Roger Debreceny [HREF1], Nanyang Business School, Nanyang Technological University, Nanyang Avenue, Singapore, 639678, Republic of Singapore. RogerD@netbox.com
Allan Ellis [HREF2], School of Social and Workplace Development, Southern Cross University, PO Box 157, Lismore, NSW 2480, Australia. AEllis@scu.edu.au
The AusWeb series of conferences commenced with AusWeb95. From its inception, the AusWeb series has used the Web to enhance the productivity of this research conference. End-to-end electronic submission, refereeing and publication has been employed. Publication of the proceedings on the Web and, with AusWeb99, pre-delivery to delegates of the printed proceedings has been designed to ensure that the delegates are fully briefed on the content to be discussed at the conference. The option to link web-based multimedia presentations to the formally refereed papers provides authors with the opportunity to expand upon, present updates and in general invite delegates to explore in greater detail the ideas and concept they are presenting was introduced this year. Conference sessions have been designed to maximise the time for the delegates and to enhance interaction between authors, moderators and delegates.
AusWeb has employed a single electronic authoring/multiple publication process: Web, paper and CD-ROM. This approach places constraints on the form and method of submission and publication but provides advantages in terms of time, cost and accuracy. Directions for the conference are discussed and readers are invited to contribute their ideas for AusWeb2K.
The AusWeb [HREF3] series of conferences commenced in 1995. AusWeb99 marks the fifth conference in the series, with AusWeb98 being incorporated into the 7th International World Wide Web conference, WWW7 [HREF4]. AusWeb attempts to bring traditional features of the research conference together with the broader reach of the Web's information dissemination and enhanced interaction and productivity. This paper outlines some of the experience of the AusWeb team in designing and managing this conference over a half-decade period. The paper sets out the logic of the design of the production and presentation of knowledge. It describes possible directions for the series. The feedback of readers of the paper are also sought to guide the AusWeb team in setting the future direction of the conference.
The remainder of this paper proceeds as follows: In the next section we describe the background to the conference and outline the management philosophy. In the following two sections we give an account of the design of the face to face conference and the simultaneous "author once/publish" production of papers on paper, CD-ROM and the Web. In the succeeding section we analyse what we perceive to be some of the strengths and weaknesses of our approach. We set out some possible directions for the AusWeb series. In the concluding section, we seek assistance of members of the AusWeb community and others who believe in a strong Australian presence in global Web research in guiding the development of the conference series. This is an essentially highly personal piece and we seek similar input from readers.
The idea for the AusWeb series of conferences originated over a bottle of good Australian red wine. One of the advantages of living in a University town such as Lismore, is that interaction between staff can readily take place out of regular university hours and in a congenial environment. We had previously been involved in using a variety of educational technologies. We regularly exchanged notes on existing and new technologies and taught each other new ways of attacking problems. Roger Debreceny had launched in early 1994 ANet [HREF5], an Internet-based information service for accountants, now part of the International Accounting Network [HREF6]. This service had used the Gopher [HREF7] protocol from the outset. It was quickly realised that Gopher only had a limited life. At our request the University Computer Centre brought up the SCU Web site in March 1994, in time for the formal launch of ANet. Throughout 1994 both authors used the Web for a variety of production, research and teaching functions. It was clear to us that this was an important technology for a variety of purposes -- we did not, however, appreciate the commercial implications of the technology. If we had, we might already be retired from University life!
We watched WWW1 in May, WWW2 in October as the WWW series was launched and then followed up with a second conference only six months later. Clearly it looked as though it would take this conference some time to get "down under". Of course, in hindsight, this was not to be the case. In August 1994 over a glass or two of red wine, Allan Ellis said "let's organise an Australian Web conference". Allan had been involved in a successful series of conferences on Open Learning that SCU had run in the early '90s. The area surrounding SCU is an attractive conference destination and we recognised that an element of 'tourism' probably plays a part in everyones' conference selection process. There was also a cooperative spirit at SCU that meant that it was feasible to manage a conference. Within a relatively short period the Ballina Beach Resort was booked and a Call for Papers issued. We had significantly underestimated the demand for such a conference. Almost immediately the conference was sold out and a waiting list had to be managed. A total of 55 papers passed the refereeing process and were presented at the conference. Most of the referees were drawn from researchers and technologists that we had only interacted with by email. A capacity group of 140 people from Australia, New Zealand and several other countries attended the conference. In addition, an enthusiastic team of student volunteers from SCU's School of Media Communications and Asian Studies prepared the Virtual Edition [HREF8] under the supervision of School Head, Chua Siew Keng. Since that point we went on to AusWeb96 and AusWeb97, both held at the much larger Conrad Hotel/Jupiters Casino on the Gold Coast, and WWW7 which incorporated AusWeb98.
From its inception the conference has been run by almost the same team at Southern Cross University. As conference co-chairs, the authors have had overall responsibility for the conference design and management of content. Various units within the SCU family have assisted with the logistics of the conferences, from conference management to computing and audio visual. Essentially the AusWeb team has been made up of less than six people. Allan Ellis has managed keynotes and sponsorship. Roger Debreceny, working closely with Allan and senior advisers, has chaired the paper review panel in most years and managed all liaison with authors.
Over the years, SCU graduates Rod Byrnes and Karey Patterson have managed part or all of the Web site and the print and CD-ROM publication processes. Conference management has been the responsibility of a small team at Norsearch Conference Services, headed by Julie Burton and including Joanne McMurtry and Lisa Messenger. John Miller from the Audio-Visual unit has provided a totally professional audio visual service at all the conferences and various staff from the Computer Centre have supported us in building the necessary networks at the conference venues. The Vice Chancellor, Professor Barry Conyngham, has opened all the conferences. This is a measure of his interest in the Web and his commitment to innovation.
Of course, the series could not be run without the support of a large number of colleagues at SCU and the broader in the research, academic and commercial communities. Many of these colleagues have been involved with the conference since its inception. Indeed, comparison of the list of reviewers, paper presenters and paper reviewers at AusWeb95 and AusWeb99 discloses a significant commonality. Reviewers are often the unsung heroes of any conference. It is their judgement that controls the core content of the conference. Set the standard too low and delegates will be disappointed. Set the standard too high and delegates will be confused. Good reviewers need to be up-to-date in their own area of expertise as well as having some feeling for the "level" of the conference delegates. Of course, reviewers only make recommendations and it is up to the program or conference chairs to make the final decisions. On a number of occasions the authors have acted as 3rd and 4th reviewers in order to resolve situations where the initial reviewers were not in agreement.
The thought of a same-place/same-time, conference on a virtual technology such as the Web seems at first glance to be a contradiction. We have never believed this. It is not that there is no place for virtual conferences such as the Royal Australasian College of Surgeons Virtual Surgeons Congress [HREF9]. The success of this and other similar conferences show that there is a demand for virtual conferences. In contrast, we have seen our market niche as providing a friendly, collegial and productive same-place/same-time conference that was facilitated by the Web. We are staunch supporters of Robert Cailliau's comment at the WWW2 conference, recorded on the final page of the WWW5 proceedings, that reads "there is no such thing as a virtual beer".
The enrolments at each of the AusWeb conferences and the IW3C2's [HREF10] WWW series of international Web conferences shows there is a demand from the broader Web community for a place to share experiences, to learn and to network, face-to-face. For AusWeb in particular, the entire continuum of the Web community has been represented. Papers range from highly technical, leading edge technology contributions to analysing the role of the Web in society. While a physical meeting is neither a necessary nor sufficient condition for community building, we believe it does assist the construction of knowledge and community.
The Web can, however, play a significant role in the processing and distribution of knowledge and the style of the conference. Much of our effort has been to use the Web to make the conference more productive. It can help create more time for face-to-face interaction and it can assist this interaction to be more meaningful. There are two areas that we have concentrated our efforts in improving productivity. The first is the reviewing and publication process, which is discussed below. The second is the preparation of delegates and the management of the conference sessions.
From the outset the conference has employed end-to-end electronic submission, refereeing and publication. This form of information management and distribution has been seen to bring significant gains in productivity (Odlyzko, 1995; Varian, 1997). We have seen this productivity in practice, with relatively short and tight lead times. For AusWeb99, for example, we sought specifically to meet the Australian DETYA's (Department of Education, Training and Youth Affairs) highest "E1" standard for research publication. This standard requires blind review of full papers by two suitably qualified independent referees. The submission of extended abstracts for review was on 1 December. We completed blind reviews of the extended abstracts and second round blind reviews of the completed papers in time for the publication of the proceedings on the Web on 15 March. This period of 3.5 months is perhaps the tightest publication timetable for a major research conference. In the view of the first author, who has to manage the process, too tight! Such a timetable could not be achieved with other than email and Web technologies. We review the media used to disseminate the proceedings in the next section.
As academics who have attended a large number of both professional and academic conferences, we came to the design of AusWeb with strongly negative feelings about the productivity of the traditional academic conference. We were determined to do better. Typically at an academic conference the papers can only be picked up at registration or, worse still, at the session or even worse still, not at all. Presenters cannot assume that members in the audience have read the paper or have a sufficient level of knowledge to make intelligent contributions to the discussion. This means that the presenter much spend most of her or his time outlining the paper. Given that the presenter has a strong commitment to the paper and considerable investment in the research, they often run over their allocated time. In so doing they remove the ability of the delegates to participate in a discussion. This is neither productive for the delegates -- they could have understood more by reading the paper prior to the session -- or the presenter -- who does not get good feedback on the paper. In designing AusWeb, we did not want to repeat the standard format of the research conference, where the chair of the session says "ah, we have time for just one question!" in the last few minutes of the session.
Consequently at AusWeb95 (Debreceny and Ellis 1995a; Vitartis et al. 1995), and less successfully at AusWeb96 and AusWeb97, we actively sought to change the discourse within the conference. The papers were made available on the Web a month before the conference, for the other delegates to read. At AusWeb99 the printed proceedings are to be couriered to participants a couple of weeks before the conference. Participants should have worked out which papers they're going to attend and, at worst, skimmed or, at best, read the papers. Having the papers in printed form allows the participants to read on the plane on the way to the conference or to make annotations that they can refer to in the conference session.
In designing AusWeb, we wanted to give the greatest ability of presenters and as many of the delegates as possible to interact in a lively and productive way. We wanted to give the delegates in each formal session the greatest opportunity to ask questions, make statements and to debate. Our measure of success was for the delegates to take over the debate informed by the contribution of the authors and by the paper itself. Indeed in some sessions delegates began directing questions to other delegates rather than to presenters.
As discussed above, this worked well at AusWeb95 and less well at AusWeb96 and 97. This can be attributed in part to the physical environment. AusWeb95 was held at the Ballina Beach Resort, which has an intimate scale which we believe fostered interaction. The enrolments at the conference were limited to 140, which is perhaps an ideal number for a conference of this size -- large enough to allow a range of delegates. Not too big to be impersonal. AusWeb96 and 97 were held at the Conrad Hotel. The facility was on a much larger scale, as were the number of participants. The conference arguably lost some of its intimacy and this had a negative effect on interaction. At AusWeb95, the moderator played an important role in binding the discussion together and raising issues.
In designing AusWeb99 we have reverted to many of the AusWeb95 features and enhanced several aspects. First, we have returned to the Ballina Beach Resort and limited conference attendance to AusWeb95 numbers. Second, we have marketed to authors from the day that they submitted their papers that AusWeb is different from the traditional conference. A number of email updates have been sent to authors throughout the review process, supplemented by appropriate pages on the conference Web site. The idea of AusWeb as a different style of conference is not an easy concept to sell. For researchers who are used to traditional conference presentations, it is difficult to get used to the idea that they have only five minutes to make their point. Nor is it easy for them to understand that the moderator has nearly as much time as they do! Third, we have introduced the concept of the online presentation. This is designed to substitute and indeed extend the traditional presenter up front conference presentation. We are also asking presenters to provide an online presentation, linked to the formal conference paper. It was envisaged that such presentations which might be (for example) a PowerPoint presentation using RealAudio's Real Presenter or a streaming audio or video address. It might be HTML .. it might be whatever the presenters' imagination runs to! Fourth, as discussed above, we are sending out the conference proceedings to delegates in plenty of time for them to read the papers before the conference, as well as placing all papers on the Web. Fifth, in seeking sponsorship for the conference we have not always chased dollars to assist the bottom line. To help promote and encourage the move to Web presentations linked to formal papers we have gained the support of Apple Australia who for AusWeb99 will award an iMac computer to the Web presentation judged as best. Both Apple and AusWeb "Think Different". Taken collectively, we hope that AusWeb99 will improve upon the AusWeb95 levels of Web-enhanced productivity in face-to-face interaction.
An important element of the conference is the keynote program. Choosing appropriate keynotes is a difficult task. The keynote must not only have stature and extensive knowledge. At the same time, it is important that the keynote program caters for the wide range of delegates that attend AusWeb (techies, educators, librarians, sociologists, media, business etc.). Of course, not every keynote will appeal to the complete range of attendees. Our biggest problem has been the alignment of expectations with what is delivered by the keynote. We enter into extended discussions with either the keynote or the keynote's sponsors to ensure that there is a common understanding of content and style. Keynotes are briefed on the AusWeb audience and negotiations are conducted on the broad details of the presentation. Negotiations with several potential keynotes have not been successfully completed over the years, because of our concerns that they might have excessive marketing in their address or where we have not been convinced that they will tailor their address to the AusWeb audience.
When we designed AusWeb95, we decided that a first priority was to place the proceedings in full on the Web. The idea of having a Web conference without disseminating the results to the wider Web community was anathema. We also decided that we would not require password access to the site. Having quickly resolved this point, we then addressed whether there should be printed and/or CD-ROM versions. We felt that for the library marketplace and for authors to be able to show their non-Webbie managers, a printed version was necessary. Further, the Web cannot -- at least up until to 1999 -- readily be carried around in the conference venue. Again, we quickly decided that a printed set of proceedings would be necessary. A more difficult question was that of the CD-ROM. The second author had had experience in producing a sampler of CD-ROM based courseware and we knew the issues involved in preparing and burning a dual platform CD-ROM. After a little thought and debate (and more red wine) we pushed ahead with a CD-ROM. At the same time, we decided that the CD-ROM would simply be used to store the HTML files of the proceedings and not some form of reformatted text.
The "author once/publish in multiple formats" model immediately set boundaries on the scope of submission. Simpler is better. We devised a new referencing format, the [HREF] reference, for printed representation of URLs. We set very tight requirements for the preparation of the HTML including standard layouts. We required use of relative rather than absolute links, so the files would work satisfactorily on the CD-ROM. We did not allow the use of CGI scripts, movies, sounds or, later, Java or JavaScript, DHTML, Shockwave or other enhanced Web functionality. All papers had to be submitted to or our server, for quality assurance and long term storage.
Very little has changed since AusWeb95 in the submission requirements. For AusWeb99 we introduced use of a W3C cascading style sheet [HREF11] to allow us to control the on-screen presentation, using fonts (Ventana) that have been designed for readability on the screen, and different fonts (Times) for the printed presentation. We required use of Dublin Core [HREF12] meta tags. We intend to re-engineer all the AusWeb papers into a Dublin Core compliant knowledge base in due course.
At AusWeb95 we had few problems with the preparation of the HTML. Authors who were researching the Web at that time were typically conversant with HTML. The spread of editing packages over the last five years such as Dreamweaver and FrontPage has not only separated casual authors from the underlying HTML code, it has also separated a minority of AusWeb authors from an understanding of issues such as the appropriate use of headings, preparation of metatags, inclusion of cascading style sheets and removal of hard-coded tags such as <FONT FACE="Arial"></FONT>. Clearly more work will be required to prepare authors for the experience of writing AusWeb papers, from a technical perspective. Given our policy of "simple is good", none of what we ask authors to do is particularly complex but it is clearly beyond the experience of some of our authors.
In moving the conference forward, there are several issues that the AusWeb team has considered, in varying degrees of depth. These include, the nature of the review process, the relationship with the wider Web research community beyond those directly involved in the conference, the role of papers and posters of differing stages of development, and the use of the Web for day-to-day management of the conference. We consider each of these issues in turn.
The review process for papers at AusWeb has followed traditional scholarly patterns. We have a review panel made up of researchers and practitioners who have demonstrated their expertise by making valuable research contributions, particularly by ongoing publication in the AusWeb and international WWW conference series. Two single-blind reviews are sought on both the abstract and final paper. Given the nature of the papers that the conference receives, adopting a double-blind review process would be unnecessary. In most cases the hyperlinks in the paper immediately identifies the research team. It is likely that we will continue with this review procedure in future as the core of the rejection/rework/acceptance decision. The single-blind review process provides high quality, unbiased guidance to the AusWeb team. We have not, however, used the Web to provide a wider source of review comments and criticism to guide the authors. A possible way forward is to allow all contributors to critique each other's papers. This would build knowledge of the authors of each other's contribution and to build a collaborative spirit between the authors. It would have to be recognised by all those participating that not all the papers would pass through the review process and be accepted for presentation.
Once a paper has been accepted, it is placed on the conference Web server. We do not provide a forum or other forms of interaction with the wider Web, other than by email to the authors. It may be that we need to consider the way in which authors, conference delegates and others interact with the rest of the Web community.
We have not used the Web to manage the day-to-day running of the actual conference. Our view has been that the conference was sufficiently small to allow natural communication patterns to work, without needing to facilitate these patterns with Web technology. At larger Web conferences such as WWW6 [HREF13] systems such as ICE (Interactive Conference Environment) have been tried but with mixed success.
Another issue that the AusWeb team will have to address is the management of the continuum between poster, short paper and long paper. AusWeb has had a single submission point only, as all extended abstracts are submitted for final presentation as full papers. For AusWeb99, we have accepted some papers as "short" papers, where we have felt that they would be best presented in a poster-like environment. Our view has been that the very short deadlines we have for the conference provides sufficient flexibility for authors to present new and interesting ideas. This may not be a view that we can sustain much longer. The significant demand for the restricted number of presentation slots at the conference means that both the reviewers and the AusWeb team are seeking contributions on research that is both well thought-out and completed. Such a model precludes presentation of papers which is floating new research ideas for feedback or work-in-progress. This is not desirable as the posters should be a predictor to the vital research to be presented at the next conference. A better view of the continuum is perhaps to see the poster as the exciting new idea that researchers are floating for feedback and guidance. The short paper is to report work-in-progress or smaller elements of a larger project. The full paper is to report thoroughly researched issues or problems. We address possible success factors for each model.
For a poster track at AusWeb to be successful, it must not be seen as of equal standing with the main paper track. A new title for the track is probably necessary (Emerging Research and Applications (ERA)? New, Emerging and Work-in-Progress (NEW)?). A separate call would also be necessary to distinguish between the NEW and paper streams. Much shorter submission deadlines would be required. A possible model is to dispense with formal reviews for the NEW stream. Everyone who chooses to, is slotted in for a NEW presentation during the early part of the conference. Delegates would then vote on which poster sessions they want to hear more about. These are then upgraded and presented as informal papers later in the conference. The best NEW paper then becomes the keynote at the end of the conference. Of course, this presents some technical challenges, to tally and display votes automatically and then to sequence the presentations and for authors to be ready to deliver more details should their presentation attract the required level of interest. A system such as ICE would be necessary. The enhanced functionality required would make this an Intelligent Interactive Conferencing Environment (ICE2)! Would any computer science faculty member like a nICE project for their senior students?
Under this model, short papers would still be reviewed, but may have a single pass through the review process. Short papers might need to be submitted almost ready for presentation. Abstracts for short papers would not be accepted. In contrast, long papers might well continue to have a two pass review process (once at the extended abstract stage and again at the full paper stage). Under this model, a longer review and paper development period might be necessary.
All of these considerations must also be matched against the very practical consideration of the limited number of places at the conference. Whether there is room for a poster/short paper/long paper model in a conference with registrations limited to no more than 150 delegates is an open, but interesting, question. Again, your views will guide us in our decision making.
In this paper we have analysed the design logic employed in AusWeb over the last five conferences. We reviewed the logic for conducting a same-place/same-time research conference about the Web. We reviewed the use of the Web to enhance productivity in the review and presentation process. We discussed the preparation of papers in paper, CD-ROM and Web format. We reviewed some possible future directions for the conference.
Finally, there is the question of the future of this series of conferences. Our view was that the market for a strictly Web focussed conference would have a limited life. Web technology would quickly become integrated into the multitude of traditional discipline-based conferences. The demand for AusWeb99 has proven this supposition completely incorrect. More than 100 extended abstracts were submitted from Australia and a dozen other countries. More than 50 papers will be presented at the conference and the conference itself was sold out more than two months before the event. There is clearly a demand for a conference with a dedicated focus on Web technologies, applications and implications. The challenge will be to continually invigorate the conference design and management to ensure that AusWeb meets the needs of the Australian and international Web research community.
Readers of this paper are encouraged to contribute their views on the future development of this conference by email to the authors or by completing our feedback form [HREF14].
Debreceny, R., & Ellis, A. (1995a). The Challenge of Integrating World Wide Web Technologies into the Delivery of Education, Governmental and Business Services. In R. Debreceny & A. Ellis (Eds.), Innovation and Diversity - The World Wide Web in Australia. AusWeb95 - Proceedings of the First Australian WorldWideWeb Conference [HREF15] (pp. 1-7). Lismore, NSW: Norsearch.
Debreceny, R., & Ellis, A. (Eds.). (1995b). Innovation and Diversity - The World Wide Web in Australia. AusWeb95 - Proceedings of the First Australian WorldWideWeb Conference. Lismore, NSW: Norsearch.
Heck, M. (1998). Meet me on the Internet. Infoworld, 20(23), 74.
Odlyzko, A. M. (1995). Tragic Loss or Good Riddance - the Impending Demise of Traditional Scholarly Journals [HREF16]. International Journal of Human-Computer Studies, 42(1), 71-122.
Ojala, M. (1999). We've got to stop meeting like this: Conferences online. Online, 23(1), 80-82.
Varian, H. (1997). The Future of Electronic Journals [HREF17] Working Paper Andrew W. Mellon Foundation and Emory University.
Vitartas, P., Debreceny, R., Ellis, A., & Mulholland, J. (1995). Conference Tourism: a motivational based typology. In E. Sogar (Ed.), International Conference of Marketing Educators and Researchers (pp. 345-360). Gold Coast: Griffith University [HREF18].
Roger Debreceny and Allan Ellis, © 1999. The authors assign 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.
[ Proceedings ]