by Scott Tilley

The IBM Centre for Advanced Studies (CAS) holds its annual conference every autumn in Toronto. This year was somewhat special for me: I had not attended a CASCON since 1996, and it's the tenth anniversary of my first participation in this distinctly Canadian event in 1991. Although attendance at CASCON 2001 was somewhat reduced (like all meetings these days), I thoroughly enjoyed my time there. For a country as large as Canada, CASCON offers the unique opportunity for everyone involved in academic software research to gather in one place and at one time to network, to discuss innovative new results, and to get a glimpse into what the world's largest software company is working on for today and tomorrow.

CASCON is the IBM Centre for Advanced Studies annual technology showcase. As predictable as Canadian geese flying south for the winter, autumn in Toronto has come to mean attending CASCON for most academics working in software systems research at Canadian universities. I first attended CASCON in 1991, shortly after its inauguration. As a CAS-sponsored Ph.D. student at the time, I found CASCON to be a very worthwhile event. In my current position as a faculty member in another country, my fundamental appreciation of IBM (and the NRC) for its sponsorship of this unique event has only increased.

From the CASCON Web page:

CASCON is the premier technology research conference in Canada, sponsored by the IBM Centre for Advanced Studies and the National Research Council of Canada. The conference offers an exciting mix of speakers, workshops, tutorials, hands-on courses, and research papers, as well as a technology showcase. Major themes for this year's conference include database technology, software engineering practices, distributed and Web-based systems, compiler technology, and electronic commerce.

CASCON 2001 was held November 5-7 at the International Plaza Hotel in Toronto. The first few years of CASCON were held at the Ontario Science Center, with participants staying at the luxurious "Inn on the Park" hotel. For the last few years, CASCON has been held closer to the Toronto airport -- a local that is easier to reach,  but more difficult to escape from. With the timeless theme of "Meeting of Minds," CASCON managed to attract 1,400 hardy souls for what amounted to a very fulfilling few days of information exchange.

As an added bonus for academics, attendance at CASCON is free; IBM/NRC foots the entire bill. As someone who has organized three conferences this year, I have an inkling of the cost, time, and effort that goes into orchestrating something as large as CASCON. Hats off to the volunteers!

[back to the top]

The Conference

CASCON consists of a mix of refereed paper presentations, workshops, keynote presentations, demonstrations, and lots of networking. The keynotes of this year's conference were Marisa Viveros of IBM Pervasive Computing Solutions, Walter Bender (Executive Director) of the MIT Media Laboratory, Joan Mitchell (IBM Fellow) speaking on the importance of mentoring, and Aristides Requicha (Director of the Laboratory for Molecular Robotics) at USC. Unfortunately, I was unable to attend any of these presentations due to tight travel restrictions. However, other people at the conference told me they liked the talks, especially Walter Benders' description of ongoing research at the MIT Media Lab.

One of the highlights for me at CASCON is the demo area. I know that IBM goes to great lengths to provide communication and networking infrastructure for all demonstration stations. The demos aren't in the same commercial category as say COMDEX, but they are far superior to most academic sessions that I've seen (and participated in) at other conferences. CASCON 2001 had over 100 demonstrations of leading-edge research, many from graduate students funded by CAS.

In addition to research paper presentations, CASCON features several half-day and full-day workshops. This year there were workshops on topics such as Web services, bioinformatics research in proteomics, and developing Java applications using Eclipse. In fact, my main reason for attending CASCON this year was to participate in two workshops, which I consider to be the highlights of the program.

[back to the top]

Program Highlights

One workshop of note was "Enterprise Integration," led by Dennis Smith of Carnegie Mellon's Software Engineering Institute. The workshop was a first in a series of gatherings on the topic on integrating enterprise legacy systems to support critical business processes in the era of net-centric computing. Dennis provided an introduction to the area and outlined the project's parameters. This presentation was followed by specialized talks on data integration by Peter Aiken, enterprise portals by Atul Bhatt, quality attributes by Mario Barbaci, control integration by Kostas Kontogiannis, and enterprise application integration using loosely coupled Web services by Arthur Ryman.

Another workshop that was near and dear to my heart was "Adoption Centric Tool Development," led by Hausi Müller of the University of Victoria. I gave the first talk of the workshop, entitled "Why Research Tools Remain Lab Orphans." This talk was a pleasure to give because the theme is something that I have long pondered. This is the  talk's abstract:

Despite the best of intentions, research tools often remain lab orphans, rarely adopted by industrial partners. There are many reasons for this situation, ranging from standard software engineering issues such a poor requirements management, to more specific issues such as a misunderstanding of the commercial environment and the professional programmer. To successfully deploy a research tool into an industrial setting, one must never forget that industry is not interested in tools; they are interested in solutions. The distinction is critical and can mean the difference between a fruitful long-term partnership and a one-off project that leaves both parties with a sour taste in their mouth.

I trust that some people in the audience were able to understand exactly what I meant. I think there is still an important role to be played by academics in software systems research, but that role needs to be clarified and reevaluated in light of current realities. Given time, I'll develop this talk into a full paper.

[back to the top]

Final Comments and Photos

As always, I captured some of the special moments with my  trusty Canon Powershot S100 Digital Elph Also as always, several of the pictures are a little fuzzy, due to the low lighting in the rooms. I was hesitant to use my flash, because I forgot to bring my battery charger on this long trip (see below). Doh!

The next CASON will be held in 2002 in Toronto (as usual).  If you are based in Canada, attendance at CASCON should be considered de rigueur. If at all possible, I strongly suggest you avail yourselves of this one-of-a-kind event. The weather may be cold (from a Southern California perspective), but the hospitality is warm.

	Welcome to CASCON 2001! This sign was hanging just outside the main demo area. In such  large hall, the flash failed to illuminate the area very well. (70KB)
	Participants in the "Enterprise Integration" workshop. It was near the end of the day, so everyone is looking a little haggard. But the Japanese dinner afterwards was great! (94KB)
	Jens Janke and Ken Wong, two Apple enthusiasts, working on material for the next day while relaxing in the hotel bar. They had an ad-hoc WiFi peer-to-peer network going. (Photo courtesy of Hausi Müller.) (335KB)
	The poster from the University of Victoria's Rigi group in the CASCON demonstration area. This is my Alma Mater. (105KB)
	Yours truly, giving the talk "Why Research Tools Remain Lab Orphans" in the "Adoption Centric Tool Design" workshop. I think at this point I was ad-libbing because some computer projection problems were still being resolved. I don't know if the audience could tell the difference. ☺ (86KB)

PS: To give you an idea of the truly global world we live in ... this article concerns a conference in Toronto, Canada. I wrote most of the piece on the plane en route from Toronto to Florence, Italy for WSE 2001. The final edits were done in Munich, Germany while I was visiting BMW. The article was subsequently published to the Web from Montreal, Canada on the way back home to California.