Are Repositories Now Enterprise?
Prologue: “Web Wonk, pushing toy technologies, tells us SOA is dead. Doesn’t he know the Gartner Hype Curve? It’s in the “Trough of Disillusionment”!” -‘Enterprisey‘ on Twitter-
Over the past six months there has been a recent set of activities taking place in the repositories community that have made me wonder if repositories are finally coming of age, so much so that they could be considered “Enterprise”?!
The trends I have noticed that I think have enabled us to reach this “Enterprise Enlightenment”, includes:
- Mandates for deposit of Open Access content by both MIT and UCL 
- Development of two new repositories that have been coupled together via OpenSource tools: one at Oxford, (EntityStore by Ben O’Steen) and the second at the University of Rocheste,r New York (IR+ by Nate Sarr and co).
- Cloud Provision by the DuraSpace Foundation.
- The establishment of ePrints as a core business system for Universities to manage the upcoming REF.
- Growing number of Research Management Systems, such as Symplectic, Mendeley, Zotero, MePrints, BibApp and other systems that are intended to sit atop repositories to leverage the individual researcher’s awareness of related research on the Web.
But what is “Enterprise” or rather how does a University view repositories in terms of being a core business process that fits in with their overall distributed technology stack? The definitions on the Web about what “Enterprise Architecture” is, compounds the fact that “Enterprise” is mostly just marketing hype, somewhat like “Web 2.0 and The Cloud” but worse as it doesn’t have any consensus amongst those who use it the most (recently asked an IBM project manager to define it for me via what technologies it used and the best they could say was “robust transaction” (please note: SOA is dead and not even good for corporations who have the budget for it, and is especially not relevant now or in the near future for Small/Medium Institutions!!!).
However, despite my general disregard for Enterprise “Architectural Astronauting” terms and the management speak that justifies it, there is still a real need by medium/large organisations (HEIs <15K students/staff) to understand what core systems and applications that their various distributed departments are using (emphasis on “distributed department” selection of systems over central procurement): Central IT Dept. (email/IM, CMS), Marketting/Business Dept. (CRM), Accounts Dept. (expenses, finance), Student Support (VLE), Library Dept. (LMS, OPAC), various Subject Specific Depts (MatLab, Office), and most significantly the rise of Web 2.0 applications that students, lecturers and researchers can micromanage themselves to change as their immediate end user needs arise.
But what about the quality scholarly content that sits in all of the above systems (on the Web, in the Central IT server and on the local departmental computer)? Basic agreement on how two departments can pass each other their data and understand it, has mostly been limited to passing spreadsheets (CSV) over email and calling each other up to explain the meanings of each column header (I’m not criticising this process entirely BTW). The core question being: how we can make this process streamlined and easily repeatable? It should be easy for two individuals from different departments to meet over coffee and have the data they were telling each other about in the hands of the other in a *meaningful* and *repeatable* way in the same amount of time it took to have that single cup or coffee. <–! THIS IS WHAT AN ENTERPRISE ARCHITECTURE SHOULD BE TRYING TO ACHIEVE (NOT SOME OVERALL BLUEPRINT OF THE WORLD THAT MAKES US FEEL GOOD BECAUSE IT FITS ON A SINGLE POWERPOINT SLIDE). If we are going to actually start understanding this thing called the Web and how our users use it, we need to stop being so deterministic that our systems are going to slot together like some big puzzle (despite what Microsoft clip-art suggests).
So where do repositories fit in my view of what a University Enterprise Stack should be trying to achieve? Well first off systems have to be simple enough to describe in terms of “push” and “pop” on a stack. This means having a core business process and defined meaning that everyone can understand. I now believe that repositories have found their core business proposition via the REF and making sure Universities list research outputs to obtain research ratings (we have not succeeeded in making the business case that IRs should be doing the job of archiving, a core library platform, or the job of an institutional demonstrator/poster space). Hence, Repositories fit in the “University Enterprise Stack” by their virtue of being a system that delivers a business solution to a real financial problem.
So if we are all in agreement that the core business case of repositories are to collect all the research that needs to be submitted for the REF (or for any other assessment body that needs to get an overview/rating of the research being produced by a department), then I think we have reached a place where repositories can be placed within the context of other institutional systems, i.e. Repositories -as described by this business case and management speak- are Enterprise.
We can now start to have other services built atop Repositories in the layer cake that is becoming the University Enterprise “Stack”, including the ability to swap out repositories systems within the stack if Institutionally required, e.g. for cheaper storage costs via DuraSpace or further customisation via EntityStore. The immediate layer that is emerging above the repository layer is “Research Management Technologies”, i.e. Symplectic, Mendeley, Zotero, BibApp, etc. Accordingly, if Repositories are about listing of research content that is worthy of assessment, then “Research Management Systems” are there to make sure that worthwhile research can compete against similar research as well as get noticed amidst competing research. Notice here the business case of lecturers/departments getting higher research ratings (repository layer), and then being able to get more research funding because they know what, where and how they can leverage their research for more funding. The latter is the emerging business case that Research Management Systems are making so they too can become a push/pop layer on the University Enterprise Stack.
Of course, this brings in the question of where the innovation space is and how repository platforms might seperate the various tools in their systems to leverage a new layer in the university Stack, e.g. be able to “sell” a new system that couples with the repository (please keep in mind I am not advocating transaction of money but rather a well defined system that can be “sold” alongside other similar products be that Open Source system of Closed Source systems, e.g. ePrints has the functionality emerging of a Research Management System but have not separated it out as a layer that sits atop the repository, hence it is really two products in one). Further questions in my mind include:
A side note: This post is partially trying to categorise the emerging layer of “Research Management Systems” (RMSs) that are emerging over the top of repositories (this includes everything from the modern Web 2.0 Systems for Researchers like Zotero and Mendeley to the ‘Common Research Integration Systems’ (CRISs) that originally populated this space . I am not suggesting all of these systems have the same functionality, but rather that these systems are competing to fulfil the same business case need.
- What does the core business case for ‘research management systems’ look like, how do you sell this new systems to University higher ups?
- What other systems lower in the stack could feed into the repository to help leverage further data that would enable a higher research rating, e.g. could the University website CMS provide further insight to either the repository or rather more likely the “Research Management System” layers in the stack?
- What systems could be coupled with both the ‘research management layer’ and the ‘repository layer’ that could result in further business cases for the institution?
- What systems sit atop ‘research management systems’ that will leverage their core business case to enable further business models.
- What should innovators be exploring to fully leverage the business case of research management, e.g. is it individual tools that inform researchers where their ideas could/should compete or should ‘research mangement’ be departmental effort to do team research in specific research fields for further funding?