Home Overview Resources Workshops OAI-PMH Tutorial Project Documents Project Partners
open archives forum

Information Resource |  Glossary |  Other Resources |  Technical Validation

Overview:    Technical Validation Questionnaire  |   OAI in Europe
Experiences:    OAI-PMH pilot  |   Implementation of OAI-Services
Inventories:    Open Archive Tools  |   Services (pdf-file, 517 KB)  |   Repositories (pdf-file, 275 KB)

Comparison of DSpace and Eprints

Programming language
Operation system

more   Printable Version   (pdf-file, 737 KB) 

more   Inventory of Open Archive Software Tools

Most the tools listed in the appendix are offering solutions for specific requirements e.g. only document archiving systems for Data Provider or special search functions needed for Service Provider or deployment of an OAI compatible interface to an existing web server/database. Like the results of the Technical Validation Questionnaire already showed within the last years obviously expectations to such tools changed.

The growing number of organisations interested in opening their archives via the Internet as well as the increasing number of digital and digitised data cause a break with early, some times experimental or isolated solutions. Large data sets produce a large need on ready, user-friendly complete solutions, which cover typical functionalities, are to be installed by relatively small expenditure, if necessary are adaptable to special requirements and causes also little expenditure with the further care of the data.

Two developments for archiving, which both pursue the objective to fulfil these requirements are comparatively confronted below under different aspects and characteristics: The GNU EPrints self-archiving software, that has been developed at the Electronics and Computer Science Department of the University of Southampton, UK. And DSpace, a newly developed digital repository as a joint project of the Massachusetts Institute of Technology (MIT) Libraries and the Hewlett-Packard Company, USA. These systems base on different technologies but are nearly identically in their functionality - search functions, document archiving, online interfaces for self archiving, integration of the OAI PMH, ...

to the top




Eprints is easy to set up: An installation script automates most of the installation processes.

It is possible to chose between a source- or binary-installation. With the source one the software has to be compiled by the programmer. The binary one is precompiled for special architectures like Solaris Sparc systems. The programmer only need to configure the software.

MySQL, Apache and mod_Perl, the components which are necessary for implementation are smooth installations - no matter if source- or binary-installation is chosen. The installation of additional required Perl modules need more time to resolve the dependencies.

There are two possibilities to support the system: One installation variant is a Solaris environment. The second variant, Linux, is easier to maintain.

If any installation problems are arising a comprehensive support is ensured. GNU Eprints has a separate website containing documentation, downloads, demonstration server and mailing lists: http://software.eprints.org/

The installation of DSpace requires a little more effort. But in fact DSpace is easy to run and maintain for any experienced systems engineer.

In order to run DSpace the following list of Software is necessary to be installed and configured before: Java 1.3, Tomcat 4.0+, Apache 1.3, PostgreSQL 7.3+, Ant 1.5. Details of the requirements can be viewed at: http://dspace.org/technology/system-docs/install.html#prerequisite

If the programmer follows step by step the installation documentation, Java, Ant and PostgreSQL are easy to install successfully.

To set up DSpace man needs to compile the DSpace source code with java tool Ant. The Tomcat server must be started by user "dspace" and user "dspace" should then create a database named "dspace".

With the installation some common problems arose, e.g. that Tomcat doesn't work when the DSpace is connected to Tomcat. Some changes in the configuration script solved that problem.

There is no support service for the DSpace installation. But there is a detailed system documentation at: http://dspace.org/technology/system-docs/index.html. And also a public mailing list for the installation questions is supported.





Both environment variants had been tested: Solaris and Linux.

Furthermore it is also possible to install Eprints2 on any computer that is running with GNU/Linux or UNIX operating system.

DSpace had been tested on Linux Suse 7.3.

In general DSpace can run on Solaris, Linux and Windows systems.


Eprints is free software which creates online archives.

It is possible to store documents in any common format that the archive administrator defined to be accepted. Each individual research paper/ eprint/ ... can be stored in more than one document format.

The archive can use any metadata schema; the administrator decides what metadata fields are held about each eprint. This is specified in three or four stages:

  • Definition of a maximal set of metadata fields that should be stored (e.g. authors, title, journal, journal volume, etc.)
  • Definition of different types of eprints (e.g. refereed journal article, thesis, technical report, unpublished preprint, etc.)
  • Specification for each type which metadata fields should be stored, and which of those fields are mandatory.
  • Decide how these metadata fields should be projected into the Open Archives world. (If necessary, interoperability can be switched off, but this is strongly discouraged.)

More functions can be viewed at http://software.eprints.org/

DSpace can be used for self archiving by institutions and faculties. It provides long-term physical storage and management of digital items in a repository.

DSpace is organised into "Communities" and "Collections", each of which retains its identity within the repository. It supports a variety of digital formats and content types including text, images, audio, and video and allows contributors to limit access to items in DSpace. All these items can be organised by an administration interface.

DSpace supports the OAI protocol 2.0 as a data provider. This OAI support was implemented using OCLC's OAICat open-source software to make DSpace item records available for harvesting.

Currently DSpace supports only the Dublin Core metadata element set with a few qualifications conforming to the library application profile. But there are still developing plans to support a subset of the IMS/SCORM element set (for describing education material) in the coming year.

More details of DSpace functionality can be founded at http://libraries.mit.edu/dspace-mit/technology/functionality.pdf


Eprints is widespread all over the world. In August 2003 there are 72 worldwide archives running Eprints software officially listed (http://software.eprints.org/).

It is not reported how many archives are running DSpace software. One example of an European repository that implemented DSpace is "Erasmus University: Research Online".


Eprints uses traditional technologies and runs on pure Open Source systems: mySQL is the world's most popular open source database, recognized for its speed and reliability and Apache has been the most popular web server on the Internet since April of 1996.

Eprints is programmed by using the script language "Perl", that is low level but powerful.

DSpace operates with new technologies such as the Postgres database, that is more advanced than mySQL and Tomcat for jsp/java web application, that has higher performance than eprints.

Dspace supports and includes also handle server, which ensures that each document has unique and persistent URL.

Optionally, DSpace can be protected by the security features (SSL) of Tomcat. It is also possible to use the redirect function (port number can be omitted) from Apache referring to Tomcat.


Eprints is freely distributable and subject to the GNU General Public License. This means that its source code is open and freely modifiable by any programmer who wishes to modify it (on condition that modifications are all free and open).

Therefore in principle an adjustment to every environment is possible even if it is different than the recommended. Naturally this may be connected with substantial expenditure.

However Eprints offers no supporting documents there are nevertheless mailing lists for support.

The DSpace system is freely available as open-source software. This allows to make any necessary changes to the downloaded copy. The system was designed to make adaptations for individual organisations as easy as possible.

In fact, several modules in DSpace will probably be customised by organizations using this tool (e.g. it might be necessary to get authorization and authentication for more than one person). Or some organisations may want to adapt a different environment than recommended (e.g. replace postgreSQL by mySQL or Oracle). At the moment, substituting a different relational database than postgreSQL will require just a few changes to the system's Browse module.

Java provides documented Java APIs that can be enhanced to allow interoperation with other systems that an institution might be running (e.g. auto-depositing in DSpace a department's web document system, or the campus data warehouse).


Eprints allows to scan each of the metadata field types in the database by simple or advanced search. Any metadata field can be searched with fine granularity by SQL querying the database.

Further information could be found at http://software.eprints.org/ files/eprints1/docs/system.html# SECTION00060000000000000000

DSpace offers two levels of text search: simple and advanced search. It's submission process also allows to use a qualified version of the Dublin Core metadata schema for the description of each item. These descriptions are stored in a relational database, which is used by the search engine to retrieve items.

to the top
Overview:    Technical Validation Questionnaire  |   OAI in Europe
Experiences:    OAI-PMH pilot  |   Implementation of OAI-Services
Inventories:    Open Archive Tools  |   Services (pdf-file, 517 KB)  |   Repositories (pdf-file, 275 KB)


The Open Archives Forum (OAF) is an IST– Accompanying Measures project (IST- 2001-320015).
The partners of OAF are: University of Bath-UKOLN (United Kingdom), Istituto di Scienza e Tecnologie della Informazione-CNR (Italy) and Computer- and Media Service (Computing Center) of Humboldt University (Germany).

information societies technologies