Viewpoint
Giving birth to next generation repositories

https://doi.org/10.1016/j.ijinfomgt.2007.02.004Get rights and content

Abstract

At a time when the ground work for Institutional Repositories has been done, and the focus in the field has shifted to higher things, we consider potentially the most significant of issues. Interoperability is a word that means many things, depending on who you ask, and most definitions are far from definitive. This article shows why Interoperability, in its many forms, is important, how it characterises the coming generation of repositories, in what ways it might be used, and even a little on how it is done. It also dares to speculate briefly on the future, and the incredible things we might achieve by working together. But it also reminds us that we do these things to improve the visibility and dissemination of scholarly work, and that this should always guide us.

Introduction

It is inspiring to look at the change in the scholarly content repositories throughout their academically recorded history. They were initially proposed as early as 1990 during Stevan Harnad's seminal “Scholarly Skywriting and the prepublications continuum of scientific enquiry” article (Harnad, 1990). At that point the web technologies that we are used to now were not nearly so well widespread, embedded or understood. It is not surprising, then, to see that early conceptualisations of the repository were centred around the idea of an FTP site on which scholarly materials could be deposited and retrieved.

It is also worthy of note that the earliest form of repository proposed was discipline specific, and the idea of an Institutional Repository (IR) was not articulated in the scholarly press until 1994 by Nobel Prize winner Joshua Lederberg (see Okerson & O’Donnel (1995) for a record of the discussion and Jones (2006) for an overview of the history). The Los Alamos arXiv1 filled the first repository niche, starting in 1991, and continues today to be an inspiration, as well as a heavily used and highly regarded repository. IRs, meanwhile, have been much later in their fruition (starting to become popular around 2000), but are now enjoying something of a heyday.

In the very first sentence of this article, we referred to all these repositories as “scholarly content repositories”, because “repository” alone is insufficient to describe what we are talking about, and the range of repository types evident today are not just Institutional or Disciplinary, but range through Learning Objects and Datasets to name just a few. But there was another motivation, and that was to suggest that these individually labelled repositories will cease to become quite so meaningful, and instead we will see a transition into a set of contextual representations of scholarly content. Bear with me.

Section snippets

Characterisation of the next generation

Originally this piece would have been called “Giving Birth to the Second Generation of Repositories”, but it has become rapidly apparent that repositories are well beyond their second generation, and that they have just come to the fore in recent years. We will not speculate, therefore, on how many independent generations you might be able to identify, only that we are now at another critical juncture where the focus of effort has shifted, and the rate of change has increased.

The “next

Some specifics on things to come

In the previous section, we suggested the sorts of things that can be achieved on a sufficiently short timescale in this field. In this section we look briefly at some of the technologies that might actually be applied to achieve these ends.

We can expect to see more of standards such as METS13 and MPEG21 DIDL14 as descriptive standards for complex objects, perhaps being moved around using OAI-PMH. We

Beyond the next generation

One advance that we hope to see become a reality in the not-too-distant future is the idea that we can link all our data together in meaningful ways. This can start with just considering your archived research paper with the pre-print, the post-print, the PDF and the TeX versions all connected together such that you always know which version you are looking at and where you can look for others. But it can end with a heterogeneous network of units of scholarly work which range from journal

Conclusions

Obviously it is not really possible to neatly divide development efforts in this area into generations in the same way that large technology companies do with their products. But the sudden swell in work towards integration, federation and devolution in repository architecture and communication cannot be ignored. It will achieve things that have yet to be thought of, but will also address our more mundane needs, such as reducing the impact of the mediated deposit paradigm on library

References (5)

  • S. Harnad

    Scholarly skywriting and the prepublication continuum of scientific enquiry

    Psychological Science

    (1990)
  • Jones, R., (2006). Institutional repositories. In K. Garnes., A. Landøy, A Repanovici (Eds.), Aspects of the Digital...
There are more references available in the full text version of this article.

Cited by (13)

View all citing articles on Scopus
View full text