Interview: Alain Chesnais, President of ACM SIGGRAPH

email
28 July 2002

What first drew you to computer graphics?

In the early 80's, while on leave from the CNRS (Centre National de la Recherche Scientifique), I was working at the Centre Mondial Informatique et Resources Humaines that Nicholas Negroponte had founded in Paris. I was doing my research on queuing theory, focusing on aspects of parallel access to shared data. My work involved running huge simulations to experimentally validate mathematical models that I had developed to describe the behaviour of concurrent systems with locking capabilities. I was using a Vax 780 to run these simulations in batch at night.

I was approached by another CMIRH employee: Henri Gouraud, to see if I had any problems with letting a group art students from the Universite Paris VIII use the Vax to do CG at night. I had no idea how compute intensive CG was and agreed. I soon discovered just how easily a handful of art students could bring the Vax to its knees by running massive render jobs all night.

Needless to say, I was concerned about this and the impact it would have on my own research. I went back to Henri to indicate that I was having problems with the amount of extra load the students were putting on the Vax. I told him that I wanted to kick them off.

Henri suggested that given my research focus I might be able to do something about the performance of the software they were using and pushed me to look at optimizing the rendering phase instead. I agreed and got hooked by the possibilities that CG offered. I ended up asking my CNRS supervisor to shift research directions to focus on CG and was actually told to suggest something a bit more serious than making pretty pictures (CG was not quite considered to be a serious domain of research within certain parts of the computer science community at that time). For a while I continued doing my official research during the day and my CG during the night, trying to schedule some amount of sleep whenever possible in between.

In 1986 I decided to leave CMIRH to start up a company with one of the students who had worked on the Vax to do computer graphics. The company, Studio Base 2, did both software development and productions, focusing on high end rendering capabilities such as ray tracing, particle systems and interactive shader tuning tools. We were later acquired by Abvent and the Studio Base 2 code became the basis for Abvent's Art*lantis Render product line. I later left Abvent and joined TDI. TDI was acquired by Wavefront Technologies, which as later acquired by SGI and merged with Alias to form Alias|wavefront. At A|w I was director of operations for the Paris office until 1998, when I moved to Toronto to work on the core Maya architecture, then moved on to be Director of Engineering and Business Manager of Design Visualization.

I've since left A|w and am now V.P. of Engineering at TrueSpectra. developing dynamic image servers. One can think of dynamic image servers as being HTTP servers with a fully featured compositing and 2D rendering engine built in.

Do you have any favorite CG mentors?

See above: Henri Gouraud.

What was the first time you contributed to SIGGRAPH?

In 1987 we submitted a piece to the electronic theater: "Systeme Particulier" and another to the art show. Both were accepted.

What year/city was your first SIGGRAPH? Which was most intense? Why?

Though I registered for the 1986 SIGGRAPH conference, I ended up not going as we were behind schedule on a major production doing the special effects for a televised science fiction series in France called "Trakal". I finally went for the first time the next year in Anaheim. I was blown away by the whole SIGGRAPH experience and it remains one of my most intense SIGGRAPH's as I went around each venue with stars in my eyes, totally amazed by the amount of things going on. It pushed me to join, first the Paris chapter, then ACM SIGGRAPH itself and volunteer to help. 1997 was another intense SIGGRAPH conference for me as I was then vice chair of the organization and also chair of the conference's international committee. The experience of organizing the 1997 conference was magical in that the group of people who came together to be program chairs under the leadership of Scott Owen had a vision that we would radically change the shape and experience of the conference itself. Everything was open for change and innovation. Each program chair worked with the others to build a cohesive view of the total experience finding synergies amongst the venues and suggesting many new features. When I go back over the initial vision of what we wanted to do and look at what we actually did in an 18 month period, it gives me a definite sense of wonder to see just how much we actually accomplished.

What contributions to SIGGRAPH are you most proud of?

Helping Scott Lang to build his vision of the chapters. The chapters have come a very long way in the 15 years that I've been an active ACM SIGGRAPH volunteer, essentially due to the hard work and dedication that Scott has put into building the chapters into a functional group that presents computer graphics on an ongoing basis in close to 40 cities all over the world. I'm proud to have been able to help.

I'm also proud of having worked on the 1997 and the 2001 conference committees, each time on the international committee. I've strived to make the user experience for non American attendees better by adding new services directly targeted at them. In 1997 we added simultaneous translation in Japanese for the panels session. In 2001 we added a presentation area to the international center and invited speakers from various sister societies around the world to come and share with the SIGGRAPH community all of the exciting activities that they sponsor. We also added an English re-reading service to the submissions process to allow potential contributors for whom English may not be the mother tongue to have their submissions checked for grammatical errors before being submitted to the jury for consideration. This service has been extremely well received, especially by the Japanese community.

I'm also proud of all the work I did during my two terms as ACM SIGGRAPH vice chair from 1995 through 1999. The most important for me were the effort we put into strengthening the relationship between ACM SIGGRAPH and several sister societies outside the United States such as Eurographics and Nikograph (now called DCAj). We have evolved to a level of trust and cooperation that I believe will be instrumental in better serving the needs of the computer graphics community throughout the world.

I'm also proud to have been elected to the position of ACM SIGGRAPH president. Being the first European president of this organization sends a very strong message to the computer graphics community worldwide that ACM SIGGRAPH is a truly international organization.

What's your favorite thing at this year or last year's SIGGRAPH?

I loved the proximity of the Riverwalk and the ability to find real food within easy walking distance of the convention center. I couldn't walk more than several hundred feet without running into someone I knew and wanted to talk to.

I also was excited to see the web graphics venue take place this year. It's high time that we focus on delivery vehicles for CG other than film and video to understand the particular needs and requirements of each, particularly with respect to the web!

What near/intermediate developments in CG do you look forward to?

I'm personally interested in looking into ways to improve graphics delivery on the web. My research focus echoes this as I believe that we have a lot to discover in terms of how best to deliver graphical content efficiently on a hugely distributed architecture like the world wide web. Graphics account for what is probably the most costly and time consuming aspects of web site development. I believe that we can address that by looking at innovations like using dynamic image servers as a basis for web imagery and hope to see substantial progress in this area in upcoming years. I presented a paper in the web graphics venue this year covering this topic and plan to submit further work for next year's conference.