Making Engage a reality: core system integration

We’ve reported previously on useful meetings with our colleagues in HR and today I met once more with Ann Hastings, HR’s ever-helpful Operations Manager. Today’s meeting was rather more practical in nature than previous discussions, with a focus on getting data Engage has identified as being necessary to support research themes into an actual core system.

Engage research theme information would sit under Additional Options in the HR system’s Self Service portal. We must be careful about the screen shots we share from the HR system, for obvious reasons, but we hope to be able to show the research theme data in place soon.

As a result of our discussion, data from the Engage prototype is now being incorporated into HR’s test system. These data include a list of already-identified keywords (which may be augmented as required), top-level research classifications (based on those from RCUK) and research methodologies. Once these data are in place in the HR system, they can be used to drive a number of new fields on our staff records. In turn, this will allow staff to view and edit the data we hold about their research via our user friendly and already-familiar Self Service HR portal.

This initial foray into adding Engage-derived research theme data to the HR system turns on its head, somewhat, the approach we initially expected to take. The HR system is, naturally, most concerned with people. So, to begin with at least, we are supplementing the existing people records with the data (keywords, research classifications) that implicitly define a research theme in Engage. Later, however, we may still want to store the themes themselves in the HR system as originally envisaged. Ann is currently investigating approaches to this, including the possibility that themes might form a non-compulsory part of the HR system’s hierarchy structure for staff records. Themes would have to be non-compulsory, because not all of the 10,000+ people in the HR system will be associated with research endeavours, but being part of the internal hierarchy would mean that research themes may be integrated into every relevant staff record. Still, this is not completely straightforward solution, because, in Engage terms, research staff may be associated with more than one theme and the HR system currently permits only one selection at each level in its hierarchy.

So, an interesting challenge, but just getting the keywords and other theme descriptors into the HR system and available to try out via the HR test system’s Self Service portal is pretty exciting. It also provides the ‘hooks’ necessary to attach our Engage prototype to a core system and see – for ‘real’ – how users might ultimately interact with a fully-fledged Engage system.

Getting to know ePrints

We are at an interesting stage with the Engage project. We have a prototype which, while still being worked upon in terms of front-end functionality, is sufficiently far long to inform our upcoming recommendations on managing the University’s research themes. We have, we think, got the measure of the various core systems with which any production version of the Engage prototype would have to interact. We’ve also got a handle on potential stakeholders’ expectations (an area of discussion which may warrant a blog post in itself, at a later date) and know where Engage would sit in the larger scheme of the institution.

One of the systems with which we’re working most closely is the ePrints-based institutional repository, Enlighten. For the Engage prototype, the University’s ePrints guru, William Nixon, has already extracted a publications dataset that represents our cohort of Arts-based testers. However, what we might ultimately like to do is leverage the ePrints API to automate our Enlighten queries and to pull out an even greater quantity and variety of data. To this end, and with one eye on the upcoming JISC-funded Encapsulate project, we are currently setting up a CentOS-based test server for ePrints 3.3 and its plug-in friendly ePrints Bazaar. The test server will feature a subset of our live publications data, obviously tied to our test users, and allow us to delve into the latest version of the API with impunity.

Also of relevance to Encapsulate, and how it ties up with Engage, is our previous discussion about choosing a research classification scheme. Much of the conversation around Engage has been concerned with how we characterise researchers and research activity in a way that facilitates useful comparison. This theme will run through into Encapsulate. The upcoming project seeks to match business or private sector interests with academic expertise, meaning that some nomenclature that may be used to describe both is required. In reality, it seems likely that this will involve some mapping of a recognised, applicable academic structure – such as the RCUK classification Engage has adopted – against an equivalent set of descriptors to which businesses can relate.

Regardless of how the Encapsulate team proceed, working directly with the ePrints API will be a valuable and interesting endeavour for Engage. From a personal perspective, too, it is unusually satisfying for one project to follow on intellectually from the last. It’s almost as if someone has a plan.

 

Engage Update

What are we working on?

• Finalising the prototype to make available to others for testing and comment
• We have received notification that our poster abstract has been accepted for the ARMA conference http://www.arma.ac.uk/news/conferences/index.xhtml
• Although a final report is not required we decided we wanted one as there were some areas we wanted to make sure we collated and disseminated information.  The report will be made available open access via our Enlighten repository. 

Engage Dissemination

We have been disseminating Engage information and handouts at our Research Staff Conference.

See our sister project blog for details:

http://www.wordpress.com/2012/04/04/research-staff-conference/

SYNERGY WITH OTHER JISC FUNDED PROGRAMMES

On Friday I attended the ‘JISC BCE* OPEN INNOVATION AND ACCESS TO RESOURCES PROGRAMME MEETING’ along with Phin Wenlock from the lead partner Essex, on behalf of our latest JISC funded project ‘Encapsulate’

http://academicexpertise.wordpress.com/

I found it very exciting to again spot opportunities for further integration of services and re-use of information.

The challenges of this *’Business and Community Interaction’ programme included tackling ‘the language barrier’ (research outputs need to be intelligible to the stakeholder)  and ‘Re-use’ of information for different needs – both issues that we are keen to understand and offer solutions to in Engage.

There was talk about the need to not just throw information at the world but to target to business and community requirements – exactly what Engage and Encapsulate are doing.

Encapsulate will take forward many of the ideas started under Engage further developing the methods of re-using exising information.

There were 11 projects represented at the meeting and there is a lot of synergy and we are eager to talk with these projects and re-use ideas.

It may be that a number of these projects will be interested in the Engage demo and outputs and we will make sure we share our findings with them.

 

Some Interesting Points from Our Team Discussion Today

Matt our technical guru has been trying out presentation of research method on the prototype basing the initial list on research methodology classifications used by ESRC.

I have asked Dale Heenan, Web Project Manager, ESRC
what/how this is used at Research Council’s.   I will also discuss with IRIOS2 colleagues (including representatives of two Research Council’s)  if this is something that might be possible and useful for Research Organisations’s to get from RCUK.

We are still pondering the issue of how to categorise research inputs and outputs by theme – rather than a list of themes at person level that may not all be relevant for a specific paper or award. Matt has developed some algorithms that nicely suggest appropriate categorisation. We could default these rather than leave the outputs unattributed to themes. Further discussion on-going.

We talked about impact and it’s definition and agreed we would refer back to MICE http://mice.cerch.kcl.ac.uk/ findings and could also keep an eye out on the new DESCRIBE project at Exeter which is looking at definitions for impact. This does not complete till well after Engage but it will still be of general interest to the University of Glasgow.

I have been discussing issues of impact definition with library. REF have one definition and collect the information at quite a high level. RCUK collect at award level. We need to establish an efficient method of collecting and re-using data where possible within the bounds of Data Protection, Freedom of Information, and other requirements.

Choosing a research classification scheme

For the Engage prototype, we looked at a number of means of controlling the categorisation of research activity before finally settling on a classification scheme used by the UK Research Councils. The idea was that we should attempt to impose some formal, broadly recognisable structure on our research themes, to facilitate discussion and to provide a cross-disciplinary vocabulary that would allow us to make meaningful comparisons. As discussed previously, this formal structure will be supplemented by allowing for much less rigidly-controlled keywords, drawn from researcher publications and assigned by researchers themselves.

The rationale for choosing the Research Councils’ classification scheme related to that fact that it was:

- Cross-disciplinary, meaning the Arts were as well represented as the Sciences, etc.

- Hierarchical, potentially allowing us to make more intuitive associations between related research

- Tied to funding, meaning that it was difficult to argue with its relevance

In addition, it should be noted that if we had discovered an alternative classification scheme already in widespread use across projects or institutions, this would likely have become our front-runner. However, our search for a de facto standard classification – which also satisfied the above requirements – did not uncover any obvious contenders. Meanwhile, the RCUK scheme has already proven popular with our sample stakeholders, not least because it can be mapped directly to our funders’ own classification (it is already employed on the Je-S system used to apply for funding from the AHRC, for example).

Some of the other classification schemes we considered include:

-          The University’s own College/School structure

-          Eurostat’s Classifications metadata

-          The EU’s Nomenclature for the analysis and Comparison of Scientific Programmes and Budgets (NABS) classification

-           The Universal Decimal Classification Summary (udcS)

Of these options, the udcS (based on the larger UDC Master Reference File) is perhaps most closely aligned to our needs, and has the advantage of being licenced for use under a Creative Commons Attribution Share Alike licence. It is also truly international in nature which, while also one of its strengths, meant that it is not so closely relatable to the familiar nomenclature of UK higher education institutions or, crucially, that of our funders. For our purposes, we found Eurostat’s impressive classifications metadata somewhat unwieldy, being focussed on economic activity rather than research per se. Similarly, the NABS classification deals primarily with scientific endeavours, which might have marginalised the Arts and Humanities in our otherwise highly inter-disciplinary project. Finally, while our own institution’s structure provided a familiar (albeit recently re-organised…) framework into which we all must necessarily fit, it lacked the granularity offered by the RCUK scheme, which delves much deeper than college, school or even subject-level classification can.

So, with the Research Councils’ classification scheme now up-and-running in the Engage prototype system, we are pleased to offer our relational database version of the scheme for download, in MySQL format:

RCUK research classifications

We hope that other institutions may find this useful, although we are unable to offer much in the way of support for its use! The data themselves are quite self-explanatory. There are three levels of classification available, and each classification has a field (‘parent’) that links back to the classification above and another field (‘level’) that tells you where that classification sits in the hierarchy (where 1 is the top level, and 3 is the lowest).

Of course, not everyone might agree with our decision: while we are more than happy to recommend the RCUK classification scheme, we also welcome comments from any interested parties (go ahead and hit the ‘Leave a comment’ link if you like…).