|
|
|
Introduction
|
In 2004 the Russian Academy of Sciences asked me to write an essay to be published in a volume commemorating the seventy-fifth anniversary of Academician Andrei P. Ershov’s birth.
Academician Ershov, the founder of a successful and prolific school of computer scientists in the USSR and in Russia, was my mentor and boss for those few years in the early 70’s when I was doing Computer Science research at the Soviet Academy of Sciences. The research was centered around project BETA.
The goal of BETA was to produce a multi-language compiler that would translate every one of the source languages it accepted into its Internal Language, which in turn would be translated into the (machine) language of a given family of computers. The compiler would perform intensive optimizations on the level of the Internal Language. The source languages contemplated at the inception of BETA were Algol-68, PL/1, Simula-67, and Pascal.
While BETA itself did indeed end without producing a working system, there is an intellectual offspring of BETA that achieved all the goals of BETA and perhaps far more. This project is the LPI Multi-Language Family of compilers (LPI-MLF) produced by Language Processors, Inc. (LPI), a company that I founded in Massachusetts in 1980.
LPI-MLF was the first multi-language compiler system that succeeded in practice. It resulted in a set of compilers that were very competitive with standalone compilers written for specific languages. The system was practical, usable, and had been actually retargeted and extensively used on several very different computer architectures -- in short, it was the first production quality multi-language compiler system. The source languages implemented by LPI-MLF were COBOL, RPG-II, FORTRAN, PL/1, Basic, Pascal, and C.
This essay, A Tale of BETA’s Wayward Son, describes LPI-MLF project, its relationship with BETA and other multi-language compiler projects, and concludes with my view on the legacy of Academician Ershov. The attachments to the essay are at the end of this page.
Michael I. Schwartzman, President
ValueSearch Capital Management
Those Who Helped
The high-technology project described in this essay was a massive and risky undertaking. While I was the founder, chairman and president of Language Processors, Inc. (LPI), the high-tech company in Massachusetts that started and developed the project, and successfully marketed the resulting highly advanced software, many people were involved in the project along the way, or at the stages when it was being conceived by me.
These people gave me their help and friendship, and forgave me my mistakes. Some of these people knew little about the project, and some gave me their help and kindness long before the project got under way.
At the considerable risk of not mentioning those whom I may forget or accidentally omit, these were, in chronological order, with their relevant affiliations when I knew them:
Prof. Iosif Krass
Professor of Mathematics, Novosibirsk University, Akademgorodok, Novosibirsk, USSR
Gennady I. Kozhukhin
Head, Department of System Programming, Computer Center of the Siberian Branch of the USSR Academy of Sciences, Akademgorodok, Novosibirsk, USSR
Prof. Yuri L. Ershov
One of my thesis advisors, Professor of Mathematical Logic at Novosibirsk University, Akademgorodok, Novosibirsk, USSR
Corresponding Member of the USSR Academy of Sciences
Dr. Igor V. Pottosin
Head, Department of Theory of Compiler Optimization Methods, Computer Center of the Siberian Branch of the USSR Academy of Sciences, Akademgorodok, Novosibirsk, USSR
Prof. Andrei P. Ershov
Head, Division of Computer Science, Computer Center of the Siberian Branch of the USSR Academy of Sciences
Professor of Computer Science at Novosibirsk University, Akademgorodok, Novosibirsk, USSR
Corresponding Member of the USSR Academy of Sciences
Vadim Kotov
Specialist in concurrency theory and its applications to the system design and analysis. Senior Researcher at the Computer Center of the Siberian Branch of the USSR Academy of Sciences, Akademgorodok, Novosibirsk, USSR
Prof. Jacob T. Schwartz
Chairman, Computer Science Department, Courant Institute of Mathematical Sciences, New York University, New York, NY
Dr. Valentin F. Turchin
Senior Researcher at the Keldysh Institute of Applied Mathematics, Moscow, USSR
Frances Allen
Specialist in compilers, large computer systems and theory of program optimization.
Principal Scientist, IBM Research Center, Yorktown Heights, NY
Ted Lehman
Investment Banker and Philanthropist, New York, NY
Daniel D. McCracken
An author, Consultant, and Vice-President and later President of the Association for Computing Machinery (ACM)
Dr. Bruce Gilchrist
Director, University Computer Center, Columbia University, New York, NY
Prof. Richard B. Kieburtz
Chairman, Computer Science Department, New York State University at Stony Brook, NY
Ron Ham
Manager, Department of Compiler and File Management Systems Development, Digital Equipment Corporation (DEC), Maynard, MA
Joe Carchidi
Manager, File Management Systems Development Section, DEC, Maynard, MA
Dave Cutler
Consulting Software Engineer and Manager, Development of Advanced Operating Systems, DEC, Maynard, MA
Bernie Lacroute
Program Manager, Advanced Computer Systems Development, DEC, Maynard, MA
Gordon Bell
Vice-President, Advanced Research, DEC, Maynard, MA
Roger Gourd
Manager, Advanced Systems Software Development, DEC, Maynard, MA
Larry Portner
Vice-President, Software Development, DEC, Maynard, MA
Michael Greata
Director, Software Development, Prime Computer, Inc., Natick, MA
Dr. John William Poduska, Sr.
Vice-President, Research and Development and Co-Founder, Prime Computer, Inc., Natick, MA
John Ankcorn
Senior Systems Software Developer, DEC, Maynard, MA
Later: Vice-President of Language Processors, Inc., Waltham, MA
Anton Chernoff , Tom Friedman, Bob Kamsler , and Louis Tsien
The first systems software researchers who joined LPI soon after its founding, specialists in compiler and operating systems development
Ellen Marsan
My assistant at LPI
William C. Rogers, Esq.
LPI’s legal counsel
Corporate Law Specialist, Gaston & Snow, Boston, MA
Brent R. Nicklas
LPI’s investment banker and board member.
Vice-President of Merrill Lynch Capital Markets, New York, NY
Bob Schechter
LPI’s auditor
Partner-In-Charge, High-Technology Practice, Coopers & Lybrand, Boston, MA
|
Reviewers of this Essay
|
The essay was reviewed by many people with whom I also had a close working relationship in the past.
They are John Ankcorn, Joe Carchidi, Dave Cutler, Bob Freiburghouse, Richard Grove, Vadim Kotov, and David Kuck. Their encouragement, praise, and occasional disagreements with me were straightforward and frank. I was not in agreement with all of their suggestions, and bear full responsibility for all of the shortcomings found in the essay.
The affiliations of these reviewers, some already mentioned in Those Who Helped section, are as follows:
John Ankcorn
Formerly: Senior systems software developer at Digital Equipment Corporation, later Vice-President of Language Processors, Inc., Waltham, MA.
Currently: Specialist in Mobile Computing and Networking,
Senior Researcher at Hewlett-Packard Research Laboratories.
Joe Carchidi
Formerly: Manager of the group at DEC charged with maintenance and development of the follow-on versions of the VAX/VMS operating system.
Dave Cutler
Formerly: Consulting Software Engineer and Manager in charge of development of advanced operating systems at DEC, Maynard, MA.
Currently: Sr. Distinguished Engineer, Manages a large team of system programmers developing several Windows versions at Microsoft Corporation.
Bob Freiburghouse
Formerly: Senior software developer at MIT, Cambridge, MA
Later: The founder of Translation Systems, Inc., Cambridge, MA, an advanced systems software manufacturer, and later a co-founder of Stratus Systems, Westborough, MA, a manufacturer of non-interruptible computer systems.
Richard Grove
Specialist in compiler development.
Formerly: Senior systems software developer and project leader at DEC, Maynard, MA; Currently: Senior researcher and a Fellow at Intel.
Vadim Kotov
Corresponding Member of the Russian Academy of Sciences, a specialist in concurrency theory and its applications to the system design and analysis.
Formerly: Director of the Institute of Informatics Systems of the Siberian Branch of the Russian Academy of Sciences.
Currently: Director of Engineering, High Dependability Computing Program, School of Computer Science, Carnegie Mellon University, Moffett Field, CA
Prof. David Kuck
Formerly: Director of the Center for Supercomputing Research and Development at the University of Illinois at Urbana-Champaign, later a co-founder of KAI Software Lab, a leading provider of performance-oriented compilers and programming tools used in the development of multithreaded applications.
Currently: Intel Fellow, Software and Solutions Group Director, Parallel and Distributed Solutions Division, Intel Corporation. Also, an emeritus faculty member of the Computer Science and Electrical and Computer Engineering departments of the University of Illinois at Urbana-Champaign.
The frankness of the exchanges with these reviewers reminded me again how blessed I was to work with each one of them at some point in the past thirty years, and how blessed I was to cross paths with these ultimate professionals for whom the old-fashioned values of the software engineering profession remain as valid today as they were when we worked together:
– never leaving the project until it is finished or cancelled no matter how high the salary offer by a competitor and no matter how maddening the current management structure is,
– allegiance to one’s current employer and honorable behavior in protecting current and past employers’ proprietary information,
– dedication to the software engineering profession requiring one’s emotions be set aside in order to achieve the best engineering product.
I am grateful to these reviewers for their comments and I am glad I had the chance to cross paths with them again.
Michael I. Schwartzman, President
ValueSearch Capital Management
|
|
A Tale of BETA's Wayward Son .
Please click here to see the document.
|
|
A Russian Edition of the Essay
A Russian translation of the essay has been published by the Russian Academy of Sciences. The attached Russian translation is nearly identical to the version published by the Academy.
Please click here to see the document.
|
|
Attachment A, a Datapro report on LPI Multi-Language Family, 1989
If you are having difficulties with the attachments being displayed correctly, be patient – they are very large files – and hit the Refresh button on your Internet browser once in a while, this will clear-up the difficulties in most cases.
Please click here to see the document.
|
|
Attachment B, an LPI advertisement of the Multi-Language Family, 1986
Please click here to see the document.
|
|
Attachment C, LPI company profile and product list, 1987
Please click here to see the document.
|
|
Attachment D, LPI company profile and product list, 1989
Please click here to see the document.
|
|
Attachment E, a memo from Dave Cutler to VAX compiler developers, DEC, 1976
Please click here to see the document.
|
|
Attachment F, a letter from M. Schwartzman to Prof. A. Ershov, 1986
Please click here to see the document.
|
|
Attachment G, a memo from M. Schwartzman to LPI R&D dept., 1986
Please click here to see the document.
|
|
Attachment H, an LPI post-merger newsletter for customers, 1990
Please click here to see the document.
|
|
|