Category Archives: resource discovery

Developing complexity and service response

Following from the release of a major upgrade to RoMEO during Open Access week, the Centre for Research Communications, which runs the SHERPA services RoMEO, JULIET and OpenDOAR, has now launched two User Surveys to gather feedback from the community – a survey for RoMEO and a survey for OpenDOAR and ROAR. These surveys are to help prepare for support of an increasingly diverse research communications environment.

As part of RoMEO we have always had a suggestion form for new publishers or for updating information and an active community of contributors and suggestions. However, we wanted to launch the current survey to more formally gather comment, opinion and wishes for the future development of RoMEO as the circumstance of its use changes over time.

Development of publisher contracts

The service originally developed to interpret publishers’ copyright transfer agreements for author self-archiving and we want this to continue as the core of RoMEO. The system started with a single aim and could interpret, summarise and present information from this single viewpoint. As time has passed the situation for archiving has grown more complex – and users’ needs have matched this. The growth of “hybrid” options for journals has made a single interpretation of a journal’s copyright contract impossible. Individual funders have come to agreements with some publishers for Open Access publishing and therefore (sometimes but not always) Open Access archiving rights also apply to publication of work they have funded. Sometimes individual publishers have recognised and matched the requirements of some (but often not all) funding agencies and for them allow their standard terms to be modified.

Complexity for Authors

All of this gives a far more complex environment for authors to work in and underlines the need for assistance in guiding authors through their options and responsibilities. It also presents real challenges to RoMEO in providing this. If any service is to be used successfully by end-users, then it has to reflect the users’ needs and fit into their workflow. If one of the current drivers for archiving work is compliance with funders’ mandates, then these need to be represented and permissions summarised.

However, many mandates have a focus on OA publication, rather than archiving. Given the number of funding agencies and the complexity of their requirements (summarised in and linked to JULIET) as these apply to every publisher, the original fairly clear RoMEO interface became quite crowded. The upgrade from last week has attempted to deal with this, in allowing “single funder views” of the data, as it were, but the diversity of possible approaches to the data remains, We are aware that archiving in an institutional repository practically takes a place within a suite of options that needs to be presented with clarity and simplicity.

This is a reflection of a larger picture – how will this look in future? What is being developed in practice within institutions to deal with the requirements of funders, authors and publishers? From a strategic point of view, what can services like RoMEO give in support of wider access to information?


We have also released a survey for OpenDOAR and ROAR. These services, run by the CRC and University of Southampton respectively, share some aspects of work in analysing the world’s repositories, but exist as separate services with individual aims. ROAR has a focus on quantitative and statistical analysis of repositories and their holdings; OpenDOAR has a focus on qualitative analysis and policy and standards development. Each of the services has healthy feedback from its users, but again, we wanted to more formally gather comments from the community on the services as they will be used in a more diverse picture of repositories.

Development of Repository Environment – Full-text holdings

Here too the situation has become more complex over the years that they have been in operation. While the original aim and distinctive difference for open access repositories was that anyone could access the full text, for many repositories this has been bypassed by conflicting needs so that for some the great majority of their content is merely metadata. Many, and probably most, repositories accept metadata entries, maybe driven by concerns to display high levels of records irrespective of full-text links; or because the repository is used for internal purposes that require no more than full-text; or because there is the hope that at some point in the future, there will be enough staff resource to chase down the full-text.

Whatever the reason, the decision to accept metadata is a significant one. It means that many searches of open access repositories now end in a bibliographic entry with no access to the full-text article, or simply a link to it held on the publisher’s website. For the researcher looking for material, this effectively undercuts the rationale for searching repositories in the first place. It is hard for any advocate to engage researchers with open access as a distinctive and different service when the full-text content is not there.

Having said that, some drivers for the adoption of institutional repositories now seem sufficiently strong (at least, to some institutions) to match the original idea of full-text access. The use of the repository as an enhanced research publications database is one example: others include it as an administrative system for projected REF needs; other requirements may be met by full-text access on-campus, even if restricted off-campus.

While the continued growth of metadata-only records remains a significant challenge for advocates and the future use of the repository network, here too, developments take their place within a wider and more complex environment of different use, structure and purpose of repositories.

Development of Repository Environment – Open Access?

Again, one of the original distinctions was that the repositories should be openly accessible. The fact that many repositories are set up as closed access in some way (registration & password systems, even subscriptions or pay-per-view) but identify themselves as open access was one of the drivers in the establishment of OpenDOAR, with a policy of a human accessing each repository and sampling holdings to check that what was being claimed was true. Since the start of OpenDOAR we have rejected between 25 – 33% of candidate repositories because they are out of scope – no full text at all, not open, test sites, junk data etc.

The types of material held in repositories has grown to include research data, learning objects, varieties of grey literature, specialist collections, and others. Some of this content brings with it understandable restrictions on access while at the same time being appropriate for a repository-like collection and (partial) exposure.

Combined with the variety of purposes that repositories are accumulating, this means that repositories of different “flavours” now take their place in a more complex, interesting and ultimately more rewarding environment. While I believe that we cannot afford to loose sight of the key goal of access to full-text research, services offered by OpenDOAR and ROAR (among many others) have to change to reflect this and allow the diverse requirements of the users of repository content and the diverse basis for repositories to be reflected in the service they provide.

Future directions through complexity

The level to which this happens with these services is a reflection of the larger question. To what extent should we all continue to press for the original OA vision if this is at the expense of the easy growth of some alternatives (metadata repositories, partial access etc)? Should future development in the field be guided by what has proved popular and practical so far, if this fails to address the original goal of full-text open access and original method and goal of author-engagement and self-archiving? Do we set goals that are the natural extension of what we see developing, or aim for the more robust and clear vision that was articulated in Budapest and elsewhere?

Have your say in how some of the support services in this developing environment will themselves develop. Do contribute to the RoMEO survey and the OpenDOAR and ROAR survey. We will be interested to see your thoughts.

Bill Hubbard

Students’ use of research content

Students use research outputs; papers, books and so on. However, students are – like many other actual and potential users of research outputs – not familiar with the landscape of academic research, and the ways in which one can discover resources that are useful and relevant. (It is worth noting that the recent PRC report on SMEs access to research output cites sources suggesting that even professionals working in hi-tech SMEs have the same trouble.) A recent JISC report by a team at UCLAN led by Stuart Hampton-Reeves sheds some light on how students discover and access research outputs. The report tells us a little more about the “Google Generation”, noting that “most students will go to their library catalogue first, then Google” (and not social networking sites) to discover research, and that they do not generally have a sophisticated understanding of peer review. The report adds to a growing body of evidence, including that from UCL’s Ciber group and an ongoing user observational study, on how we can improve the “discover-ability” and accessibility of research outputs. It certainly seems that there are ways in which both tutors and libraries might better help students act as “apprentice researchers”, navigating the unfamiliar research landscape more effectively. Perhaps more fundamentally, infrastructure – including repositories – is making it possible to add more research outputs to the open web, which is where these users are.