ADMIRAL 20100916 project partners meeting
From ImageWeb
Contents |
ADMIRAL project partners meeting: 16-Sep-2010
- When: 16-Sep-2010, 11:00
- Where: Room D31, Zoology Department, University of Oxford (http://www.zoo.ox.ac.uk/info/contact.htm)
- Present: SR, AR, MW, SP, DMS, GK
Lunch will be provided after the meeting.
Please report to the Zoology Reception on arrival, and ask to be directed to Room D31 (up two floors in lift at back of building, then left out of lift, right down central corridor, last door on left).
This meeting page: ADMIRAL 20100916 project partners meeting
Previous meeting page: ADMIRAL 20100715 project partners meeting
Attendees
- Expected to attend: David Shotton (PI, IBRG), Graham Klyne (Project manager, IBRG), Sally Rumsey (PI, OULS). Anusha Ranganathan (OULS).
- Other partners and members of the Advisory Board: Simon Hodson, Paul Jeffreys, Alex Kacelnik, Peter Murray-Rust, John Wilbanks and Max Wilkinson are notified for information and welcome to attend if they wish to and are able, but are not expected to attend.
Goals
(David is on vacation as I prepare this announcement, so details may be changed early next week.)
- the main focus of this meeting will be to report on progress
- towards deploying a second instance of the ADMIRAL data store, and
- on our interactions with the Library Services' data repository
- to review progress in other aspects of the project
Agenda
- Introduction, setting the scene, agenda review (DMS).
- Review recent progress:
- Recruitment: replacement for DG (GK, DMS)
- Progress towards stable specification for RDFDatabank (AR/SR)
- Progress toward ADMIRAL submission and display of datasets in the RDFDatabank Entity store (GK)
- Further deployment of filestore to research users (GK)
- Update on interactions with test user groups.
- Other developments: Shuffl/JRON (GK)
- Technical and/or other detailed discussions (all).
- A.O.B.
- Wrap up, confirm actions (DMS).
Minutes
Review recent progress
Recruitment
Took on temporary 50% developer in August, who arrived, did some work, then resigned citing contract problems and management style. Intending to review and document lessons learned.
We discussed the possibilities for project extension: with another developer from mid-Oct, maybe an extra month falls within budget. Further details TBD when an appointment has been made and more data is available. Have held tentative discussions with programme manager, but nothing definite agreed until we are in a position to make a formal request.
We have 6 applicants for the advertised post, and are due to have shortlisting meeting later today.
Update on interactions with test user groups
Progress has been put on hold, other than informal discussions. DMS had been hoping to push ahead with the new developer on return from vacation, but resignation means this won't happen now.
Further deployment of filestore to research users
In preparation, have fixed some problems in system build scripts, and improved the build documentation. We are planning to create a deployment for the Evolutionary Development group, but specific work on this has not yet started.
Submission and display of datasets in the RDFDatabank Entity store
- submission testing: awaiting fix for metadata merging to generate sameAs for supplied URI
- have created an initial dataset display utility (demo later)
- have noted that retrieval performance from RDFDatabank is rather slow
ACTION: GK/AR: review test suite and update to reflect intended usage
Disciussion: Progress towards stable specification for RDFDatabank
The specification is understood to be stablilized, but documentation is not yet available.
ACTION: SR ask Neil for write-up of databank V1.0 specification.
Discussion: DOI assignment and metadata
Looking forward: discussion of DOI assignment (DataCite minimum metadata). SR and NJ are are preparing a response.
Informally: minimal requirements have 2 levels
- level 1: title, creator/author, pub date, publisher, discipline, DOI
- level 2: more, not discussed
Discussion of impact on citations ontology work. What/who is the publisher? What is the role of ORA? As publisher? Or what?
"Publishing" is a difficult term, especially if formal legal definition is considered. Nomenclature will always be a problem. Pragmatically, will collect what we want. At this stage, not requiring metadata to be submitted. We call Dryad, ORA, etc "data publishers" - because they encourage taking responsibility.
Publishers are those BL provide DOIs to, and will be required to take responsibility for the datasets. BL guarrantees DOI persistence/uniqueness, but the publisher must take responsibiity for being clear about the status of the dataset (which may or may not persist - essentially, persistence of metadata and status?).
What can be published as a DOI-citable dataset? "Stuff that is refused by crossref"?
ACTION: MW to email all about DOI issuing API
ACTION: AR schedule activity to create a very simple test suite to run against the BL API for DOI issue and maintenance
David asks about DOI stability when issued pre-publication. Can a DOI to be included in document dataset? What does a DOI resolve to? The landing page or the dataset itself? What about multiple items within a dataset. (DOIs used in workflows vs DOIs for human access.?) It was noted that standard Web content negotiation might be used to support resolving to either landing page or the actual dataset.
Asking about timescales: when will Datacite fix the core elements? Consultation closes on 20th Sept. Datacite metadata WG meeting several weeks later.
ACTION: MW to provide timescales to determination of minimum metadata for DOIs
ACTION: DMS: respond to DOI metadata consultation by 2010-09-17
ACTION: MW progress access for OULD access to DOI test server
ACTION: GK/AR circulate notes about implementation of content negotiation (non urgent, important)
Other developments: Shuffl/JRON
GK describes the role of JRON for migrating JSON web applications to use RDF/linked data. Implementation of JRON/RDF mapping for Shuffl has been created and tested, but is not yet integrated.
Technical and/or other detailed discussions
Dataset viewer demo given later, in lab.
SOme technical discussions. GK/AR have shceduled a pair-working session to finalize the APIs to be supported and the corresponding submission test suite.
A.O.B.
DMS wants to discuss data containers; analogy of containerization for shipping. General discussion held later over lunch.
Review actions from last meeting
(Not covered in meeting)
GK: provide simple example of RDF data merging - added to slide set: link in meeting minutes.
PJ: coordinate meeting with SUDAMIH/EIDCSR for "show'n'tell"
DMS: bibliographic RDF metadata example for Paul
Confirm new actions
Summary of actions:
ACTION: GK/AR: review test suite and update to reflect intended usage
ACTION: SR: ask NJ for write-up of Databank V1.0 specification.
ACTION: MW: to email all about DOI issuing API
ACTION: AR: schedule activity to create a very simple test suite to run against the BL API for DOI issue and maintenance
ACTION: MW: to provide timescales to determination of minimum metadata for DOIs
ACTION: DMS: respond to DOI metadata consultation by 2010-09-17
ACTION: MW: progress access for OULD access to DOI test server
ACTION: GK/AR: circulate notes about implementation of content negotiation (non urgent, important)
Next meeting
Per schedule, this will be 21-Oct-2010
Notes
- Notes from technical meeting with Library Services: ADMIRAL_20100825_technical_meeting_with_library_services
- Live project plan: ADMIRAL planning
- ADMIRAL project announcements blog: http://admiral-announce.blogspot.com/
- Previous project partners meeting: ADMIRAL 20100715 project partners meeting
- Next project partners meeting: ADMIRAL 20101021 project partners meeting
- https://datacite.org.uk/
