Welcome to WWW7!
This is one of the most "distributed" WWW conferences yet to be held. With the Programme Committee and Review Committee dispersed over Europe, the Americas and Asia in addition to Australia, selecting this programme for you has been a challenging and interesting experience. Everything has been done electronically submissions, reviewing, notification, final versions and correspondence. In almost every sense, we have had virtual programme committee. Because of their geographic dispersion, the entire committee will not have met in the same room together until the conference itself!
It is not only the Programme Committee that come from many countries. We also received paper submissions from nearly every continent. In particular we are pleased to be able to offer you a selection of papers from Asia. Australia has been increasingly becoming part of an Asian economic community and holding the conference here has made it more accessible to the Australasian Web community. It has also increased the Web community's awareness of internationalisation issues. In the past, we have been concerned with research to make documents more accessible to people who may not speak English, but the Asian community serves to remind us that some people not only have a different language but a different script and alphabet as well. These pose additional challenges in rendering and translating documents, as well as searching and querying.
In selecting the work to appear in these proceedings, we followed a double-blind refereeing process. That is, the identity and affiliations of authors were kept separate from the actual work under review. The reviewers did not, in most cases, know who wrote the papers they refereed, and the programme committee likewise made its selections without knowing the authors' details.
One of the most remarkable results side effects of the double-blind review process is that it accurately reveals some thought-provoking trends. It was only when we began notifying authors of the results of their papers that we discovered that well over one third of the full papers we had selected came from authors in industry; from Hewlett-Packard, NEC, Microsoft, Digital Systems and in particular IBM. This strong showing of industrial competence suggests that the development of the Web is increasingly consumer-driven perhaps this is one of the few research fields whose agenda is determined by the urgent requirements of the end user. In response to this, industry is putting substantial resources into the development of the Web. And this investment of resources is paying off academically as well as commercially the work being done has been independently recognised by an academic review team as leading-edge research and development.
What does this mean for academic institutions, especially in times of tightened government funding? Is it possible for a university to compete with a well-resourced company? In one respect, the answer is no, they probably cannot compete directly with commercial interests. However, it is worth noting that industrial interests are perhaps more influenced by economic rationalism where research investment is often expected to show foreseeable commercial returns. The more long-term research, "blue-sky" research where the risks are higher and benefits are not so readily apparent, is still the rightful domain of non-commercial research institutions. The very concept of hypertext was originally the thought experiment of a government scientific advisor and researcher, and while the earliest instantiations of hypertext were prototyped in a commercial research institution, it was a physics research institute that finally brought us the Web as it is today.
It was not only in the final papers that we found interesting trends but in the complete set of submitted papers. We analysed the number of papers submitted for each of the topic areas. About 20% of the 218 papers submitted as full papers were tagged by their authors as being in the area of Information Retrieval and Modeling, and 17% in the area of Search and Indexing Techniques, although these were but 2 amongst over 35 areas*. This is nearly double that of the next largest areas/topics for which papers were submitted: namely, Browsers & Tools, HCI, and Education, each with approximately 11%. Metadata, Markup Languages, and Hypertext and Hypermedia came next with about 9% each.
Among the areas/topics with least submissions were: Mobile Code, Security and Authorisation, and Charging and Payment, each with 2%.
These percentages warrant some comment. It is worth stating that there is no apparent correlation between these percentages and the order of topics/areas in the Call for Papers. Nor can these percentages be due to any biases of the Programme Committee or the Reviewers these percentages relate to the number of submissions, not acceptances.
Can they then be said to simply reflect the research interests, activities, and progress of the WWW research community? To a large extent, the answer must surely be "Yes". But with qualifications.
The concern with resource discovery on the Internet pre-dated the World Wide Web, being a central issue for Gopher, WAIS, etc. The advent of the Web saw large WWW-based public document indices and search engines being developed, which provided capabilities for locating potentially relevant documents based not solely on document identifier or location, but especially on document content. It is clear, however, that the techniques used by these search engines typically suffer from poor precision, and often untimely and erratic coverage, and while they currently boast impressive performances, the techniques employed may not scale well, either in terms of volume of material to be indexed, or in terms of future query loads. Of additional concern is the behaviour of the Web spiders and crawlers that feed these search engines with new or modified Web documents; in particular, the resource-intensiveness of the "pull" techniques, the erratic and slow cycle times, and so on.
However, it must be remembered that the area of Information Retrieval (IR) is one of the oldest in the history of modern computing, and that IR research has a broad and deep body of results. While it is not especially surprising that WWW-related search and IR are receiving such research attention, one of the reasons that this attention has resulted in such a significant contribution to the papers submitted to WWW7 is no doubt the maturity of IR itself current Web-related search issues have a substantial body of extant research on which to build.
The differences between those areas that had 8-12% representation in submitted papers for example, browsers, metadata, markup languages, hypertext and hypermedia are insignificant, and all these areas/topics are reasonably well represented amongst submissions to WWW7.
Just as the areas/topics of Search and IR were noteworthy for the percentage of submitted papers claiming to address these issues, so too for the areas of Mobile Code, Security and Authorisation, and Charging and Payment but in this case, it is the fact that the percentages are so small. There are a significant number of research groups and projects that are focussed on one or more of these three areas so it seems unlikely that the small number of submissions to WWW7 in these areas is due to lack of effort in the research community. Could it instead be that the research problems being faced is these areas, regarding the Web, are not merely difficult, but truly novel?
To those of you who submitted your work for our programme, we extend our warmest thanks. The success of the conference will be measured partly by your contributions. We want to thank our reviewers for their hard work and careful attention to ensuring the quality of the programme. We also want to give especial thanks to the programme committee for their wonderful work in selecting the tutorials, workshops, special interest groups and panels. Finally, none of this would have been possible without the hard work of the Local Organising Committee whose efforts cover everything from the logistics to the entertainment. All of these people are listed in these proceedings, and we hope you will join us in our thanks for their contribution toward making this conference a success.
Now it is time for us to hand over to you, our delegates. We wish you all the best with your research and development and hope that the work in these proceedings helps you toward your goals.
*Authors tagged their papers with up to three tags; consequently, their was some overlap between the papers tagged with these two tags.