3 ms·
My one email exchange with Brooks had nothing to do with Mythical Man Month. In the 1990s I was was the junior co-founder and, for a while, main developer of V
by dalke 4y ago
My one email exchange with Brooks had nothing to do with Mythical Man Month.
In the 1990s I was was the junior co-founder and, for a while, main developer of VMD, a program for molecular visualization. I wanted to include molecular surface visualization, but me being me, would rather integrate someone else's good work.
I looked around and found "Surf", a molecular surface solver written by Amitabh Varshney when he was at the University of North Carolina. (See "Computing Smooth Molecular Surfaces", IEEE Computer Graphics and Applications, https://ieeexplore.ieee.org/abstract/document/310720 https://ieeexplore.ieee.org/abstract/document/310720 .)
Brooks, you may not know, heard Sutherland talk about using the screen as a window into another world, which got Brooks interested in VR. Back in the 1970s, at UNC, they started experimenting with head-mounted displays. Brooks worked on VR for the rest of his career.
The UNC VR group worked on many different VR approaches, including haptic (tactile) feedback. As I recall, the first was a used hydraulic-powered robot arm. People had to wear a lab coat and helmet when using it because it would leak, and had a tendency to hit people.
One of the experiments, the NanoManipulator, hooked up the VR and haptic feedback (not that same robot!) to an atomic-force microscope, so people could feel the surface and move nanoscale objects around. http://www.warrenrobinett.com/nano/ http://www.warrenrobinett.com/nano/ .
Brooks felt that VR would be very useful for molecular visualization, and developed the GRIP Molecular Graphics Resource. Quoting https://apps.dtic.mil/sti/pdfs/ADA236598.pdf https://apps.dtic.mil/sti/pdfs/ADA236598.pdf , some of its early achievements were "the first molecular graphics system on which a protein was solved without a physical model", "using remote manipulator technology to enable users to feel molecular forces", and "Real-time, user-steered volume visualization of an electron density map".
As that document points out, their goal was to "wildcat radical new molecular graphics ideas to the prototype stage. Winning ideas are spun off to the thriving commercial industry or into autonomous research projects."
Surf fit very well in those lines, as VMD was an "autonomous research project".
My exchange with Brooks and UNC was 1) to get permission to distribute Surf as part of the VMD distribution, and, 2) a few years later, to provide numbers about how many people had downloaded VMD with Surf.
- parag_chandra 4y agoYes, the NanoManipulator! I remember getting a demo of it during the grad student “research assistant job fair” when I first started there over 20 years ago. It was like magic, and what got me excited to study computer graphics. Unfortunately after taking the intro class (taught by Brooks himself) I realized I wasn’t cut out for all the complex math involved, so I switched to medical imaging instead. Still, Dr. Brooks was a great lecturer who injected a lot of his own personal experiences into his classes, and I ended up taking his computer architecture class later on.
- gregcoombe 4y agoSide story: the haptic feedback for the NanoManipulator was through a hydraulic system (kinda like this? https://www.sarcos.com/wp-content/uploads/history_5-339x280.jpg https://www.sarcos.com/wp-content/uploads/history_5-339x280....). There were hydraulic lines that were piped through the building down to the machine room (where the SGI Infinite Reality Engine was!). Someone read through the manual and realized that the force that the arm was capable of could easily break someone's arm, and since it was usually grad students working late at night programming it, they decided it would be safest to just decommission that. I think I got one of the last demos during a UNC grad school recruiting event.
- dalke 4y agoI may be mixing up my robot arms! I visited the lab around 1995 and saw a smaller robot arm than the one shown in Figure 2 of https://dl.acm.org/doi/pdf/10.1145/166117.166133 https://dl.acm.org/doi/pdf/10.1145/166117.166133 . The big arm from Figure 2 is "an Argonne III Remote Manipulator". Oddly, I can find no mention of that ARM outside of its use for the NanoManipulator. I did find https://www.ks.uiuc.edu/History/VMD/ https://www.ks.uiuc.edu/History/VMD/ (my old haunting grounds!) say: > Computer scientist Frederick Brooks describes his chance encounter with the man who designed the manipulator as providential. In the 1950s, at Argonne National Laboratory near Chicago, Raymond Goertz and his group developed the ARM, the Argonne Remote Manipulator, a force-feedback device used to manipulate radioactive material in contaminated areas unsafe for humans to enter. Users gripped a device and moved it with their hand, and then signals were transferred to a robotic hand inside the contaminated area, which the users could see through glass. In the late 1960s or early 1970s Brooks met Goertz, the primary developer of the ARM, and Goertz arranged for Brooks to receive a manipulator that was no longer in use. ... > While trying to use the donated remote manipulator with a computer in the 1970s, Brooks realized that he needed at least a hundred times more computer power than was feasible at the time, and he sidelined his work with the ARM until 1986, the arrival of the VAX computer. ... Oooh! And you can see a few pictures of a young me in that UIUC link!