This week I have been attending the fantastic Delivering RDM Services Workshop organized by GEANT and others. Individual sessions have covered procurement, open source business models, and partnerships involving National Research and Educational Networks. In my view serious discussion around problems with delivery of RDM Services is long overdue, and this workshop is a welcome and groundbreaking step that hopefully will be the start of an ongoing conversation leading to fundamental improvements.
It’s been an exciting and stimulating week, and there is far too much of interest to attempt to summarize in a short post. The wrap up session is tonight, and in a somewhat unusual, and perhaps even cheeky, effort to make sure that a point made in the presentation I gave in the Procurement session is not overlooked, I am going to elaborate briefly on it here, prior to the wrap up session.
IT’s central role
The point I highlighted was that the disconnect between RDM managers and IT managers lies at the heart of most of the problems with service delivery and that until that issue is faced head on it’s not going to be possible to make a lot of progress on enhancing RDM Service Delivery. There was not sufficient time in a lightning talk to explain why I said this, so here is some support for my contention that IT is central to RDM Service Delivery at every stage of the process.
Exploration, design and requirements gathering, evaluation and procurement
IT very likely will have a say, often a deciding say, in budgetary decisions about RDM services. Their support will be needed to determine how the service will fit into existing infrastructure — here a critical consideration is cloud versus on premises deployment and related issues of cost, data location, and convenient ongoing maintenance. IT also will be involved in the process of refining and presenting requirements and evaluating prospective vendors.
Service delivery does not end with procurement! Procurement is just a step in an ongoing process. After a decision has been made to procure a particular service from a particular vendor, before the service can be delivered IT needs to work with the vendor to get it deployed. Without IT’s active and engaged support in working through the many issues that will come up — e.g. cloud versus on prem, which may have been left undecided to this point, and integration with institutional infrastructure and SSO — deployment and hence delivery cannot take place.
A note on the role of University IT in National Research Educational Networks
Here at Research Space we are highly enthusiastic supporters of National Research and Educational Networks (NRENS). For example we’ve worked with NRENS in Switzerland, the Netherlands and Norway on deployment of RSpace in a way that will make it accessible to researchers in multiple educational institutions in those countries in conjunction with other research services enabled by the NRENS. And we are looking forward to the advent of the European Open Science Cloud, which will among other things open up exciting new avenues for multi-organizational collaboration. In the US we are exploring deployment of RSpace on the New England Research Cloud, an initiative of Harvard, MIT, Boston University, Northeastern and the University of Massachusetts which has a mission and structure roughly analogous to EOSC. I’ve written previously on the possibilities that a national research data service could open up.
But here again university IT departments are a necessary part of the equation and without their active involvement efforts to deploy RDM services in conjunction with NRENS are stymied. For example, at the request of researchers at a university recently we have been attempting to activate an integration between RSpace and a national file sharing service. To do this we need the support of the university’s IT department, which so far has not been forthcoming. Fair enough, they have other priorities and limited time. But the unfortunate end result, so far at least, is that a workflow the researchers view as useful/essential, and in principle is exactly the kind of workflow the national file sharing service was set up to facilitate, is not available to them.
Concluding thought: let’s bring IT managers into the conversation!
Notwithstanding the somewhat pessimistic tone so far, it clearly is the case that the divide between research data managers, who conceive and deliver research data services, and IT managers, who deploy and support them, has narrowed considerably over the past five years or so, and that the influence of research data managers in budgetary and policy decisions relating to research data services continues to grow. That’s progress! In our experience, moreover, many institutions are moving towards a more collaborative approach and framework — formal and/or informal — that results in research data managers and IT managers working together more effectively.
That is encouraging, and if it were not the case the prospects for making progress on enhancing delivery of RDM services would not be good. It is the case, however, and that is all the more reason to include IT managers in the conversation this inspiring workshop has taken to a new level. My simple plea, therefore, is that in followups to this workshop — and I hope there will be many of them — IT managers are invited to the table to join the discussion as the equal partners they need to be.