Computer Architecture and Networking
My first research project involved transmission of video signals by external modulation of a free-space laser beam.
I later returned to optical networking at Bell Labs in the 1970s, developing the first optical modem
deployed for computer networking in the US, as well as systems for video transmission
through optical fibers.
As manager of the hardware laboratory in the Hebrew University Computer Science Department in the 1980s, I built two 16-processor parallel computers as part of the Makbilan Parallel Computing Project and developed specialized communication hardware, bridging the two machines in support of collective communications (scatter/gather).
Foundations of Electrodynamics and the Problem of Time
This research program extends the Feynman-Stueckelberg
"proper-time" approach in quantum electrodynamics (QED) to a
wider set of issues in relativistic dynamics associated with the
role of time. The work is theoretical in nature, seeking to
resolve known difficulties in the Standard Model of elementary
particle physics. Standard relativity, induced from
relationships among measurements in moving reference frames,
imposes a priori constraints on the underlying observables of
space and time. The resulting mixing of spatial and temporal
coordinates leads to significant conceptual and computational
difficulties in areas ranging from relativistic spectroscopy to
questions of time travel. To avoid these constraints, the
Feynman-Stueckelberg approach to quantum electrodynamics (QED)
introduced a universal "proper time" parameter, not directly
related to the time coordinate in relativistic spacetime, and
treated the phenomena of pair creation and annihilation by
describing antiparticles as negative energy particles propagating
backward in time. Extending Stueckelberg's classical and quantum
framework to include dynamical interactions, the monotonically
increasing "proper time" parameterizes interacting event
evolution through space and time in a manner consistent with
standard relativity, but with significant formal, computational
and explanatory advantages, including the problem of the static
"block universe" and the "grandfather paradox".
In the context of this research, Stueckelberg's proposal has
been developed into a consistent and integrable microscopic
dynamics underlying standard electromagnetic theory that
approaches Maxwell's equations in its equilibrium limit. The
earlier detailed study of electrostatics will now be extended to
the electrodynamics of simple macroscopic radiating systems, in
order to identify practical experiments to test the theoretical
framework.
Publications in the past three years address the relativistic
harmonic oscillator and the self-interaction problem for
radiating particles and a detailed analysis of phenomenological
electrostatics with a discussion of experimental signatures.
This work is now being extended to describe the electrodynamics
of simple macroscopic radiating systems.
The International Association for Relativistic Dynamics (IARD)
was founded in 1998 to encourage collaborations among researchers
from diverse fields interested in these areas. I have presented
papers at the biennial IARD conferences since 1998 and have
served as IARD president since 2006. The IARD website is at www.iard-relativity.org.
The understanding of time also plays a significant role in the humanities, and yet much of that work is based on conceptions of time introduced by Newton in the 18th century. Since 2005 I have collaborated with Prof. Jonathan Boyarin, an anthropologist at the University of North Carolina, on a multi-disciplinary project intended to communicate modern physical conceptions of time to the humanities and allow physics to draw on a wider spectrum of cultural insights into temporality. In the past three years, we delivered a joint paper and presented a joint seminar on our earlier writing. During the next year we will continue work on the relationship of physical notions of time to Jacques Derrida's conceptions of iterability and the spectral. We plan to deploy this work in a moderated public website, encouraging contributions from interested scholars from around the world.
Social Impact of Computing Systems and Responsible Computing
The rapidly developing power of computing systems and the equally rapid drop in the cost of their implementation and operation are well known, and the social impact of these developments is frequently discussed. Less attention however is given to the conflict that can develop between these trends. While expensive computing systems become increasingly reliable, with MTBF on the order of decades, lower cost is often associated with compromise to RAS — reliability, availability, and serviceability. As a consequence, the pursuit of greater ‘productivity of investment’ often results in critical and essential public infrastructure systems implemented using products intended for momentary enjoyment. As early as 1996 in the US, the President's Commission on Critical Infrastructure Protection (PCCIP) warned that the incorporation of inadequate and unreliable technological components in systems supporting public and individual safety represents one of the greatest vulnerabilities to modern society. These warnings were born out directly in the events of September 11, 2001, including the massive failure of communication systems operated by the emergency services in NYC and the inadequate fire protection in the World Trade Center. As computing professionals critical of ‘spaghetti code’ and ‘crappy design’ we are in a position to understand the risks associated with using a cheap consumer-oriented operating system in medical emergency equipment, or making that equipment accessible to hackers through a public network. In the 1990s, I collaborated with Prof. Yonah Alexander on questions of this type, linking the vulnerabilities of critical infrastructure associated with ‘cyber-terrorism’ to those that derive from poor planning. Unfortunately, as the phenomenon of malware has grown in quantity and sophistication, the trend away from planning for RAS is growing in tandem. Today these questions cast a shadow over plans to deploy massive distributed control systems for ‘smart electric grids’ and centrally controlled transportation based on analysis of ‘big data’. I have touched on these problems in a number of papers and talks.
Stepping out of Violence
I appear in a 2012 educational film in dialog with psychoanalyst Jessica Benjamin of the Postdoctoral Program in Psychotherapy and Psychoanalysis at New York University. The film attempts to understand the interaction of psychological and political forces involved in resolution of inter-communal conflict.