#

Activity

Cost

Timescale

Complexity

Who

Notes

1

Undertake a significant round of workflow/usecase modelling, building on significant work already undertaken, and combined with iterative prototyping, to identify “good enough” solutions for a range of stakeholders.

6 person months

In parallel with other activities; total of 24 months elapsed time

Distributed activity across different domains, single place to share results

·         Any whose funders wish to support this

·         incidental activity in existing projects/services

·         dependent on people with the right Agile dev/reqts skills to do this well

2

Review and update existing relevant mindmaps provided as input to the workshop (and keep up to date for life of project as a resource)

1 person month

Short

Single place to share/maintain results (in coordination with other activities)

Alma Swan (funded through Sherpa)

May need a way of pivoting/refactoring existing mindmaps to align with this program of work

3

Repository identifiers: build on things like DOAR and ROAR to produce a registry of available repositories. This would need to accommodate both open and closed access repositories, the ability to refer to things not in the registry, some sanity-checking on entries, temporal scope, and automated pings to detect ‘repository death’

12 person months

12 months

·         Probably mix of centralised and federated solution

·         Hub and spoke model?

Europe TBD

·         Build in an incremental way

·         Need benefits from inclusion

·         Draw on existing lists from other domains

4

Organisation identifiers: by geographical region, need to deal with temporal scope changes

12 person months

12 months

·         Probably federated solution

·         Hub and spoke model?

Everyone who cares

·         Many to many relationship between organisations and repositories

·         Base on existing lists (incl. DNS)

·         Governance model required

5

People identifiers: develop a people identifier collection service (both human and machine queryable) to enable people to create equivalence/non-equivalence assertions between a subset of their different digital identities, and to store these equivalences

36 person months for a prototype, plus user testing and marketing

18 months

Could be federated or centralised

·         SURF?

·         And anyone else who cares

·         Builds on existing author identifier systems

·         May be benefits in parallel approaches as a risk-mitigation strategy

·         Support for optional semi-automatic assistance with identifying possible candidate identities

·         Add different personas in version 2.0

·         What about authority files?

6

Object identifiers – phase 1: recognising the existence of a wide range of existing object identifier schemes, provide a simple object equivalence service. In version 1, this would operate at the level of FRBR Manifestations, and enable the ability to say (for instance) “the object in repository A identified by this Handle is the same as the object in repository B identified by this ARK”. Service would be queryable by both machines and humans.

12 person months dev time, plus subsequent marketing

9 calendar months

Could be federated or centralised

·         Potential for JISC to take this forward

·         Also links with DRIVER and other national activities

·         Motivated primarily by citation tracking, enhanced publications,  usecases

7

Object identifiers – Phase 2: Move to Expression level

?

·         Work or Expression version of this is more like people equivalence service

·         Build on RIDIR, and VALREC projects, as well as plagiarism detection

·         Other types of relationships could be picked up here