To the Multicellular Computing Home Page
Biographical info about the author

Steve Burbeck -- Areas of Research Publications, Presentations Patents

The various areas of research appear below in approximate time order with the most recent first. There has been considerable overlap between projects at various times.

Theory of Mind

This area of research has been a long collaboration with Sam Adams in IBM Research. The problem we attacked was how to model a theory of mind in which a General Artificial Intelligence (AGI) system is assumed to be embodied, i.e., it has internal and external sensory input and it is in constant active interaction with a changing world. That represents a radical departure from the earlier assumptions of AI. We also took the position that AGI is not possible without modeling "emotion" (i.e., affect), and "superstition" and "forgetfulness." (Those words are in quotes because they are only approximations of the more formal properties we actually modeled.)

This work was recently presented by Sam Adams at the 2007 Singularity Summit in San Francisco, Sept. 8-9, Transcript and slides, audio of presentation, video interview and commentary are on line.

Publications

Archiving Digital Records

From 2006 through 2008 I worked with the Collaborative Electronic Records Project (CERP), a joint effort of the digital archiving groups at The Rockefeller Archive Center and the Smithsonian Institution Archives.

Historical archives differ from corporate email archives. Historical archives must be preserved for much longer, they must deal with messages generated by a much more diverse set of email systems, and records must be preserved in a way that they can be interpretable decades hence. We designed and built a working prototype system that converted email in the common .mbox format to an XML representation conforming to the Mail-Account schema, a general XML schema co-developed with the North Carolina State Archives. The code, which is in Squeak Smalltalk, is open source and freely available here.

Archiving Conference Presentations

The CERP project was completed December, 2008. The work has been presented at several archiving conferences. Examples include:

The Interface Between Biology and Computing

As many have noted, computing systems and biological systems exhibit similar characteristics (see for example, this 2005 National Academy Report). Both fields are especially relevant these days, as is the interface between them. Researchers in each field find metaphors from the other field useful. Biologists increasingly use computers in their work. The new biological area called "Systems Biology" is strongly dependent upon computational techniques. And insights about how very complex biological systems are architected can give computing researchers ideas about how to design and manage complex computing systems, especially those immersed in the Web.

Publications

Podcast

Presentations

  • Burbeck. Evolution of Multicellular Computing: Parallels with Multicellular Life (pdf). Seminar presentation, Department of Computer Science, University of Birmingham, Birmingham, UK, Dec, 21, 2009.
  • Jordan, K. & Burbeck. What is Systems Biology: an Opportunity for Computational Science, Math and Engineering. Joint SIAM-SMB Conference on the Life Sciences, Raleigh, NC., July 31 - August 4, 2006.
  • Burbeck. TII/Vanguard Conference on The Challenge of Complexity, Los Angeles, September 27-28, 2004
  • Burbeck. An assessment of Computational Systems Biology from a computing perspective (presentation pdf). Genentech/CMEA Ventures Symposium on Systems Biology, San Francisco, June, 2003

Book Acknowledgments

  • Catalyzing Inquiry at the Interface of Computing and Biology, J. C. Wooley & H. S. Lin, The National Academies Press, 2005. Preface and Table of Contents available here (pdf). Discussion of the multicellular metaphor here.
  • Bio-2003, Published by Burrill & Co., San Francisco, 2003.

Service Oriented Architectures (SOA), Web Services and Peer-to-Peer (P2P) Software

To some this juxtaposition of topics may seem strange. What they share is that they are different varieties of distributed (what I now call multicellular) systems. SOA and Web Services seem at first to be much more structured than P2P software. But they simply exploit different kinds of structure. Tim O'Reilly makes a similar case. The messaging between P2P nodes is quite structured even though the identity of the collaborating systems is unknown in advance. My earliest work on this topic was at IBM Research where I investigated the scaling properties of a random forwarding message architecture for distributing information in randomly connected networks. That work was done in 1997, before P2P file sharing burst upon the scene.

Publications

Conference Presentations

  • O'Reilly Peer-to-Peer Summit, San Francisco, September 19, 2000. -- invited participant. Some raw notes from that summit are available here.
  • Intel Peer-to-Peer Working Group (San Jose, October, 2000) -- Speaker
  • World Internet Center "Thinktank on Peer-to-Peer" (Palo Alto, March, 2001) Keynote Address
  • European Conference on Peer-to-Peer (Amsterdam, February, 2001) -- Speaker
  • O'Reilly Peer-to-Peer Conference (San Francisco, February 14-16, 2001) -- Program Committee and Panel Speaker
  • THESEUS International Management Institute Conference on Peer-to-Peer Software (Sophia Atipolis, France, March, 2001) Speaker
  • O'Reilly Peer-to-Peer Conference (Washington DC, November, 2001) -- speaker ( see third photo)

Book Acknowledgerments

  • Peer-to-Peer: Harnessing the Power of Disruptive Technologies, Andrew Oram, Nelson Minar & Clay Shirky, O'Reilly & Associates, 2001

Open-Source Software (OSS)

A great deal of credit for IBM's current good standing in the OSS community is due to the two guys James Barry and Yen-Ping Shan, who in 1998 formed the alliance between IBM and Apache. That very successful project became the existence proof that getting IBM to participate in OSS was possible, if far from easy. As it turned out, Shan took so many arrows in his back that he left IBM soon thereafter.

The Apache deal was a tactical move that was driven in large part by the fact that IBM's own Web Serving software was losing out to both Microsoft's IIS and Apache. I was one of the small group of very early leaders within IBM (where there's never just one leader) arguing that it would be strategically advantageous for IBM to fully embrace open-source software and fold it into IBM's business wherever possible. I participated in the Corporate Task Force that formed IBM's Linux strategy and led the group in 1999 (together with Dan Frye under the auspices of the Corporate Technology Council), that developed the corporate strategy and the business case for IBM to embrace Open Source software. It was bad form then, and perhaps even now, to explicitly acknowledge that the strategy was aimed primarily at Microsoft and, to some degree, at Sun. However, since I instigated and led the strategy team and wrote the majority of the resulting report to the CTC, I can confirm that the 2002 ZDNet story had it mostly right when they said, "Open-source, widely viewed as a way for the development community to participate in the evolution of software that's owned by nobody but shared by everyone, was now a competitive weapon..." Much of the rest of that article is fairly accurate too.

Getting IBM to take a bold strategic leap is a bit like mating with a female Black Widow spider. The anecdotes in the last section of this Salon story give the flavor of the infighting and misunderstandings that the OSS revolution stirred up within IBM. It mentions just a few of the players and tends to get their roles wrong to boot. But IBM is so large, riven by rivalries, and confused that it would be a massive undertaking to reconstruct an accurate history. Now that Open Source has completely remade the landscape of the Software Industry, everyone who could possibly make a claim to be involved fancies that their efforts, or their executive decisions, were instrumental in that success. Credit tends to be attributed to those with access to the biggest megaphone: IBM's corporate PR machine.

Publications

  • Burbeck, Capek, P., et al. Open-Source Software: Implications for IBM. IBM Academy of Technology Report (IBM Confidential), September, 1999.

Conference Participation


Object-Oriented Software

My interest in Smalltalk and Object-Oriented Software began in 1985 when a small group I led at the Linus Pauling Institute ported Xerox PARC's Smalltalk-80 to the IBM PC-AT. We then spun out a little company called Softsmarts (long since defunct) to commercialize that port. One thing led to another and it became a central focus for my work until the late '90s

Publications

Conference Presentation

  • OOPSLA-86 (the first ACM Conference on Object Oriented Programming) -- Executive Committee
  • OOPSLA-87 -- Program Committee, Technical Reviewer
  • MacWorld (Boston '88 and 89, and San Francisco '89) -- Conference Faculty
  • SCOOP '89 -- "OOP: Past, Present & Future" panel
  • Software '89 -- Object-Oriented Programming panel
  • SCOOP '90 -- Workshop on OO Analysis and Design
  • C++ at Work '90 -- Workshop on OO Analysis and Design
  • Software Development (Santa Clara '92) -- Speaker
  • ObjectExpo (New York City '92) -- Speaker
  • IBM International Conference on Object Technology (San Francisco, June 1996) Speaker

Book Acknowledgerments

  • Object-Oriented Information Systems. David Taylor. John Wiley & Sons, 1992.
  • Developing Object-Oriented Software for the Macintosh: Analysis, Design, and Programming. Neal Goldstein & Jeff Alger. Addison-Wesley Publishing Co., 1992.
  • Object-Oriented Analysis. Peter Coad & Edward Yourdan. Yourdan Press, 1991.
  • Object-Oriented Design. Peter Coad & Edward Yourdan. Yourdan Press, 1991.
  • Programming with MacApp. David A. Wilson, Larry S. Rosenstein & Dan Shafer. Addison-Wesley Publishing Co., 1990.

Bioinformatics

From 1980 to 1988, I directed the scientific computing group at the Linus Pauling Institute of Science and Medicine in Palo Alto, CA. That group helped to pioneer techniques for computerized quantitative analysis and identification of proteins separated in 2-D PAGE gels -- what is now a part of the field of Proteomics. I also collaborated with Dr. Emile Zuckerkandl on DNA sequence analysis research (unpublished) using novel Fourier Analysis techniques. We sought to find repetitive short sequence motifs at the period of the nucleosome (165 - 200 bp) in regions of the genome containing human globin genes.  However the globin DNA sequences available at that time were too short to obtain reliable results. Recently such periodicity has been found by "wet lab" techniques. I also collaborated with other LPI researchers on several other bioinformatics projects.

Publications

  • Leavitt, J., Sun-Yu Ng, Varma, M., Latter, G., Burbeck, Gunning, P. & Kedes, L. Expression of Transfected Mutant beta-actin genes: Transitions toward the stable tumorigenic state. Molecular and Cellular Biology, 7, 1987, 2467-2476. (PubMed)
  • Ross, M., Latter, G., Burbeck, & Leavitt, J. Reduced area two-dimensional gels for direct digital imaging of radioactive protein profiles. Electrophoresis, 8, 1987, 249-250.
  • Leavitt, J., Sun-Yu Ng, Aebi, U., Varma, M., Latter, G., Burbeck, Kedes, L., & Gunning, P. Expression of transfected mutant beta-actin genes: Alterations of cell morphology and evidence for autoregulation in actin pools. Molecular and Cellular Biology, 7, 1987, 2457-2466. (PubMed)
  • Burbeck. The complexity of computerized microdensitometry: Implication for the design of a 2D-gel workstation. Invited paper presented to the EMBL 2D-Gel Workshop, European Molecular Biology Laboratory, Heidleberg, Germany, March 1986.
  • Leavitt, J., Latter, G., Lutomski, L., Goldstein, D. & Burbeck. Tropomyosin isoform switching in tumorigenic human fibroblasts. Molecular and Cellular Biology, 6, 1986, 2721-2726. (PubMed)
  • Goldstein, D., Djeu, J., Latter, G., Burbeck, & Leavitt, J. Abundant synthesis of the transformation-induced protein of neoplastic human fibroblasts, plastin, in normal lymphocytes. Cancer Research, 45, 1985, 5643-5647. (PubMed)
  • Burbeck, G. I. Latter, E. Metz & J. Leavitt. Neoplastic Human Fibroblast Proteins are Related to Epidermal Growth Factor Precursor. Proc. National Academy of Sciences. USA, 81, 1984, 5360-5363.
  • Latter, G., Burbeck, Fleming, J. & Leavitt, J. Identification of Polypeptides on Two-dimensional Electrophoresis Gels by computerized Amino Acid Analysis. Clinical Chemistry, 30(12), 1984, 1925-1932.
  • Latter, G., Burbeck, Fleming, J., Metz, E. & Leavitt, J. Measurement of amino acid composition by computerized microdensitometry: An aid in the identification of proteins on 2-D gels. Paper presented to the 4th annual conference on 2-D Electrophoresis, Argonne National Laboratories, June, 1984.
  • Burbeck. Direct digital imaging of radio-labeled 2-D gel beta emissions using micro-channel plate image enhancement. Electrophoresis, 4, 1983, 127-133.
  • Burbeck, G. I. Latter, E. Metz & J. Leavitt. Simultaneous Amino Acid Analysis of 100 Polypeptides in 2-D Gels by Computerized Microdensitometry. Paper presented at Electrophoresis '83, Boston, 1983.
  • G. I. Latter, E. Metz, Burbeck & J. Leavitt. Measurement of amino acid composition of proteins by computerized microdensitometry of two dimensional electrophoresis gels. Electrophoresis, 4, 1983, 122-126.
  • R. Marcuson, Burbeck, R. L. Emond, G. I. Latter, & W. Aberth. Normalization and reproducibility of mass profiles in the detection of individual differences from urine. Clinical Chemistry, 28, 1982, 1346-1348.

Mathematical Cognitive Psychology

This was the field of my PhD dissertation under Professor R. Duncan Luce at UC Irvine and Harvard University. My coursework and dissertation writing occurred at UC Irvine. The experimental portion of the research was done at Harvard with Duncan Luce and Dave Green (in wonderful William James Hall). I developed novel statistical techniques for using hazard functions to analyze reaction time distributions. For an up-to-date list of references to that work, Google [Burbeck Luce hazard "Reaction time"].

Publications

Conference Presentations 

  • Burbeck. Recovering decision latency distributions from reaction time experiments. Paper presented to the Acoustical Society of America, Pennsylvania State Univ., University Park, Pennsylvania, June 1977.

Mathematical Sociology

Upon finishing a BA in mathematics at California State University, Long Beach, I joined a research project (the Urban Disorder Project) that was investigating the internal dynamics of the large scale urban race riots in the '60s, e.g., the Watts Riot in Los Angeles. That research involved mathematical modeling and statistical analysis of census data and detailed spatial/temporal data on many of the individual "riot events" such as arson, looting, rock throwing, etc. See Clark McPhail's review for background and a short description of the results of this project in the section titled "Temporal and Spatial Variation."

Publications

  • Burbeck, W. J. Raine, M. J. Abudu Stark. The dynamics of riot growth: An epidemiological approach. Journal of Mathematical Sociology, 6, 1978, 1-22.
  • M. J. Abudu Stark, W. J. Raine, Burbeck, & K. K. Davison. Some empirical patterns on a riot process. American Sociological Review, 39, 1974, 865-876.
  • S. M. Moinat, W. J. Raine, & Burbeck. Black ghetto residents as rioters. Journal of Social Issues, 28, 1972, 45-62
  • W. J. Raine, M. J. G. Abudu, Burbeck, & K. K. Davison. Black ghetto violence: A case study inquiry into the spatial patterns of four Los Angeles event types. Social Problems, 19, 1972, 408-426.


IBM Patents

Contact: sburbeck at mindspring.com
Last updated September 12, 2011