Roger E Critchlow Jr
I am very interested, and very effective, in using computers, computer graphics, and graphical user interfaces to facilitate the understanding of
mathematical, physical, chemical, biological, informational and business processes. I am currently working full time.
Although my formal training ended with a bachelors degree in chemistry, I am quite comfortable researching the scientific literature in many fields, including population biology, computer science, software engineering, machine learning, molecular biology, combinatorial chemistry, experimental design, computer graphics, agent based modeling, geometric algebras, and optimization. I can take original technical papers in a variety of disciplines and translate them directly into working code.
I have been successful in applying my programming and scientific skills for a great variety of clients. I have produced code for software engineers, computer scientists, chemists, biologists, computer game companies, physicists, mathematicians, logistics experts, and artists. I am able to listen to and talk with people about their desires, needs, and requirements. I can find and use the parts of the scientific literature applicable to their problems. I am able to produce a quality product that meets their specifications.
Most recently I've been working as an independent consultant. My work has included simulating fiesta crowds for the Santa Fe
Police Department, maintaining combinatorial chemistry software for the Chiron Corporation,
From 2000-2002 I worked for the BiosGroup in Santa Fe, NM, a consulting firm founded by Stuart Kauffman. My principal projects there were with Walgreens, United Airlines, the US Transportation Command, and Southwest Airlines. The Walgreens' project involved building, on very short notice, a Java simulation of their largest warehouse to validate its ability to handle the unprecedented peak loads expected in the fall of 2000. We found that the warehouse had insufficient capacity for restocking pick areas under peak conditions. With Tony Plate I developed and delivered a simulation of operations and maintenance for the United narrow body fleet in order to explore prioritization strategies for maintenance items under different airline operating conditions. I reviewed the existing tools at US Transportation Command for simulation of troop deployment scheduling for air and sea transport and persuaded them that incorporating more modern scheduling technologies was both possible and desirable. With Fred Seibel I analyzed the Southwest schedules for 2001 and 2002 with respect to maintainability and on time performance. We found that Southwest had some lessons to learn from the scheduling construction practices of its competitors.
In 1998-2000 I worked as an independent software contractor, develping Tcl/Tk based tools for clients in the pharmaceutical and electronic games industries. My collaborator Steve Dipaola has a very nice page describing FaceLift, a character face generator we developed for Maxis/EA as part of the release of The Sims.
In 1997-1998 I worked for Daylight Chemical Information Systems at their Research Office in Santa Fe, NM. I developed scripting language interfaces for their toolkits.
In 1995-1997 I worked with Jeff Blaney's computational group in Drug Discovery Research at Chiron Corporation, Emeryville, CA, first as a contractor and later as a full time employee. Working with Eric Martin, I automated a combinatorial synthesis design system using Tcl/Tk and other tools. This system is used to help chemists choose maximally diverse sets of reagents, given the constraints of the chemistry involved, for use in combinatorial syntheses, where hundreds, thousands, or ten-thousands of compounds are produced in parallel. The automation of the tool reduced the elapsed time for designs from weeks to hours [6,7].
In 1989 I became one of the founding employees of Arris Pharmaceuticals in Cambridge, MA, a startup founded to combine advances in cell biology, combinatorial synthesis, and artificial intelligence for drug design. I learned to use Tcl/Tk at Arris and built systems which visualized the operation of the directed learning systems we built to identify structure-function relationships in molecules with biological effects [3,4,5].
In 1982-1989 I worked for the Mark Williams Company in Chicago, one of the great also-rans of the computer age. I retargeted their c compiler to the 68000, participated in the port of Coherent, a UNIX V7 clone, to the IBM-PC, produced the MWC c development system for the Atari-ST, and actually ported Coherent to the Atari ST, although that never became a product. I also participated in the design and production of ANSI C: A Lexical Guide .
In 1979-1981 I worked for the Sidereal Corporation, a manufacturer of computerized teletype terminals then located in Portland, Oregon. I retargeted the Kernighan and Ritchie UNIX c compiler to the 68000 architecture at the same time that Chris Terman was doing similar work at MIT using Steve Johnson's PCC package.
In 1978-1979 I worked at Reed on an NSF grant, starting under Dave DeSante and finishing under Steven Stearns, to develop computer programs for teaching population biology. This produced demonstrations of predator-prey interactions, competitive exclusion, the chaotic possiblilities of discrete logistic growth, and other models, all programmed in my own language which implemented matrix and vector operations in a free interpretation of Dirac's bra-ket notation. It also produced a paper criticizing Joel Cohen's Food Webs and Niche Space for misunderstanding the mathematical methods it introduced for analysis of trophic relations.
I graduated from Reed College in 1978 with a degree in Chemistry and a concentration in Biology. My senior thesis examined the expected frequencies of palindromes in DNA sequences by comparing their frequencies in native DNA to their frequencies the same sequences transformed by
global transition, transversion, and transition-transversion mutations, and employed lots of nifty monochrome graphics printed from the Tektronix 4014 terminals onto their dedicated thermal printer.