What is the current state of digital repositories for research output in the European Union? What should be the next steps to stimulate an infrastructure for digital repositories at a European level? To address these key questions, an inventory study into the digital repositories for research output in the European Union was carried out as part of the EU-financed DRIVER Project [1]. In this article the main results of the inventory study [2] are presented and used to formulate a European agenda for the further establishment of an infrastructure for digital repositories for research output.
The DRIVER inventory study aimed to provide a complete inventory of the current state of digital repositories in the 27 countries of the European Union. It is a follow-up of an earlier SURF study carried out in 2005, which included 10 European countries [3]. The study was started in June 2006 and completed in February 2007. By a combination of a Web survey, publishing results on a wiki and telephone interviews, an attempt was made to make the inventory as complete as possible and to generate feedback amongst participants in the study. In total, 114 respondents with digital repositories participated in the Web survey. This study focused on OAI-compliant repositories containing research output.
In this section the most important results about repositories themselves are presented from the following perspectives: coverage, contents, access forms for the full-text records, work processes, software packages and accessibility by search engines, gateways or portals.
The inventory focused on institutes who had already a digital repository implemented. From various sources addresses of potential digital repositories were collected and these were invited to participate in the Web survey. A combination of a wiki, an additional Web survey and telephone interviews were used to find additional information. From the data collection process in this study it was estimated that there are approximately 230 institutes with one or more digital repositories with research output in the European Union (of which nearly 50% participated in this study). The situation per country differs:
Figure 1: Contents of the Digital Repositories
What are the contents of those digital repositories? Based on figures given by 104 repositories, it appears that average digital repositories contained nearly 9000 records (8984, as assessed in the second half of 2006). The large majority of these records (90%) related to textual research materials: these records can be divided into metadata-only records (61%) and full-text records (29%).
5% of the records relate to non-textual materials such as images, video, music and primary datasets. The 5% 'other materials' relate to learning materials, students papers etc.
Figure 2: Types of Textual Materials
What types of textual research materials are deposited? More than half of the textual materials relate to journal articles (53%), a smaller number relates to books or book chapters (18%). Theses, proceedings and working papers - often labelled as grey literature - represent some 30%.
Figure 3: How Full Text Materials Are Made Available
What forms of access for full-text records are offered by the repositories? Is Open Access the only form of access, or are other variants offered? The most important other variants are Open Access with embargo for a certain time period, Campus Access or not publicly accessible at all (archival purposes only). The results are presented in the bar diagram. Clearly, most repositories (95%) offer Open Access accessibility. Open Access with an embargo period for full-text records is only offered by 18% of the repositories. About a quarter of the repositories (26%) offer Campus Access or contain records with no access (14%). Other forms of access are offered by 8% of the repositories, such as available for a fee, in response to an e-mail request or restricted to members of a project team.
| Which statement best describes the work processes of depositing of materials in the repository? | N | % |
| Self-depositing by academics, quality control by specialised staff members | 32 | 28.1 |
| Delivery by academics, depositing by specialised staff members | 30 | 26.3 |
| Collected by staff members independent of the academics | 8 | 7.0 |
| A combination of the above | 32 | 28.1 |
| Other | 12 | 10.5 |
| 114 Answers | 100.0 |
How is the material deposited in a digital repository? The results of a question about the work processes of depositing the materials in the repository are presented in the table above. It appears that a procedure of self-depositing by the academics, with quality control by specialized staff members is most common (28%), closely followed by a procedure of delivery by the academics, and depositing by the specialized staff members (26%). Only 7% of the repositories followed a procedure, whereby the materials were collected by staff members independent of the academics. However, 28% of the digital repositories of the participating institutes followed a combination of the above-mentioned procedures.
Which software package is used for the digital repository? The results of a question about this are presented in the table. The main results are:
In total 17 different software packages have been mentioned, while 19 respondents reported a locally developed software package. This means that the digital repositories in the European Union use at least 18 and probably more than 30 different software packages.
| Which software package is used for the digital repository? | n | % |
| GNU Eprints | 27 | 23,7 |
| DSpace | 23 | 20,2 |
| Locally developed software packages | 19 | 16,7 |
| OPUS | 12 | 10,5 |
| DIVA | 6 | 5,3 |
| ARNO | 4 | 3,5 |
| Fedora | 3 | 2,6 |
| CDSWare | 1 | 0,9 |
| iTOR | 1 | 0,9 |
| EDT | 0 | 0,0 |
| other | 18 | 15,8 |
| 114 Answers | 100,0 |
| The contents of your digital repository are searchable via which of the following general engines/gateways/portals: | n | % |
| General internet search engines such as Google, Yahoo, MSN etc. | 74 | 64.9 |
| OAIster | 66 | 57.9 |
| Google Scholar | 59 | 51.8 |
| Open DOAR | 47 | 41.2 |
| OAI Search | 39 | 34.2 |
| Scirus | 21 | 18.4 |
| BASE | 20 | 17.5 |
| OPUS | 14 | 12.3 |
| OASE | 8 | 7.0 |
| MetaGer | 7 | 6.1 |
| MEIND | 6 | 5.3 |
| Citeseer: Computer Science | 5 | 4.4 |
| PLEIADI | 5 | 4.4 |
| Other | 21 | 18.4 |
| 114 Answers |
Via which channels is the digital repository searchable/accessible? The results of this question are presented in the table above. Over 50% of the participating digital repositories are searchable via general Internet search engines such as Google, Yahoo or MSN, via OAIster and via Google Scholar. All other search engines or portals access less than 50% of the participating digital repositories. It has to be emphasised that these findings reflect the answers of the respondents to the questionnaire and not actual searches using the search engines/gateways/ portals mentioned. Therefore the results might reflect only the awareness of respondents about the searchability of their repositories. However, if their awareness is accurate, there appears to be no single search engine, portal or gateway that can access all participating digital repositories.
After these factual questions on their repositories, the respondents were asked to give their opinion on a number of issues. Firstly, they were asked which services should have priority for further development at a European scale. The top four answers (selected by more than 25% of the respondents) were:
In two questions, the respondents were asked to select the three most important stimuli for the development of their digital repository and the three most important inhibitors out of 14 factors. In the table below all stimuli and inhibitors selected by more than 25% of the respondents are listed. The following factors are seen as most important [7]:
| What do you see as the most important stimuli for the development of the digital repository and its contents in your institute? | N | % |
| increased visibility and citations of the publications of the academics in our institute | 53 | 46.5 |
| our simple and user-friendly depositing process | 50 | 43.9 |
| awareness-raising efforts among the academics in our institute | 33 | 28.9 |
| interest from the decision-makers within our institute | 30 | 26.3 |
| What do you see as the most important inhibitors of the development of the digital repository and its contents in your institute? | N | % |
| lack of an institutional policy of mandatory deposit | 57 | 50.0 |
| situation with regard to copyright of (future) published materials and the knowledge about this among academics in our institute | 56 | 49.1 |
| lack of requirements of research funding organisations in our country about depositing research output in Open Access repositories | 31 | 27.2 |
What is the current state of digital repositories for research output in the European Union? From this inventory study it is clear that digital research repositories are already well established throughout many countries in the European Union. In 2006 approximately 230 institutes had one or more digital repositories for research output implemented. In addition, from the contacts with respondents in various EU countries it appears there is a growing and active interest in implementing digital repositories at other institutes. Recent surveys in the USA show similar results [8]. Clearly, digital repositories for research output are on their way to becoming a permanent part of the scholarly communication and documentation infrastructure.
What should be the next steps in driving forward a connective and integrative infrastructure for digital repositories at a European level? The further deployment and development of the digital repositories will follow a two-tier approach:
With regard to such a two-tier approach, an agenda for activity at the European level can be formulated. Based on the above results of the inventory study, such an agenda should include the following seven action points:
The author thanks Bill Hubbard (SHERPA), Leo Waaijers and Kasja Weenink (SURFfoundation) for their stimulating comments and suggestions.
Ariadne is published every three months by UKOLN. UKOLN is funded by MLA the Museums, Libraries and Archives Council, the Joint Information Systems Committee (JISC) of the Higher Education Funding Councils, as well as by project funding from the JISC and the European Union. UKOLN also receives support from the University of Bath where it is based. Material referred to on this page is copyright Ariadne (University of Bath) and original authors.
Issue 52 July 2007