Note – this is a draft proposal, interest in its development is valued. Comments welcome or email jo.walsh@ed.ac.uk
During our time of Focus we will create Linked Data resources from different data services supporting scholarly communications, and connect them to other projects. We look to build up a picture of the data and publications metadata needed to practise Open Scholarship. We hope to find and present evidence of the value in following the approach to scholars and administrators.
- Develop several proof-of-concept prototype applications exposing and re-using Linked Data in EDINA services and nearby.
- Provide Guidance for publishing and re-using linked data for open scholarship
- Develop Linked Data skills and knowledge amongst developers and project officers at EDINA, the JISC National Datacentre at the University of Edinburgh
- Provide a focus for encouragement and support of Linked Data development across UK teaching and research institutions
- Make creative re-use of Linked Data resources currently supported by JISC
Background
Linked Data holds many promises – the potential to make web-based resources more discoverable, to facilitate their re-use in ways the original data providers may not have imagined, to create new insights and connections.
The Lucero project at the Open University is an early example of a JISC-supported effort to make information about research and teaching efforts available as Linked Data. The jiscEXPO strand of JISC projects has a wide scope, including bibliographic, multimedia, and geographic data.
Making data available according to the four Linked Data principles is just the beginning of the story. To realise the benefits, steps should be taken to
- Make the network of data denser – linking resources to another, re-using data and schemas.
- Make the network of people around the data denser – spreading knowledge about how to re-use and make sense of the resources available.
Proof-of-concept
The project timeline describes work with several EDINA services that provide scholarly communications data, and efforts to link it to open bibliographic and citation data published as part of recent jiscexpo project work.
Infokit
Produce a blog documenting the experience of developing Linked Data resources and of persuading the maintainers of different resources to support the effort. The blog, including discussion of others’ work in similar areas, will build towards an infokit describing use cases in making scholarly communications data available as linked open data and the potential value therein as well as the cost.
Awareness
Maintain awareness of work going on in the area of scholarly / institutional linked data and document related work and discussions on the blog; open the blog to guest posts from different institutions and projects.
Hold regular informal seminars for EDINA developers
Bi-weekly “Linked Data learn-in” where developers can compare experiences and try out new tools and services. Quarterly we’ll invite developers from elsewhere in the UK research support community, with emphasis on those finding their way around Linked Data applications and standards (sharing the journey rather than the destination.
Provide support to JISC services and to other related projects or services.
Where services are keen to provide Linked Data we will provide a point of contact, and developer help where possible. Consultancy efforts will build on the network outwith EDINA, using social media tools to link developers to others working on the same problems where possible, trying to work with the DevCSI developer support initiative.
Investigate the development of a data.ed.ac.uk resource
To practise what we preach, try to engage different information publishers at EDINA’s host institution, the University of Edinburgh, with the prospect of following the model that the Open University is setting in providing access to institutional resources as Linked Data, tracing the steps that lead to setting up of a data.ed.ac.uk resource.
We would create a case study in the organisational and social engineering effort needed to convince the Administration of the benefits of providing Linked Data resources. We will build on existing efforts to create networks of resources (the LAIRD project linking the ERA publications repository to the DataShare repository).
I think this has a lot of synergy with a proposal I failed to send into JISC last week which is outlined here:
http://www.geog.leeds.ac.uk/people/a.turner/personal/blog/archive/2010/11/#JISC2
(Sorry that blog is so Web 1.0. I need to sort it out!)
I wonder if the image connection to the following is more than just a shared library:
http://www.science3point0.com/evomri/2010/11/21/where-have-all-the-data-gone/
Thank you
Appreciate the links, Andy. I always liked your lo-fi lo-bandwidth blog, keep it that way and it will be neatly post-Web when the time comes.
AKSW in Leipzig have done solid work turning OpenStreetmap into a Linked Data service at http://linkedgeodata.org/
Sorry to hear you had a near-miss for 15/10 on a project with a dense local network looking firmly outwards from academia, partnerships out to the edges of the public sector, (“memory institutions”), probably something I’m missing here, food for thought.
Great to see so much interest on linked data in education going on lately.
Would be nice if all the different initiatives could somehow connect.
Mathieu (http://data.open.ac.uk)
Mathieu, right, this proposal is very directly inspired by what you’re doing with the Lucero project ( http://lucero-project.info/lb/ ) and http://data.open.ac.uk/
Questions:
- Do you plan to write about the more social engineering side of the project? That is, what has been the experience of convincing university administration to publish Linked Open Data about what it’s been up to – and also, what kind of outcomes/results would be helpful to ensure institutional support to do more?
- What actions do you think would help most in connecting existing initiatives / encouraging more to grow?
Good questions:
1. Yes! (we have to, it was in our proposal). More seriously, there is a lot to say around this, and we are only starting. Convincing data owners is actually not too hard, if you don’t ask them to do any work and the data is already “somehow” public. Getting higher-ups to understand the value of linked data, and to engage with it is another story.
2. There are quite a number of ways to do this:
– share experience, through workshops/seminars/etc.
– share code
– share ontologies, and more specifically, make sure we use the same bits of ontologies to represent the same things, so that data from different universities can be aggregated. Of course, that also means possibly discussing collectively the way to represent certain common and tricky things (e.g., there is an ongoing discussion led by Southampton on the representation of places on a university campus).
– share applications: I suspect that something that makes sense for one university also makes sense for the others.
re infokit (as well as related programmes/projects) the jiscMRD programme have several projects working with the D of ABCD open scholarship which I will be collecting names and details soon, so will pass those onto you.
[...] This blog describes the Linked Data Focus project being planned at EDINA and also serves as its project proposal. An overview of the work planned can be found on the page discussing our Aims, objectives and outcomes. [...]