Search   |   Back Issues   |   Author Index   |   Title Index   |   Contents

Conference Report


D-Lib Magazine
October 2004

Volume 10 Number 10

ISSN 1082-9873

Report on the 8th European Conference on Digital Libraries (ECDL 2004)

12 - 16 September 2004, Bath, United Kingdom


Jonas Holmström
Swedish School of Economics and Business Administration
Helsinki, Finland

Red Line


The ECDL 2004 was held in September at the University of Bath, UK, one mile from the center of the World Heritage Site of Bath. The conference offered a total of 47 papers [1]. The conference papers had an acceptance rate of 32%. It also included three keynote sessions and two panel sessions and a poster and demo session. (Reports on three of the four ECDL 2004 workshops are provided separately in this issue of D-Lib Magazine, and the fourth workshop report will appear in the November 2004 issue of D-Lib.) The report of the main conference presented here is by necessity limited to but a few of the presentations. I have chosen to highlight the presentations that I thought were distinctly different in some particular aspect. The selection was also made based on my own interests and reflections and therefore is admittedly subjective. The conference program as well as information about how to obtain the conference proceedings is available on the conference website [2].

The conference presentations gave an excellent overview of current digital library research projects. Digital libraries are created for an increasingly diverse and distributed user community. Digital libraries are also focusing more on the individual user, and different kinds of personalization features are emerging as key success factors. Related to this is the integration of digital libraries directly into e-research and e-science, which puts the digital library in a larger context. The importance digital libraries have in e-science and the 'Semantic Grid' was emphasised in the keynote by Professor Tony Hey, Director, UK e-Science Programme.

E-learning also received attention. As an increasing proportion of learning is mediated by the Internet, it becomes important to integrate the digital learning and digital library environments with each other. Director Neil McLean from IMS Australia addressed these integration issues in his keynote.

Lorcan Dempsey from OCLC (Online Library Computer Center) spoke about the rapidly changing library landscape and introduced some interesting new terminology. One of the most apparent changes libraries face is the change in expectations created by Google and Amazon—or 'Why can't the OPAC work as Google/Amazon?' This change in expectations was labelled the 'Amazoogle' effect. Dempsey also argued for the need for a theory or big picture without which libraries are vulnerable to—among other things—'Marchitecture' and 'Techeology'. As Dempsey defined it, Marchitecture denotes an architecture produced by a vendor for marketing purposes, and techeology is a mixture of technology and ideology. Dempsey made some very good points outlining common biases he feels are hampering the advancement of digital libraries.

MacKenzie Smith from the Massachusetts Institute of Technology (MIT) talked about the DSpace digital repository system. She noted that DSpace—originally developed by MIT and Hewlett Packard Labs—was released as open source in November 2002. The original producers of DSpace are now trying to foster the growing community of users. Managing, taking part in, and contributing to an open source project is a novel approach in the development of digital libraries. DSpace may not have been the first digital library software project to go open source, but they may be the first to, at least publicly, realise the importance of actively nurturing the community. It remains to be seen if DSpace will successfully manage the transition from a sponsored project to an open source project and turn users into active contributors. Nevertheless, they seem to be off to a good start since BioMed Central during, but not at, the conference announced that they are going to offer a hosted institutional repository service based on DSpace [3]. During the same session as Smith's, George Pyrounakis from the University of Athens also talked about institutional repositories and their decision to adopt Fedora instead of DSpace for multiple heterogeneous collections.

Kurt Maly from Old Dominion University talked about a system called Kepler, which can best be described as a personal or team repository, contrary to an institutional repository. This brings a new meaning to the term 'personal library'. For instance, rather than serving as a repository for my collection of documents produced externally, Kepler is a repository for my collection of documents I (or my team) produced. The personal version of Kepler can be installed in a matter of minutes, whereas the team version takes about an hour to install. Since Kepler may be installed on an author's laptop, this raises reliability issues, which can be solved by integrating Kepler into an institutional repository.

During the new models and tools session, Geneva Henry from Rice University gave a presentation about a somewhat different project. The Connexions project is not a digital library but rather an alternative way to publish course materials. In Connexions authors create knowledge modules that can be integrated into courses. These knowledge modules can be modified and reused and are licensed with a Creative Commons "By" license that only requires attribution for reuse. This form of knowledge production and sharing is what Benkler [4] calls commons-based peer production. It was also intriguing to learn about the effort made to solve legal issues. The Connexion project has addressed licensing issues in the areas of open source licences of software tools, open content licensing, and site license for Connexions.

Ann Apps from MIMAS at the University of Manchester spoke about the development of a Web Services interface to the zetoc current awareness service. (In another session, Ken Eason's presentation, described later in this report, also dealt with zetoc.) Ann Apps' project is an interesting one, as we will certainly see more and more use of Web Services within digital libraries. Indeed, a recent initiative called the Vendor Initiative for Enabling Web Services (VIEWS) [5] was launched in June 2004 by vendors and library service organizations and is aimed at the enabling of web services between disparate applications used in libraries.

The session on personalisation and annotation included one paper by Agosti et al. about annotations in digital libraries (which drew my attention since I frequently annotate and mark-up all articles I read with a pen or a magic marker). The paper explored similarities between digital libraries and collaboratories. The focus of collaboratories is to facilitate interaction among a team—and annotation is certainly one kind of interaction. The annotations I do are mostly on printed versions of e-journal articles—and those annotations are seldom read, or are even readable, by my colleagues. On the other hand, I also use the annotation capabilities of both Microsoft Word and Adobe Acrobat, but in that case it is mostly to annotate pre-prints of my colleagues' papers. On the Internet most annotations are made as comments to articles such as you can do on the Slashdot [6] site. Agosti's ECDL paper elaborates on a comprehensive model of annotations, and I will be interested to follow up on his team's future research.

Katriina Byström and her colleagues presented the result of a holistic evaluation of an information system. The evaluation was made both from the system perspective and from the user perspective in the actual environment in which the system was used. The study revealed that there are discrepancies between the two evaluation results and that the two methods of evaluation complement each other well.

Ken Eason from the Bayswater Institute presented the results of a usage survey of the zetoc current awareness service. The zetoc service is used to electronically deliver table of contents data to 13,000 users. Zetoc has been OpenURL enabled, and this allows users with an institutional OpenURL resolver to get direct electronic access to the 'appropriate copy' of the articles (that is, if they have subscribed to them). Some users reported that the alerts they themselves had set 'haunted them'. I have exactly the same feeling, as sometimes it feels like the alerts I have set for myself are self-inflicted spam. The result of the survey showed, perhaps not unsurprisingly, that users want electronic 'join up'- direct access to electronic full text. But what was surprising was the extremely high satisfaction expressed by the users who had access to an OpenURL resolver and at the same time had access to a lot of electronic journals. This survey makes a small but strong case for the use of institutional OpenURL resolvers.

Vicky Weissman from Cornell University spoke about developing a policy language that can be understood by both humans and computers. A policy says that under certain conditions, an action—such as downloading a file—is permitted or forbidden. A policy may be used by a content provider to specify how its works may be accessed, and subsequently enforces the policy. XrML is a language for writing policies, but it requires a significant amount of training. The solution suggested by Weissman is called Rosetta, a language that allows policies to be expressed as plain English and then translated into XrML. While Weissman's presentation focused mainly on expressing policies regarding access to content, I related this to my own presentation and to the possibility of expressing privacy policies in an easily understandable way. This issue was also raised by others in the question and answer period following Weisman's talk. It would be interesting to see how Rosetta could be employed by digital library users to express their privacy policies. Social networks such as LinkedIn, Friendster and FOAF were discussed during one of the panel sessions. In these social networks people are already revealing quite a lot of private information, and it would be very interesting to see if any application like Rosetta could be used to express the privacy policies of these users.

At the conclusion of the conference, it was announced that next year's ECDL conference would take place in Vienna, Austria. A Call for Papers has already been issued. More information about ECDL 2005 can be found at <>.


[1] Surprisingly one paper on the agenda was never presented.

[2] See the ECDL 2004 web site at <>. The conference proceedings are available from Springer-Verlag as Research and Advanced Technology for Digital Libraries. 8th European Conference, ECDL 2004, Bath, UK, LCNS 3232, <,11855,3-152-22-34393454-0,00.html>.

[3] Open Repository, <>.

[4] Benkler, Y. (2002). Coase's Penguin, or, Linux and The Nature of the Firm. The Yale Law Journal, 112(3), 369

[5] VIEWS: Vendor Initiative for Enabling Web Services, <>.

[6] Slashdot, <>.


Copyright © 2004 Jonas Holmström

Top | Contents
Search | Author Index | Title Index | Back Issues
Previous Report | Next Report
Home | E-mail the Editor


D-Lib Magazine Access Terms and Conditions