Difference between revisions of "Tribute to Bill"
Line 51: | Line 51: | ||
My best work, not my most cited work was a chapter in my dissertation. Yeah, I bloomed early. | My best work, not my most cited work was a chapter in my dissertation. Yeah, I bloomed early. | ||
− | < | + | Summary:<br> |
− | + | Goof ball – 17 citations<br> | |
− | Goof ball – 17 citations | ||
My best work – 12 citations | My best work – 12 citations | ||
− | |||
It went poorly for me. We laughed about this for years. Literally, not a beer session would go by without shared jabs on this topic. I will miss the running gag. | It went poorly for me. We laughed about this for years. Literally, not a beer session would go by without shared jabs on this topic. I will miss the running gag. |
Revision as of 14:03, 9 January 2020
Bill was a major influence in the areas of computer graphics, visualization, medical computing, and software process. He was loved by many throughout these communities. Here we have captured some of the many tributes contributed by friends and family.
Contents
Obituary
Bill grew up near Troy, NY where this local obituary was posted.
VTK Discourse List
Bill was a significant contributor to the VTK software. He was one of the three original authors of the Visualization Toolkit textbook including the companion software. Bill contributed to VTK until the last weeks of his life, in later years contributing an extensive compendium of examples and code. He was also a driving force in converting the original book into Markdown and LaTeX (with Andrew Maclean and other community members). Bill could always be counted on to provide praise and nag scofflaws who were slow to add software tests, or correct errors on the VTK dashboard.
David Banks
I met Bill when he visited UNC in the early 1990s. He gave a talk about Marching Cubes and Dividing Cubes, and was tremendously affable throughout his visit. Although his work in visualizing 3D MRI was interesting, I didn't consider it particularly relevant to my own efforts to display and manipulate mathematical surfaces in R4. More than a year passed before I realized that surfaces in R4 arise as level sets (Andrew Hanson at Indiana was simultaneously thnking along the same lines). I worked with Chris Weigle to develop a 4-dimensional version [1] of Marching Cubes, and then worked with Kevin Beason to compute global illumination on heightfields as a pre-process [2] before running Marching Cubes. Kevin demonstrated the results on 3D MRI, bringing the work full-circle back to Bill's original application-domain. Paul Stockmeyer (at William and Mary) visited my lab to work out the group-theory details underpinning Marching Cubes and its variants [3]. I ended up being very influenced by Bill and his informal presentation at UNC Graphics Lunch. It was an honor to meet him and discover his work.
[1] Weigle, Chris, and David C. Banks. "Complex-valued contour meshing." In Proceedings of Seventh Annual IEEE Visualization'96, pp. 173-180. IEEE, 1996.
[2] Banks, David C., and Kevin Beason. "Decoupling illumination from isosurface generation using 4D light transport." IEEE transactions on visualization and computer graphics 15, no. 6 (2009): 1595-1602.
[3] Banks, David C., Stephen A. Linton, and Paul K. Stockmeyer. "Counting cases in substitope algorithms." IEEE Transactions on Visualization and Computer Graphics 10, no. 4 (2004): 371-384.
Greg Jones - The Goof Ball Story
Bill and I got to spend some time over the years, one night during a DARPA meet up we began discussing earlier pieces of our careers – high and low. Bill lit into a rendition of his work in golf, a passion of his. Turns out GE Research created a package to map the path of a putt. This involved scanning and modeling golf greens and then doing a solve for the likely path of a putt. They published a paper on this work; one of Bill’s lowest cited papers as it turns out.
GE made a deal with CBS for the Masters’ Tournament to show a predicted putt trajectory before the golfer actually made the putt. Of course, the initial conditions of a putt can be varied and GE’s initial conditions differed significantly from the golfers’ initial conditions. Pro golfers hit the ball quite a bit harder than the GE team assumed. The projected trajectories where not similar to the actual trajectories and the Wall Street Journal review mentioned such with a reference to GE’s “goof ball” system.
So, some of Bill’s work earned the name “goof ball” from a major media outlet. At this point of the story we were 2-3 beers into the evening and I was enjoying the great Bill Lorensen’s humbling story and might have mentioned to Bill how much I enjoyed the humbling. Further, I might have asked Bill to give me an estimate of how much marching cubes did to offset the impact of goof ball on the world of science.
Bill retorted along the lines of, “Let’s compare citations. Your best work to my goof ball work.” Yep, that's right Bill had an edge to him, albeit a funny edge. He could give as well as take when it came to heckling.
My best work, not my most cited work was a chapter in my dissertation. Yeah, I bloomed early.
Summary:
Goof ball – 17 citations
My best work – 12 citations
It went poorly for me. We laughed about this for years. Literally, not a beer session would go by without shared jabs on this topic. I will miss the running gag.
Sandy Napel
I think I first met Bill in the late 1980s. I was a visiting assistant professor at the University of Western Ontario, and I attended, with my colleague, Dr. Brian Rutt, a GE Research Meeting at Snowbird, Utah that brought together people using Bill’s (and others I am sure) “MRConsole software.” It was highly advanced for its time, ran on affordable SUN Microsystems workstations, and facilitated the visualization of volumetric scans. It was also ahead of its time in that the source code was shared freely among GE’s users, predating the concept of open source software so important and powerful in today’s world. When I got to Stanford in 1991 I was still actively using it to develop advanced visualization methods for CT. Many of my earlier papers were supported by results obtained using this software.
But it was more than just software. Bill was a wonderful colleague, always open to new ideas, always willing to make changes to the code. And always fun to be around. At that same meeting in Utah I recall sitting in the bar with Bill and Ron Kikinis and discussing this new idea for distributing software called java and java beans. I thought they were insane, but they were really insanely great, well before anyone else used that term.
After those early years we only crossed paths at meetings, during a sabbatical he spent at Stanford with Marc Levoy, and more recently at advisory board meetings for Ron’s P41. It was always a pleasure reconnect, to discuss science and technology, and to share a good time over a beer. He was a great technology innovator and an all around great guy. I miss him already.
Hanspeter Pfister
I have known Bill since I attended my first IEEE Visualization conference in 1992. At the time I was a first year PhD student in Arie’s lab and did not know much about visualization. Arie introduced me to Bill by saying something like “this is the guy who invented marching cubes.” We exchanged some pleasantries, and later that day I had to look up what marching cubes is.
Bill was one of the inspirations for my research in point-based graphics. We used to talk about the GE hardware they built for dividing cubes, which we now would call a point-based rendering algorithm for volume data. Bill was very proud of dividing cubes, and he showed a lot of interest in our work in point-based graphics. As always, he was encouraging and enthusiastic about our work, which in turn inspired us to continue.
Over the years, we participated in several workshops and panels on volume graphics. Bill’s usual role was to defend marching cubes and polygon rendering, and my role was to present ray casting as the solution to all volume rendering problems. Taking these extreme positions made the panels especially fun, even though our arguments were mostly tongue in cheek and not really serious.
In 2000, I organized the “Transfer Function Bake-Off” panel, and invited Bill to be the referee to choose “the best” method for transfer function selection in volume rendering. Bill made a show of it, and selected Gordon Kindlman as the winner. The prize was an easy-bake oven, which, many years later, I saw proudly displayed in Gordon’t office.
Bill and I interacted a lot during my tenure as the VGTC chair. I invited him to be the chair of the VGTC technical awards committee in 2004, which gave me the opportunity to have lunch with him and the awardees at several VIS conferences. Bill handled this important position with class and dignity, and he and the committee made excellent selections for the awards until he stepped down in 2011. More about this in the email snippets below.
Later I had the pleasure to be part of the ITK-4.0 project, sponsored by NIH and led by Terry. It was fun to see Bill’s occasional technical emails fly by, mostly corralling people to get behind his software architecture and coding standards. Bill was a programmer’s programmer, and the depth of his programming knowledge was infinite.
Needless to say, Bill has been a role model for me and many other young researchers. His easy-going manner, his good humor, his love of visualization and geeking out over a beer made everyone feel at ease around him. He is a person who achieved so much without letting it get to his head.
He will always be in my heart, and I truly miss him.
I found some old pictures with Bill from IEEE Visualization 2002 (Boston) and 2004 (Austin, TX)
Theresa-Marie Rhyne
It was 1990 when I first met Bill Lorensen. He was co-author of a forthcoming book on "Object-oriented modeling and design" and was involved with the 1990 Workshop on Volume Rendering. It was a time of large workstations where Silicon Graphics (SGI) and Sun Microsystems (Sun) were the leaders. I was a Unisys employee responsible for running the U.S. Environmental Protection Agency (EPA) 's Scientific Visualization Center. I needed help on the appropriate workstations to purchase so that EPA's visualization activities could grow beyond GIS modeling, often performed on Data General workstations, into 3D time series animations. He advised me to go with both SGI and Sun workstations along with accurately predicting that we would eventually need a powerful SGI workstation to keep the center running. The SGI Oynx workstations would be introduced in 1993. Bill knew I was a newbie when it came to workstations so he gave me enough coaching to help me appear to have some hardware knowledge of the situation. I had to stand up to specialized consultants in supercomputing who knew far more about high performance computing, but not necessarily computer graphics, than I did. Bill helped me do that. He was a master of both technology and kindness. He encouraged me rather than making me feel foolish as I evolved into a technical leader. I will always remember his thoughtfulness in that regard.