SOSIG

Using SOSIG to support Social Science Teaching and Research

The Internet can provide an invaluable resource for supporting education and research in the social sciences; it offers access to people, data and resources on a hitherto unparalleled scale. However, even with the advent of the more user-friendly World Wide Web, it is as yet a far from ideal work environment. The sheer enormity of information available and the corresponding lack of organisation of this information can prove an effective barrier to potential users.

The Social Science Information Gateway (SOSIG) allows social science researchers and practitioners easily to discover and access relevant high-quality networked resources and services world-wide including data archives and statistical software. It also provides a comprehensive list of relevant UK-based sources. SOSIG points to hundreds of resources on subjects ranging from anthropology to statistics. All the resources that appear on the gateway have been catalogued and described to aid users.

Introduction

The basic definition of the Internet is a global network of networks, a physical infrastructure linking hundreds of thousands of computers and enabling, via various software tools and information 'protocols', the exchange of information between them and between more than 30 million people across the world. With electronic mail, discussion and news groups it provides a method of communicating with people locally, nationally or internationally on an individual or group basis in a way that is both fast and economical. The Internet also allows individuals to use services and systems - such as library catalogues and data archives - 'hosted' on computers anywhere in the world when they do not have access to such services locally.

The culture of the Internet is still very much one of co-operation and a readiness to share knowledge and expertise with others, so, whatever your interests or specialisms, there are likely to be groups of people who share those interests on the Internet. Traditionally academics have shared resources such as working papers, articles, library catalogues, data archives and so on and the number of these has grown as publishing information on the networks has become easier. However, as more commerical institutions and businesses are becoming involved in the Internet a whole new range of information and services are also appearing.

The Internet - this global network of networks and the means of exchanging information - has been around for 20-30 years, but the advent of the World Wide Web (the set of tools and standards which allow images, sounds and animations to be transmitted and offer a 'graphical user interface' to information) has made publishing and accessing information much 'friendlier' and, in theory, much easier. When Internet access first escaped from the domain of the computer experts in this way, many others in the academic community 'gave it a try' and found it wanting. The most common complaints were the lack of availability of useful materials and the difficulty in locating the few that did exist. These problems persist with the continuing rapid expansion of the Internet: imagine a vast second-hand bookshop with a constantly expanding and changing stock but with little more than the spines of hundreds of thousands of books visible. Rose Growing and Essays on Stress jostle on the shelf next to Developments in Cognitive Psychology. Just like browsing the Internet, random selections from the shelves may uncover something up-to-date and worthwhile, but may just as easily disclose nothing more than outdated or frivolous comics on subjects of little or no interest. The time and effort taken to clamber to the top shelf to retrieve Essays on Stress will have been wasted if eventual perusal of the back cover or a quick flick through its pages reveal it as a collection on stress fractures in bridges (or worse still, simply a list of titles of other such works) rather than on the psychological and physical effects of occupational stress you were looking for. A recognition of these difficulties led to the eventual establishment of the Social Science Information Gateway project in the UK.

Background to the Project

In 1992 the UK's Economic and Social Research Council (ESRC) appointed a Networked Information Support Officer to examine the potential for the use of networked information amongst the UK social science community and to encourage further development. There was a perception that the social science community were not reaping the benefit from networked resources in the same way as their colleagues in the natural and physical sciences. Through holding training sessions and workshops during these early days, it became clear that, whilst researchers could eventually locate resources where guided by an instructor and assisted by training documentation, the picture was very different when they tried to incorporate these newly-acquired skills in their day-to-day work. Disappointment and frustration soon followed the exhiliration and excitement of the first trips on the Information Superhighway. Neither the inclination nor the time were available for extensive browsing in an attempt to find something useful. It was decided to try to provide the social science research community with an easy way of locating electronic information and data which could be used in their work. The project which grew from this idea was the Social Science Information Gateway (SOSIG). A pilot service of the gateway became operational in June 1994.

Use of World Wide Web

The primary objective of the project was to establish a 'one-stop- shop' of UK social scientists, to connect the user seamlessly to relevant resources regardless of their location. World Wide Web was chosen as the software tool to deliver this solution for various reasons: it provides the consistent and easy-to-use interface that was required and was the fastest-growing NIR (Networked Information Retrieval) tool at the time, with development work being carried out in Europe and the United States. It also allows access to many other processes and protocols such as gopher servers, ftp sites, telnet sessions and many more of the earlier systems and services hosting useful social science data. The subsequent development and popularity of the Web for delivery of both academic (teaching and research) and commercial and recreational information serve to vindicate that choice now the Gateway is well- established and others are following the model.

Structure of SOSIG

There are over one thousand links to social science resources on SOSIG at present, with new resources being added on a regular basis. Users are presented with an uncluttered 'home' page with self-explanatory and simple-to-use buttons (with text alternatives) allowing browsing of resources by subject heading (subjects now range from accountancy and anthropology to sociology and statistics), Universal Decimal Classification (UDC) order or searching of the SOSIG database of resource information. Users have the option of connecting directly to resources they locate through SOSIG or reading descriptions and keywords before making the decision whether or not to connect. Unlike many other gateways or subject listings, the project tries to maintain a level of quality control. Resources do not appear on SOSIG until they have been filtered, recorded, classified and described.

Identification and Collection of Resources

A variety of sources are scanned for relevant resources including:

Mailing lists and newsgroups: Various lists and newsgroups are scanned for announcements of interesting social science resources, some of these groups are subject related but others are general lists set up specifically to monitor and disseminate information about new network resources.
Printed guides and catalogues: These often provide a good starting point for identifying resources within a particular subject area.
Networked search tools: There are a number of tools or robots that aim to build indexed catalogues of resources available over the Internet, such as Veronica, Archie and Lycos and the ever- growing number of other Web search engines and virtual libraries.
Other networked services: Following links from other services on the Internet.

The project also has a number of volunteer LISTeners. These are generally subject specialists in the social sciences who can advise on the quality of resources. In addition to recommendations for the LISTeners, the project also regularly receives recommendations from the users of the service. An on-line form is available via the 'Add New Resource' button on SOSIG's home page for users to e-mail suggestions and additions (as well as occasional corrections) to the gateway. These are subject to the same quality checks before they are added to SOSIG. Regular, automated checks also avoid 'dead' links which occur when sites move leaving no forwarding address or vanish altogether - although as one of the selection criteria for SOSIG is 'stability' this happens less often than elsewhere.

Filtering Resources

An important role for SOSIG is to filter out 'junk' - resources that are of little or no use to our users. Resources are chosen according to selection criteria that include areas such as relevance, reliability, stability and currency. The popular Web search engines have much larger databases and often retrieve thousands of 'hits' which at first glance look exciting, but which are often in most cases irrelevant. With SOSIG, the user searches a 'focussed' and selective database of resources.

Recording Resources

There is very little meta-data or descriptive information about resources available on the Internet, often no more than a file or directory name - the equivalent of the spine of a book. This can result in users choosing a link or downloading a file, waiting possibly minutes while it transfers to their system (often from the USA) only to find it wasn't what they wanted at all. All the resources that appear on SOSIG have been catalogued using a standard pro-forma or template. The template, which includes a description of the resource, underlies the search mechanism which is available on SOSIG. A keyword search will provide you with a list of resources that match your criteria, each of which will dynamically link you to the resource described, wherever it is in the world.

Classification

Each resource is classified using the Universal Decimal Classification (UDC) Scheme. Use of this scheme was agreed with two other national service providers, NISS and BUBL, to allow for collaboration amongst the projects. Although the UDC underlies the organization of the resources, a strict hierarchical scheme is not enforced, so if a subject has recently become important enough it can find a place on the top menu. Individual resources can also be cross-classified so that they can be found under several different subject areas. The subject categories can be viewed alphabetically (the default) or according to the UDC.

Developments

In 1995 SOSIG received additional funding from the ESRC to allow continuation of the established service for a further two years. Funding from the Electronic Libraries' Programme (eLib) - which has its roots in the Follet Report on IT in Libraries - has also increased the programme of training and awareness of resources for social science teaching and research with the appointment of a Documentation and Training Officer. Workshops running over two days are provided at host sites for research students, academics, support staff and subject librarians, with hands-on sessions built around self-paced on- line and paper materials and a number of presentations and demonstrations. Training materials are available on-line and for downloading, tailoring and use by UK HE institutions free of charge for non-commercial use with appropriate attribution from the SOSIG home page. Evaluation of the training and the general effect and usefulness of subject-based services is also being undertaken as part of the eLib project.

Underlying SOSIG is ROADS (Resource Organisation and Discovery in subject based Services), a two-year collaborative project funded by eLib. ROADS provides the software tools and standards for SOSIG and a number of other subject-based gateways now being built on the SOSIG model, including the templates for recording resources and tools for indexing and searching. The project builds and extends on the work on resource descriptions already in process on SOSIG. The project will also implement a system that allows users to search across several different subject based services seamlessly. This will initially be piloted on the SOSIG service, OMNI (Organising Medical Networked Information) and the Electronic Libraries Information service at the UK Office of Library Networking (UKOLN).

EU funding for the DESIRE project involving eight European partners will build further on the software tools and standards underpinning the gateway and will allow SOSIG to offer a European focus as well as the existing UK and more general 'worldwide' sectors. Consideration will also be given to the extension of cross-database searching, improvements to search mechanisms, meta-data, indexing and cataloguing standards and practices as well as issues raised by the multi-lingual nature of many of the resources.

For more information about the ROADS project contact the authors or see the URL: <http://ukoln.bath.ac.uk/roads/>

Information about DESIRE is available at: <http://www.nic.surfnet.nl/surfnet/projects/desire/>

Access to SOSIG

Users with World Wide Web clients such as Netscape or Mosaic can access SOSIG by typing the URL <http://sosig.ac.uk/>

Users without WWW clients can access the service using the Lynx client. This will give you a text-only based interface to the service. Make a telnet call to:

sosig.ac.uk
login as: sosig

Debra Hiom, Research Officer
Lesly Huxley, Training Officer
SOSIG/DESIRE, University of Bristol