Daily Archives: March 3, 2009

Jack Schwartz: Work is a Signed Quantity

The sagest advice I ever received from Jack Schwartz was:

Work is a signed quantity.

Advertisements

John Markoff of the New York Times to Write the Obituary for Jack Schwartz

I just got off the phone after a call with John Markoff of the New York Times, who has been asked to write the obituary for Jack Schwartz, who died yesterday at the age of 79. Peter Capek had given my name to John, and John asked me to talk of my years with Jack.

John and I spoke for over half an hour, during which I learned that John knew both John Cocke and Ken Kennedy. Both, like Jack, are sadly no longer with us. Ken was the first Ph.D. in Computer Science at NYU/CIMS; Jack was his thesis advisor. Jack was also my thesis advisor.

It will be interesting to see which, if any, of my observations make it into print.. Here are a few of the key points I tried to make:

Jack was a world-class mathematician. While at Yale as a post-doc, his collaboration with Nelson Dunford resulted in one of the fundamental works in mathematics in the last century, a multivolume work known to mathematicians worldwide simply as Dunford and Schwartz. This work established both his reputation and that of the Yale Mathematics Department. [1]

Jack was similar to the great mathematician David Hilbert in that, as Wikipedia puts it, “He invented or developed a broad range of fundamental ideas in many areas.”

In Jack’s case these fundamental areas included: creating SHARER, the first timesharing system for the CDC 6600 computer, then the world’s fastest computer; program optimization, as expressed in the classic “Cocke and Schwartz; Programming languages and their compilers : preliminary notes;” creating SETL, a programming language based on finite set theory, that served as the inspiration for the Python language; ultracomputers, and his work on robotics. [2]

Jack’s collaboration with the legendary IBM computer scientist John Cocke is but one instance of the several cases known only to his colleagues in which Jack’s efforts brought recognition to people who would otherwise have not gotten the credit that was their due.

For example, Jack was the first person to bring to the attention of the Western world the work of Andre Ershov, who did fundamental work in program optimization in the USSR in the early 1960’s.

Jack also helped to publicize the pioneering work on program optimization byf John Cocke and Fran Allen of IBM. [3]

Greg Chaitin of IBM Research, like Jack world-class mathematician, would be much less well-known were it not for Jack. While an undergraduate at the City University of New York in the mid 1960’s, Greg independently discovered what is known as “Kolmogorov Complexity.” Jack traveled widely, and around 1972 or so he was in Argentina when he learned that Greg was working in Argentina as a systems programmer for IBM. Jack immediately got on the phone to John Cocke at Research, and told John to bring Greg to Yorktown. When he arrived at Yorktown, Greg joined that legendary “801 Project” that resulted in RISC architecture. Greg worked on the then unsolved problem of register allocation. He solved it, by first showing that it was computationally near intractable in that it was equivalent to a problem in coloring a graph, and then finding an efficient approximation known as Chaitin’s algorithm.

Jack was also one of the first computer scientists to visit China following President Nixron’s historic trip to China in 1972. [4]

On a personal note, Jack was also the brother-in-law of the literary critic Alfred Kazin.

During our call I mentioned that Jack was on my personal list of the five smartest people, and that John’s colleague at the New York Times, Tom Friedman, was Smart Guy #5. I said that one of the goals in writing this blog was to get Tom to enter a comment to my post HELLO Tom Friedman. John said he would pass on the request to Tom, so readers of this blog will be able to join me in learning if John has real clout.

Notes:

1. My youngest daughter attended Yale, and whenever I happened to walk by the small building that housed the Yale Department of Mathematics, I thought of Jack and his years in New Haven.

2. See my recent post On “Farewell World” and Programming for the influence of SETL on Python.

SHARER was possible only because of a marvelous programming hack due to Jack. I will write about it in a future post, “Jack’s Hack.”

3. Jack mentioned to me that he wrote most of “Cocke and Schwartz” while serving as a juror on a trial that went on for several months.

4. Jack knew the nephew of Chou En-Lai’s interpreter. (I forget the name, but recall he taught at a college on Staten Island.) Jack gave a series of lectures in China, during which he began the translation of the terminology of computer idioms into Chinese. He found “hash” particularly hard to translate.

Jacob “Jack” Schwartz Dies at 79

I learned yesterday that Jack Schwartz had died early in the morning at the age of 79.

I learned this sad news from Peter Capek, who had been told by Fran Allen. Hank Warren, a mutual friend for almost forty years, was on the cc: list. I replied as follows:

Sad news, indeed. Though I hadn’t spoken to Jack in years I thought of him earlier today, before learning of his untimely death.

I checked out a copy of Steve Lohr’s book about programming, “GOTO,” a week or so ago, and finally got around to reading it last night. (I noted with interest that you, Fran, Dick Goldberg,Lois Haibt and Watts Humphrey were among those Steve consulted in putting the book together.)

I knew all except Watts well from my days at Research, and once met Watts when Jack and I traveled to Poughkeepsie early in the SETL days to brief Watts on the project.

Those memories got me to thinking about the best programmers I have worked with during my career.

Hank was the first fully professional programmer I met; I recall him quietly sitting at his desk at CIMS (he was on loan from IBM to CIMS for about a year) churning out flawless PL/I code hour after hour.

Robert B. K. Dewar of CIMS and Chet Murthy of IBM are the two best programmers, in the usual sense, that I have worked with. But, taken all in all, I place Jack at the top of the list.

I recall almost to the hour the moment I encountered Jack around 90th and Broadway about September 1969. He knew I was looking for a thesis topic, and said that he had been working at IBM for the summer, and had been pursuing an idea.

The idea was to base a programming language on finite sets, to make “finite set theory executable.”

I thought it the most brilliant idea about programming I had yet heard, and still do today.

He said that since he had done the work on IBM’s dime he had to offer it to them, but he expected they would decline, and if they did then he would start a project on it at CIMS.

That insight, and the first programs he wrote in SETL, as he refined the language by taking algorithm after algorithm and expressing them in SETL, make him the best programmer I ever worked with.

Please keep me informed of any memorial services.

Karin and I both mourn his lossing. Not only was he the most brilliant person I ever worked with, he was, along with Fran, the nicest.

I understand the Ed Schonberg is arranging a memorial event, to be held on or about March 27th. I expect the details of this will appear at the CIMS site,

I plan to write a memoir about my years with Jack and some of the stories thereof, but want to take my time doing so, so I can savor the memories as I write it.

Jack Schwartz — May His Memory Be a Blessing.

  • Pages

  • March 2009
    M T W T F S S
    « Feb   Apr »
     1
    2345678
    9101112131415
    16171819202122
    23242526272829
    3031  
  • RSS The Wayward Word Press

  • Recent Comments

    mrrdev on On being the maintainer, sole…
    daveshields on On being the maintainer, sole…
    James Murray on On being the maintainer, sole…
    russurquhart1 on SPITBOL for OSX is now av…
    dave porter on On being the maintainer, sole…
  • Archives

  • Blog Stats

  • Top Posts

  • Top Rated

  • Recent Posts

  • Archives

  • Top Rated