Well, this is a very personal conclusion. As I said previously, the Web community aggregates not only people from all over the world, but from different backgrounds, and this includes Web development, comp.scientists, high school teachers, physicists, engineers, physicians, psychologists, educators, publishers, etc. And this diversity is really what the Web is all about, this richness is what makes the Web so unique. But, if on one hand this is what it makes all so special, I believe that on the other hand it can also be extremely prejudicial if not handled with adequate care. We all want different needs from the Web a high school teacher wants it to deliver their classes; a researcher wants it to serve as an collaboration instrument; an Education Minister wants it to replace human-labour to save money; physicians want to be able to simulate a surgery, or perhaps assist a surgery being performed on the other side of the world. Ultimately, there are an infinite number of applications to the Web, and this makes creating these applications a very difficult job indeed. As I pointed previously, people seem to be ignoring previous research done into a topic when the Web is the medium. HCI is totally ignored when a browser is the GUI; database theory is discarded when the Web is the storage mechanism; psychology and pedagogy is not relevant when designing a teaching environment through the Web; work on CSCW is not considered when designing cooperative environment on the Web.
We must take care, and not ignore research previously done in different areas just because we are now talking about the Web. Of course we have to rethink some of them, but rethink does not mean discard. And this brings me to my final comment, which is related to inter-disciplinary research. Inter-disciplinary is by nature extremely difficult and not easily defined. So, I would like to briefly introduce a paradigm of disciplines that came to my mind when attending all those paper sessions and observing the lack of inter-discipline in such multi-discipline ventures. One could start by establishing what disciplines are involved in one's project; identify the exact role each discipline will have on the project, its interactions, and how and what would they communicate amongst themselves. Perhaps we could think as such projects as a collection of objects, where each object represents a discipline, and like Object Oriented technology we would define what classes they belong to and the methods we should create in order to make them accomplish their role within the project.