Bunji Sakita 1930 - 2002
Professor Bunji Sakita, distinguished professor of physics at the City College of New York (CCNY), passed away on 31 August while in Japan, after a year-long struggle with cancer. Sakita was born in 1930 in the Toyama area of Japan. He received his first degree from Kanazawa University in 1953, and his Masters from Nagoya University in 1956 as part of Sakata's group. He was among several students recruited by Robert Marshak to come to Rochester University, and received his doctorate in 1959 under Charles Goebel. He went on to a postdoctoral position and a professorship at the University of Wisconsin. During this time, and while visiting the Argonne National Laboratory, Sakita wrote a series of influential papers on the quark model, introducing the new symmetry group SU(6), which generated considerable scientific interest among physicists as it united spin and isospin.
At Wisconsin, Sakita wrote some of the fundamental papers on the "dual resonance model", which now forms the foundation of string theory. With Goebel, he generalized the Veneziano amplitude to the many-particle case. With K Kikkawa and M Virasoro, he showed how to correct a crucial defect in the theory (unitarity) by including loop diagrams, much like Feynman diagrams. With his student C S Hsue, and with J L Gervais, he then generalized this to the functional formalism, based on Riemann surfaces, which today provides the most powerful formulation of string theory. Sakita and Virasoro also showed how ordinary field theories, in the infinite loop limit, can create fishnet diagrams which approximate string theory (which helped form some of the basis of the 1/N approximation). In these seminal papers, we see the foundations of string perturbation theory, the string functional formalism, and conformal field theory.
After J Schwarz, A Neveu and P Ramond introduced spin into the dual resonance model, Sakita and Gervais revealed the supersymmetry underlying the theory by writing down the first linear supersymmetric action, which today forms the basis of the superstring action. (Different versions of supersymmetry were also discovered in the Soviet Union at around this time.) B Zumino and J Wess, stimulated by the paper of Sakita and Gervais and by a seminar Sakita gave at CERN in the spring of 1973, then generalized this to a variety of quantum field theories defined in four dimensional space-time, rather than the two-dimensional world sheet. This led to the beginning of the application of supersymmetry to the physical world. (Sakita fondly remembers, in his memoirs, talking to Zumino in the CERN coffee lounge. Sakita said: "If you allow me to use anti-communting c-numbers, Gervais and I have written down a transformation of a fermi field to a bose field in the Nuclear Physics paper." Zumino replied" "It's OK to use anti-communting c-numbers. Schwinger has frequently used them.")
With the rapid expansion of the graduate physics programme at CCNY in the 1970s, Sakita followed Robert Marshak (who became president of the college) and joined the faculty at CCNY as distinguished professor in 1970. He presided over a rapid growth of the high-energy group, which developed string field theory, superconformal gravity, and research in strong coupling theory and collective co-ordinates. He received the Guggenheim Fellowship in the 1970s and was awarded the Nishina Prize in Physics in 1974. With Gervais and his student A Jevicki, he wrote a series of papers presenting the general formalism of collective co-ordinates and its full perturbation theory, allowing one to link point particle theories into those describing extended objects. This was extended to gauge theories with his student S Wadia, and Gervais. With his student J Alfaro, he applied the method of stochastic quantization to large N theories.
Sakita's interests were broad and varied, always seeking out the fundamental basis found in physical systems. In later years, he turned his attention to problems in solid-state physics, especially two-dimensional systems which exhibit W symmetry, and also physical characteristics found in high-energy physics (for example the fractional Hall Effect) with his colleagues S Iso and D Karabali.
Sakita leaves behind two children, Mariko and Taro. His warmth, leadership, modesty, and vision will be sorely missed by his students and colleagues all over the world and especially at CCNY.
Antal Jevicki, Brown University; Michio Kaku and Parameswaran Nair, CCNY; and Spenta R Wadia, Tata Institute of Fundamental Research, Mumbai.
Karl Brown 1925 – 2002
Accelerator pioneer Karl Brown died on 29 August in Stanford, California, where he had spent most of his working life. Professor emeritus of applied research at the Stanford Linear Accelerator Center (SLAC), Brown pioneered the development of linear accelerators for research, as well as for cancer treatment.
Brown attended the University of Utah as an electrical engineering student, but in 1946 he transferred to Stanford to work on particle accelerators. This move was the beginning of a Stanford career spanning more than half a century. He earned his doctorate in 1953 for the commissioning of the Mark II accelerator, following which he joined the Stanford physics department's Hansen laboratories. In the early 1960s, when Wolfgang Panofsky conceived the idea for the Stanford linear accelerator, Brown became a member of the core team of young scientists who designed and built the 2 mile long accelerator under Panofsky's direction.
In 1958, Brown was the first to use matrix algebra to calculate magnetic-optical aberrations in charged particle spectrometers, used by physicists for the precise analysis of nuclear and subnuclear structure. He developed a computer code called TRANSPORT to facilitate the equipment design process. This code later became a tool used worldwide to design spectrometers, beamlines and accelerators ranging in energies up to 1 TeV.
Brown also introduced the use of sextupole magnets to enhance the performance of spectrometers at SLAC. In the 1960s, he proposed making a colliding beam machine using two linear accelerators at SLAC. Later, he designed achromatic magnetic optical systems, which focus beams largely independent of their energies. His designs made it possible to achieve beam spots of a micron or less. They found application in particle colliders as well as in medical diagnosis and treatment. Travelling worldwide to assist in design of spectrometers and beam transport systems, Brown took sabbaticals in 1959 at Orsay in France, from 1972 to 1973 at CERN to work on the SPS and LEP, and from 1992 to 1994 at the ill-fated Superconducting Super Collider in Texas.
Though internationally renowned as an expert in beam optics for spectrometers and high-energy particle accelerators, Brown's greatest satisfaction came from his contributions toward the development of small linear accelerators for radiation therapy. As a graduate student in the 1950s, he was part of a small research team at Stanford that designed the first linear accelerator in the US to be used successfully to treat a cancer patient. In the late 1960s, Brown initiated and led the development by Varian Associates of the first commercially successful line of such machines, the CLINAC series. The present-day incarnation of the CLINAC treats more than 100,000 patients a day worldwide.
A fellow of the American Physical Society, Brown was awarded the 1989 prize for achievement in accelerator physics and technology by the US Particle Accelerator School.
"He is probably best known internationally for his development of the programs which make it possible to easily trace the path of particles through complex magnetic beam transport systems," said long-time friend Panofsky. "However, his contributions go well beyond that, and we all are extremely sad about his passing."
Swiss company Sentron AG announces two new integrated Hall sensors that respond to magnetic field parallel to the chip surface. Both are complete integrated CMOS systems including Hall elements, biasing circuit, amplifier and programming of gain, offset and temperature coefficient. The CSA-1 is a single axis sensor with sensitivity up to 300 mV/mT. The 2SA-1 is a two-axis sensor. Information is available at http://www.sentron.ch.
Slovenian company Instrumentation Technologies has announced a reconfigurable integrated beam-position monitoring system DBPM2. The instrument boasts sub-micrometre resolution and over 100 dB dynamic range. It consists of analogue processing, data acquisition and digital signal processing building blocks with a simple memory-mapped VME interface. Email info@i-tech.si or see http://www.i-tech.si.
Fortran still going strong
In "Particle physics software aids space and medicine" (CERN Courier June), Maria Grazia Pia and Jürgen Knobloch write: "The predecessors of the Geant4 toolkit - which were written in the now almost obsolete Fortran language..."
Fortran may be falling into disuse in the particle physics community, but that it is "almost obsolete" is a myth. There have been three standards since 1966, and a fourth is nearing publication. Perhaps the authors, and the rumour they are repeating, were referring to Fortran 66. In that case, they were understating their case: Fortran 66 is not "almost obsolete" - it is definitely obsolete, and has been so for 26 years.
It has been said that more people are using Fortran than ever before, but a smaller fraction of the people who develop software are using it. That is, the market is growing, but the market share is shrinking. I cannot find concrete support for this assertion (at least the first part of it), but there is some anecdotal evidence; at least one Fortran compiler vendor cites a continuous yearly increase of sales volume.
I would be curious to know why some consider Fortran to be "nearly obsolete"? If one compares Fortran to other languages, one doesn't find it wanting for features. Indeed, one finds many features tailor-made for its application domain - numerical computation - that are lacking in other languages.
Van Snyder, Jet Propulsion Laboratory, California Institute of Technology.
Jürgen Knobloch replies:
The term "obsolete" in our article was certainly inspired by the situation in the particle physics community. The major experiments in the field have decided to develop their new software in object-oriented technology - currently mostly in C++. The FORTRAN and C-based CERN library has not received any new developments since 1996, and a final release was delivered this year.