Members
- Dragan Ivanovic, VIVO Technical Lead, will direct the efforts
- José Ortiz, Software Developer, from Ecuador
- Michel Héon, Enterprise Architect, University du Quebec a Montreal
- Abhishek Raval and Narendra Kumar from DSquare Technologies will also be participating
Goal
Integration of DSpace and VIVO platforms
- Streamline process for academics
- Avoid duplicated bibliographic data management
- Adding semantic web aspect to existing DSpace repositories
- Adding depositing files and monitoring of Open-access policy compliance to existing VIVO instances. VIVO instance might be used as a front-end for researchers where they can be motivated to create their own profile page including a list of research results (publications, datasets, etc.). For any single research results researchers might provide DOI/URL in the metadata and/or initiate depositing files to DSpace through VIVO user interface, while the rest of the files processing will take place in Dspace by librarian/officers.
- Growing community for both platforms
- Knowledge transfer between selected team members, improving their capacity through collaboration, and engaging new developers for the VIVO and DSpace community.
Phases
Phase 1
Functional requirements
- VIVO doesn’t store full text articles and other research outputs (e.g. dataset) - it transfers them to a digital repository
- Adding a file (or files) to VIVO entities (publications, datasets, etc.) through VIVO UI
- Use DSpace REST API to deposit file items and update metadata in DSpace
- URL to file deposited in DSpace is preserved in the VIVO and visible to VIVO users
- Crosswalks - definition of the mapping between VIVO ontology and DSpace internal model
- XML settings files used to specify how data fields are mapped between the two systems
- Default crosswalks supplied for standard fields
- Mappings for both directions
- Inbound DSpace -> VIVO
- Outbound VIVO -> DSpace
- Migration batch
- Harvest all DSpace items and ingesting in VIVO with clear database
Non-functional requirements
- Communication between VIVO and DSpace should be based on CSRF Tokens [7]
- All new features should be covered with
- Logging
- Tests
- Wiki documentation
Phase 2
Functional requirements
- Support to copyright checking in DSpace
- VIVO deposits items to the DSpace workflow
- VIVO checks and shows status of an item (e.g. First deposit, Published), when DSpace item is published, the URL is visible to VIVO users
- Enable adjusting crosswalks to take advantage of evolving systems
- Crosswalks capabilities
- String manipulation (split, concatenate, format, etc.)
- Boolean algebra
- Map multivalued compound data types
- Dictionary lookups
- Registry lookups (based on identifiers such as ORCID)
- Regex
- Support multiple DSpace repository connections in VIVO
- e.g. separate repositories for datasets & publications, or
- in different departments
- selection of a DSpace repository in which metadata and a file (or files) should be deposited
- Migration batch
- Harvest all DSpace items and matching to existing publications in VIVO
- Monitoring for changes in DSpace
- VIVO preserves in its database the date of last DSpace harvest
- daily or weekly update might be configured
- OAI-PMH ListIdentifiers request with defined from parameter to collect identifiers [5], and to use them for collecting item information through DSpace REST endpoint [6]
Non-functional requirements
- All new features should be covered with
- Logging
- Tests
- Wiki documentation
- Create a dockerfile for VIVO-DSpace custom build
Calls
- Weekly, on Thursday at 7am Eastern Time
- The zoom link - https://lyrasis.zoom.us/j/82670709536?pwd=MzF3NDladE1DKzEvUml4SGQ5eUFsQT09
- Minutes available at Calls