Nicole Van Kempen , Coordinator, AVEL - Australasian Virtual Engineering Library [HREF1] , Physical Sciences and Engineering Library [HFEF 2], Hawken Building, University of Queensland [HREF 3], St Lucia 4072, n.vankempen@library.uq.edu.au
The Australasian Virtual Engineering Library (AVEL) has been an important contributor to the development of Australian subject gateways since it began in 1999. The site was officially launched in July last year at the Institution of Engineers' Australia, Excellence Awards. Designed to provide easy access to quality IT and engineering Web-based resources, it was initially created to support the information needs of researchers working in the fields of engineering and information technology (IT). Originally conceived as a searchable, browseable database of high quality resources it has continued to change and add services in order to better meet the needs of its target audience. The database now contains over 3300 high quality records, covering all areas of engineering and IT.
This paper describes the evolution that AVEL has undergone in order fulfill, as well as anticipate, customer information requirements. It discusses the methods that were used to gauge the needs of customers and details the way that customers are using the site. The paper also outlines some of the current and future services that will meet these demands. The organisational structure of AVEL, its existence as a virtual partnership and the technical infrastructure behind AVEL are also explored.
The Web has been called a "chaotic repository for the collective output of the world's digital printing presses "(Lynch, 1997). The latest statistics from the Web Characterization Project [HREF 4]conducted by the Online Computer Library Centre (OCLC) show there were 7,128, 000 unique Web sites in existence in 2000. From 1997 to 2000 there was a 371% growth in the number of Web sites on the Web. Inktomi [HREF 5] estimate that there are over one billion individual documents on the indexable Web. Not surprisingly many Internet users find it difficult and time consuming to locate relevant information on the Web. Not only is the sheer size of the Web daunting but the quality of the information is often questionable. Anyone can, and does, publish on the Web. Engineers and IT specialists are not immune to the challenges that the Web presents to researchers, indeed they have their own specific issues, requiring tailored solutions and services. These issues are discussed at length further in this paper.
Like many Australian gateways AVEL came into existence with the successful application for a Research Infrastructure Equipment and Facilities Scheme Grant from the Australian Research Council / Department of Education, Training and Youth Affairs in 1998. It was initially created to support the information needs of researchers working in the fields of engineering and information technology (IT) and to provide easy access to quality IT and engineering Web-based resources. It began as a searchable, browseable database of high quality resources.
AVEL is a collaborative venture. Significant financial and in-kind contributions were received from the partner institutions. The AVEL partnership includes the:
The AVEL Team works in a virtual environment which enables team members to complete their contributions to the project anywhere, anytime. The development and sustainability of a collaborative working relationship between geographically dispersed partners has been a key to AVEL's success. This is facilitated through the strategic use of phone, email, and online discussion groups, complemented by limited face-to-face meetings. Achieving a level of transparency and visibility is a key enabling strategy, made possible via an evolving Web page. A Web site for the AVEL Team has been in existence from the beginning of AVEL. This contains all the policy documents on such things as the metadata schema, the resource selection criteria and instructions on how to enter metadata, the thesaurus schema and areas of responsibility for partners, as well as presentations made by AVEL staff.
The need for AVEL was evident from the beginning. The anecdotal experience of librarians liaising with departments and conducting information skills classes for IT specialists and engineers suggested that these researchers found it difficult to keep up to date or lacked the skills or time to find relevant information on the Web efficiently. They were often unaware of new or significant information resources on the Web or were confused about how to access or download these resources. In addition, although high quality engineering and IT gateways existed elsewhere in the world, there was no comparable Australian-based, comprehensive source of quality Web information. Free subject gateways such as EEVL (Edinburgh Engineering Virtual Library) [HREF 14], EELS (Engineering Electronic Library, Sweden) [HREF 15] provided good initial models for AVEL.
Like these gateways AVEL uses human skill and knowledge to mediate the retrieval of information on the Web . Gateways provide a form of quality control by using standard metadata schemas to describe Internet resources. The services of subject or information specialists are used to evaluate information. Experts add value to Internet resources by discovering them, describing them and making them searchable and browseable. In the case of AVEL, resources are selected based on a set of defined resource selection criteria. Engineers, IT professionals, subject specialists and other interested parties also suggest sites for inclusion. All selection and record creation is done "by-hand". Although the software underpinning AVEL supports the use of automated metadata harvesting this has not been implemented.
AVEL Record Creators manually produce a record for each site and they are described using a metadata schema that is based on Dublin Core. In essence metadata [HREF 16] is a "structured set of data which describes the key defining characteristics of a resource". This might include the title of the resource, who created it, when it was created, where it is located and its format. A metadata editor (REG) designed by the Cooperative Research Centre for Enterprise Distributed Systems Technology provides a template for creating records, a MySQL repository stores the records and HOTMETA software provides a searchable / browseable interface.
Customer expectations of gateway services as well as their information needs are dynamic. To be successful, the innovative gateway needs to identify these needs and tailor their services to meet them. They also need to continuously monitor customer demands and react accordingly by adapting or adding new services and information resources as required. In order to identify, anticipate and fulfill the needs of our audience AVEL has employed a number of strategies and undertaken a range of qualitative and quantitative research. Some of these methods are detailed below.
Developing a formative and open relationship with the prime user communities has been one such crucial strategy. It has enabled AVEL to more easily define what users want as well as given increased credibility to the site. One of the ways AVEL has fostered such communication is through workshops and presentations at the Australasian Association for Engineering Education and the Australian Council of Engineering Deans both prior to and during the formative development of AVEL. Presentations to engineering faculties, schools and departments at various universities have also been important as has the inclusion of news about AVEL in engineering publication. AVEL's strong working relationship with the peak Australian professional society for engineers (The IEAust) has also increased the profile of AVEL as a credible information resource.
The target audience for AVEL is diverse and includes university researchers, undergraduate and postgraduate students, researchers in the R & D departments of Australian companies and Australian government departments as well as engineering and IT professionals working in industry.
A series of focus groups with user groups were also important in ensuring that the design of the system was user-driven and informed by real customer demands and expectations. The participants were engineers and IT researchers, academic staff and postgraduate students representing all Engineering Departments including the School of Information Technology at The University of Queensland. A second focus group was also held with staff and students from QUT.
The aims of the focus group sessions were:
Participants were asked to describe what type of information they looked for on the Web. Were they interested in finding information on a broad topic, or on a specific problem? Did they require the expertise of a colleague or were they looking for free full-text information? Not surprisingly, the responses indicated that users look for value-added features that distinguish AVEL from its rivals, especially from the larger, established search engines such as Yahoo! and Google. Respondents echoed established research, which suggests that engineers primarily use personal networks and contacts to seek and exchange information. AVEL was used to find technical reports written by colleagues, as a tool to source grant funding and as a method of locating current research, especially research conducted by University Cooperative Research Centres. Respondents wanted to build additional vertical features on to AVEL to bolster it as a interactive and dynamic "community" resource. These included the addition of discussion groups, an expertise directory and Standards information.
Specific questions were asked about the browsing function. Was it used, and how was it used? How could it be improved? AVEL wanted to implement HyperIndex browsing and we were particularly keen to get feedback on the usefulness of this refinement. Data from the online survey conducted after the focus groups indicated that around 1 in 5 users like to browse the database. While this is not a huge proportion, we recognised the importance of providing multiple paths to resource discovery and we wanted to implement a browse option that had a close "fit" to the information seeking behaviours of our customers. In response to suggestions made by focus group members we made a number of significant changes to the browse structure and presentation. Top level terms were included on the Home Page, in much the same way that Yahoo! directory structure operates. A tree structure is used once a term is chosen so that customers can easily identify were they are in the browse hierarchy. Respondents indicated that they were often confused about how to display records once they had "drilled down" to the required level. A prominent red "Display" button was added to make this easier. More than one approach to browsing for information was also required. To this end we supplemented the thesaurus terms with an alphabetical index at the top of the browse screen. Resulting terms are then listed in alphabetical order. To reduce customer frustration we "hid" any terms that did not have records attached. Records have subsequently been added for all thesaurus terms.
The focus groups were extremely useful in that they enabled us to identify future services of importance to our customers as well as improve and expand on existing services. Based on the feedback from the session changes were made to browse function that made it easier to use and identify results. Key subject areas and resource types were identified to give more breadth and depth to the database. Regular link-checking procedures were implemented and database maintenance procedures were established to ensure that information was current. Emphasis has also been placed on collecting more full text documents.
Another method of measuring the ways users currently use AVEL, as well as gauging the types of features they would like to see offered in the future has been the use of an online survey. The survey was put on the AVEL Website, and was also distributed to subscribers of the AVEL newsletter. The survey focussed on gathering basic demographic information, current service usage and customer satisfaction with current services as well as identifying the types of services respondents would like to see added in the future.
Thirty-two surveys were completed. The bulk of the responses came from Australian customers, however responses were also received from users based in Malaysia, India, South Africa and the United States. Interestingly AVEL web usage statistics also indicate that significant traffic is coming from these countries, especially South Africa. Most of the respondents were Librarians and Information Professionals, however a significant proportion were engineers, academics or students. Almost half the responses came from people employed in the academic or research sector with 19% coming from the government sector and 16% coming from the private sector.
When asked what services they were currently using, full text access and current awareness services featured strongly. 21% of respondents said they used full text papers, 20% used the newsletter, and 14% used the "Latest Resources" feature of AVEL. Other services such as the "Online Bookshop" and the "Conference Announcements" section received an 8 - 10% score.
In terms of future services that customers would like to see added, the majority of respondents echoed the desire for full text information and current awareness services. Most said that "Links to Electronic Journals" (21%) or an "Alerting Service" (20%) were features that they would like to see added. 13% favoured an "Expertise Directory" while 12% each wanted to see "Online Seminars by Experts" or an "Ask and Expert Service".
Although this is a fairly small return rate some tentative observations can be made from the data collected. The survey certainly serves as a starting point from which other investigations can be made and the data can be used in conjunction with information gathered from the other research methods that AVEL has employed. The full results of the survey [HREF 17] are available from the AVEL Website.
In addition to these measures, other methods have been used to evaluate usefulness and effectiveness of AVEL on an on-going basis. These include: analysis of suggestions and comments sent to AVEL Website, the number and variety of subscribers to the AVEL e-newsletter, and Website usage statistics to analyse both the number of users over various periods of time and where the users are coming from, and where they exit the AVEL page.
The results of the focus groups and the online survey echo what has long been established in the literature about the information seeking habits of those working in engineering, science and technology. Researchers use the Web not only to find information but also to "to maintain their identity, to engage in discussion and to circulate information" (Ballantyne, 2000).
With regard to engineering in particular, the emphasis is usually on solving a defined technical problem. As a result engineers rely heavily on colleagues, internal sources of information, and personal knowledge networks. But as the engineering discipline moves towards linking with non-traditional areas such as the environmental, health, biological and social sciences, both problem solving and keeping current are becoming more difficult. Dependence on personal networks and work colleagues for information will no longer be sufficient.
Perhaps equally as significant are the results of other studies that suggest that despite their technical backgrounds engineers and IT researchers often fail to take full advantage of electronic information sources. One reason proposed for this is that many information systems are not designed to support the problem-based methods that they employ, nor to they mimic the knowledge networks they rely upon. They are also often perceived as difficult to use (Leckie, 1996).
AVEL has made a number of significant changes to its interface in order to make it easier to use. The Web page has been redesigned and is now divided into three columns. All search, browse and service options are immediately obvious to the user. The central column contains the browse and search options (basic and advanced), while the outer columns detail services. The Web page colours have been changed and the options clearly segmented by different banners and colours. Changes have been made to the browse functions. Top level terms are identified by bolder type , and a complete hierarchy is available to select from. If a thesaurus term contains no record, it is hidden from public view.
The results of our own research, as outlined above, indicated that AVEL users were interested not only in links to useful Web sites, but to actual full-text documents. The AVEL metadata schema was altered to add the full-text field to the metadata. This makes it possible to search only for full-text documents if the users wanted. All research higher degree theses in the fields of engineering and information technology available in pdf format from the Australian Digital Theses site are indexed and accessible from AVEL.
Responding to the need for engineers and IT specialists to use the AVEL as a virtual space, enabling interactive communication and knowledge exchange we have added a number of services. AVEL is in the process of moving from the limited definition of gateway [HREF 18]: " a Web-based mechanism for accessing a collection of high quality, evaluated resources identified to support research in a particular subject discipline" to provide a space that encourages and enables dialog.
In order to keep professionals updated with the latest developments in their fields we have added features such as a "What's New" section, where the resources added to AVEL in the last month are displayed. We list announcement s for conferences for the current and coming year. We offer a "Jobs and Employment" Section, as well as links to an "Online Bookstore". A free monthly newsletter is produced by AVEL staff that acts as an alerting service, keeping subscribers up to date with new sites and developments at AVEL. We also mirror the SciTech newsletter, which is produced in the United States by the National Science Foundation.
In the near future it is planned that AVEL will continue to add new features which capitalize on users needs for a virtual research space. A model is contained below. As detailed in the model, we plan to link to and interoperate with existing library, information and research infrastructures. AVEL will build on the proven strengths of existing leading virtual libraries and gateways (such as New Zealand Digital Library, Australasian Virtual Engineering Library, Edinburgh Engineering Virtual Library, Electronic Engineering Library Sweden) and streamline current processes to improve research efficiencies and build a new level of productivity targeted at Australian researchers.
Other targeted services we plan to provide include:
The model below represents a vision of AVEL's future the services and structure:
The need for subject specific, focused, quality Web resources is self-evident. The Web is currently too unwieldy for researchers and information seekers to efficiently navigate unaided. In fields such as engineering and IT the importance of human involvement at the point of resource selection and description remains crucial. For the immediate future there is no automated harvesting process that can add the same value to resource discovery as a real, live subject expert.
Customer expectations of gateways services as well as their information needs are dynamic. Innovative gateways need to identify these changing needs and evolve to meet them. The qualitative and quantitative research undertaken by AVEL, as well as much of the current literature, indicates that engineers and IT professionals are suffering from information overload. They welcome the chance to customize their access to quality information. They also value gateways that parallel the way they operate in the real world. For AVEL this means operating as a network, facilitating community interaction and enabling information exchange.
Ballantyne, P. and Addison, C. 2000, 'Cooperation on the Web: trends and practice in the development sector', Information Development, vol. 16, no.1, pp. 15 - 23
Leckie, Gloria J. , Pettigrew, Karen E. and Sylvain, Christian. 1996 'Modelling the information seeking needs of professionals: a general model derived from research on engineers, health care professionals, and lawyers,' Library Quarterly, vol. 66, no.2, pp. 161 - 189
Lynch, C. 1997, 'Searching the Internet', Scientific American, vol. 276, no. 3, p.52 .
Schwartz, Evan I. 1999, Digital Darwinism : 7 breakthrough business strategies for surviving in the cutthroat Web economy, Broadway Books, New York
HREF 1
http://avel.edu.au
HREF 2
http://www.library.uq.edu.au/pse.html
HREF 3
http://www.uq.edu.au
HREF 4
http://wco.oclc.org/main.htm
HREF 5
http://www.inktomi.com
HREF 6
http://www.uq.edu.au
HREF 7
http://www.umsw.edu.au
HREF 8
http://www.monash.edu.au
HREF 9
http://www.unimenb.edu.au
HREF 10
http;//www.qut.edu.au
HREF 11
http://www.ieaust.org.au
HREF 12
http://www.cmte.org.au/
HREF 13
http://www.dstc.edu.au
HREF 14
http://www.eevl.ac.uk/
HREF 15
http://eels.lub.lu.se/
HREF 16
http://www.library.uq.edu.au/iad/ctmeta4.html
HREF 17
http://avel.edu.au/results.html
HREF 18
http://www.nla.gov.au/initiatives/sg/servicetypes.html
Nicole Van Kempen, © 2000. The author assigns to Southern Cross University and other educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Southern Cross University to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers and for the document to be published on mirrors on the World Wide Web.