It is a very useful exercise to consider the overall ecology of systems in the university which store and communicate information about research projects, researchers, and research outputs. These could include:
- HR (for names of employed researchers, associates etc)
- Student information system (for research student information)
- The email/directory system as a source of identification and authentication information
- Research grants and awards (for projects, principal investigators, team members)
- Publications databases (central and/or faculty and departmental)
- Webpages for individual staff and departments
- Library systems for bibliographic data
In terms of an approach to this mapping, it is difficult to improve on a presentation by Bill Hubbard for the RSP in 2009:
- What related interfaces/systems/databases are there in your institution?
- What system tie-ups are desirable?
- What ways can the repository be used?
- What replication of information exists across systems?
- What about workflows for gathering and using material – and what systems/interfaces support these?
He suggested creating a model of interfaces, systems and databases and considering the relationships between them:
- Identify all of the related interfaces, systems and databases in your institution
- Place these on the model of interface/ system/ database
- Identify overlaps, synchronicities, gaps – which could be filled with innovative and useful systems
- Draw out workflows between each component
- Identify owners of each component and each workflow
- Think about relationships between systems
- Think about political aspects of control and authority
- Ingest once, store once, use many times
- Ensure authority for each point of information can be backtracked to an appropriate authority
Mapping where the same pieces of data are entered into different systems and databases on separate occasions offers the potential for quick wins in terms of reducing effort and improving consistency.
Similarly, mapping where metadata can be inherited from one system to another offers the chance to enhance information and provide added value by linking pieces of data with each other and with research outputs while reducing the ‘keystroke barrier’.
The issues surrounding bulk importing of records from internal and external systems and deposit via publication management systems such as Symplectic are discussed in maximising deposit through embedded workflows section, metadata creation and flows section and interaction with external systems section.