have been launched
      • What happens when the systems founds error on the descriptor
    • Removing a registry from the private area will remove it both from the list of registry/registers and the federated search.
  • Federated Search

Discussion and proposals for improvement:

  • [Proposal] Rename "object" to "Register" in the harvesting job table.
  • [Proposal] Michael proposes to indicate more explicitly in the results page thrown by the federated search whether if an item is a registry or register for example in brackets.
  • Question from Chris (EEA): What happens with what was already harvested if we delete the registry (is it set as retired?)
    • Daniele and Michael reply, that no, it simply removed as the function of the RoR is to share data not to version "local" data. The RoR must be seen as a broker to share data.
    • [Proposal] Michael suggests to change the wording  in the user interface
      • Instead of  "add registry description" or "delete descriptor" use "share" or "stop sharing", that should help clarifying the functions of the RoR
  • Question from Chris (EEA):  where can you see the status of an item?
    • Daniele replies that only the relations status are shown, and regarding the items only "valid" items are imported by the system
    • [Proposal] Michael indicates that it would be useful to include and index as well superseded and retired items.
  • Question from Chris (EEA):  is the INSPIRE registry content already fully available in RoR?
    • Michael replied that it is not available for the time being as the Re3gistry is not fully supporting the RDF yet as required for the RoR.
    • Only a subset is included in the testbed: those registers and code lists from the central INSPIRE registry that are being extended/profiled by the other registers in the testbed.
  • Questions on the uniqueness of the fields:
    •  What if you have two IocalID, which are equal?
      • As they will belong to different extensions, so there will be no error. Instead if you intend to add a URI which is already in the system you will get an error because as the URI is a unique field.
  • What happens if you are harvesting a correct descriptor file of the registry but the items are not in the right format?
    • Daniele replied that items must be correct initially otherwise an error will appear.

[Actions] for the next web meeting:

  • Testbed participants should update the descriptor files provided so far including the new requirements of the exchange format. They should then try to import them in the RoR and provide feedback on what works or does not work etc.
    • Daniele to open an issue in the MIG platform to include feedback on the RoR testing
  • All to continue adding FAQ questions on the issue space of the MIG platform
    • Christian will set up a wiki page with the questions to start discussing on them.

AOB

  • Michael thanks Chris for providing some FAQs (#2746)
  • Chris asks if there will be a FAQ section in the RoR system, Michael says that they will try to answer the questions first.
  • Zoe indicated that her IT team is having problems for installing the Re3gistry software and she would like to know who they could contact to solve them, as there are many platforms for that. The Re3gistry team prefer the use of the email. Besides, she suggests adding a compilation regarding problems with the Re3gistry installation to add them potentially to the FAQs as they will be probably asked more than once.
  • Doodle for the next meeting: http://doodle.com/poll/tx3x566fp7un3imh
  • No labels