One Boy's Boston*

During the period 1959 - 1965, apart from an interlude in Bologna, the Web siteholder and his wife lived in the Boston area, mostly in the Back Bay at 169 Marlborough Street, and after that in Bonn. Apart from the alliteration of their names, the three cities have much in common: all are leading university and medical centers, all are small enough for the entire downtown area to be comfortably covered on foot, and all are considered provincial as seen from such metropolitan centers as New York, Berlin and Rome.

For those in computing (the term computer science had not yet come into vogue), the early sixties was a heady time to be in Boston and Cambridge. While this was not exactly a Siglo de Oro in computing history, there was a definite feeling of being where it's happening. At MIT, time-sharing on IBM 7090s and 7094s was getting off the ground, allowing users effectively to run and debug programs online,** and other applications of such large new mainframes were being explored. Under Marvin Minsky's direction, research on artificial intelligence was well under way at Tech Square amid speculation as to how soon we will "have AI". John McCarthy and his group
(David Luckham, David Park, et al.) were inventing and implementing LISP, the principal AI programming language, and James Slagle and later Joel Moses used it for programs to do symbolic integration in particular and algebraic manipulation in general. Another computer algebra pioneer, Robert Fenichel, was writing an algebraic simplification program at Harvard.

Encouraged by Hao Wang, Martin Davis and Hilary Putnam, and inspired by their recently published algorithms, individual researchers at Harvard and MIT such as Joyce Friedman and the Web siteholder were busy constructing automatic theorem provers, and word of J. A. Robinson's new resolution principle, which was to revolutionize automatic deduction, was starting to get around. Although PROLOG had not yet made the scene, its immediate forerunners in exploiting predicate logic as a programming language, Carl Hewitt's PLANNER and Gerald Sussman's MICROPLANNER, were not far off. J. C. R. (Man-computer symbiosis) Licklider was laying the theoretical and financial foundations of the Arpanet and interactive computing, Ivan Sutherland's Sketchpad launched interactive computer graphics, Nicholas (Being Digital) Negroponte was preparing to enter the brave new world of multimedia, RLE director and future MIT president Jerome Wiesner was appointed presidential science adviser, and Martin Greenberger was making "startlingly prescient predictions" in The Atlantic Monthly about the future of computers as Vannevar Bush had done nineteen years earlier.

In the area of computational linguistics, Victor Yngve's bustling machine translation project invented COMIT, a string-processing language that later evolved into SNOBOL4 and SPITBOL, and applied it in writing a series of computational German grammars, while at Harvard Susumu Kuno and his group constructed the predictive analyzer for parsing English sentences and Anthony Oettinger started a section on computational linguistics in Communications of the ACM. (The ALPAC and Lighthill reports, which put dampers on MT and AI research at least for a while, had yet to be delivered -- in 1966 and 1973, respectively.) The advent of the internet has triggered a renewed interest in machine translation, with the wide availability of such tools as Yahoo! Babel Fish and Google Translate, though their output is often at best an approximation to a correct translation, and both are totally buffaloed by the three-word title of this essay. The renowned linguists Noam Chomsky and James D. McCawley were both for a time members of Yngve's project, while McCawley was becoming a leader of the generative semantics movement, and Chomsky was already a "transformational figure", as Colin Powell later described Barack Obama. A number of other denizens of the by now legendary Building 20, then and later, are introduced and discussed in The First Word by Christine Kenneally and The Linguistics Wars by Randy Harris. Two excellent sources on the history and present status of computational linguistics and machine translation are Yorick Wilks´s interview of Victor Yngve, and the web site of John Hutchins.

A number of early natural-language question-answering systems that combined some of the natural-language processing and algorithmic problem-solving techniques mentioned above were constructed during this time. Dan Bobrow devised METEOR, a version of COMIT embedded in LISP, and used it to write a program for solving algebra word problems, while Bert Raphael wrote a semantic information retrieval program in LISP, and the Web siteholder used COMIT for a program to solve logic word problems. Several QA systems of this sort were chronicled in detail in two surveys by Robert F. Simmons, who considered them in direct line of descent from McCarthy's advice taker model. Natural-language question answering is currently enjoying a renaissance on the Wolfram Alpha web site, though this program "isn´t sure what to do with" questions involving deductions from given premisses. Many of the researchers who got their start at MIT during this time soon left for the (west) coast, where they were instrumental in the rise of computer science in such centers as Stanford University, Stanford Research Institute and Xerox PARC. (More about MIT during this period, including many good photographs, used to be found at the Research Laboratory of Electronics Web site, rleweb.mit.edu. The Web siteholder does not know where these pics may have migrated to, and regrets their disappearance.)

While covering the city on foot so as not to forfeit one's parking place (Tom Lehrer's excuse for riding the MTA, as it was then called), one might stroll out to the Fenway (where a dog was heard being summoned by "Here, Peabody!") and visit the museums or the Red Sox according to one's tastes, or down to the Esplanade to watch the rowing shells work out on the Charles, or cross the bridge to MIT and maybe carry on to Harvard Square and marvel at the "design research" that went into the fabrication of "Bead Bird". Nowadays, of course, the more "wired" among us can let their digits do the walking to such Web-linked shops in the square as WordsWorth and the Coop, or even cover the whole Greater Boston area, restaurants and all, from their desktops. Apart from such virtual connections, however, the Web siteholder has no current acquaintance with the Boston area, having missed not only what Susan Conant in Ruffly Speaking termed "postfeminist Cambridge" but also any intermediate stages there may have been.

Nearby dans l'arrière-baie were Joseph's, a favorite restaurant of Thomas Crown, and Maître Jacques, the forerunner of today's Maison Robert. Or one might cross the Common for Sunday brunch at the Parker House (where an elderly diner was heard to protest "This is instant oatmeal! That's not a good advertisement," a claim stoutly denied by the waiter who insisted that the porridge in question was the slow-cooking variety), or to Faneuil Hall Market and the North End for shopping or restaurants stopping perhaps at Coogie's clam bar on Hanover Street (six cherrystones on the half shell for 45¢ or seven for 50¢), or run into state rep Michael Dukakis on Boylston or State Street and get the latest on Bay State politics. One was constantly reminded of the curious way Bostonians have of expressing themselves, as in their seeming passion for redundancy ("French or Gallic dressing?" -- query often heard in restaurants), for hedonism ("Pleasure Vehicles Only" -- sign on Soldiers Field Road) and for highbrow sports reporting ("Red Sox Applaud Adult Attitude of Rookie Yastrzemski" -- Christian Science Monitor), or a play-by-play announcer might report that someone was on first base "by virtue of" a single to right field.

Radio fare included the newscasts of Louis M. ("Well, here's the news") Lyons, who read excerpts from the book from which this web page takes its title, and Fulton Lewis, Jr., whose opinions held little that Northeast Corridor types might agree with apart from his byword "America begins at the Alleghenies," depending on which way one is going. The Harvard station broadcast classical music "orgies" that included every recorded work of a given composer (even the Mozart KVs 231 and 233), while striving for purity of the language (if not of the air) in their advertisement "Winston tastes good, as a cigarette should," and emphasizing the contribution a product might make toward the acquisition of wisdom as in "Be wiser Bud, buy ... " (you guessed it).

For part of this time, the President of the United States was a Bostonian, a state of affairs that looked like continuing for some time, the Speaker of the House of Representatives was likewise a Bostonian (as were two other holders of this office since 1947), the governor of the Commonwealth was Endicott Peabody '38 and the Celtics seemed set to go on forever, all of which even Cleveland (The Proper Bostonians) Amory could not have anticipated in 1940 and which merely enhanced Beantown's Hub-of-the-Universe quality.

There was of course a darker side to life in Boston. The Boston Strangler was at large during most of this time; the anxiety this caused was compounded by peoples' recent vivid memories of Psycho. Somewhat before these events took place, in the late forties the Web siteholder and family went on a rubberneck bus tour of the home of the bean and the cod, where the driver kept up a stream of self-styled witticisms. Passing the Charlestown prison, he cheerfully informed the passengers: "That's where they fried Sacco and Vanzetti -- they was bad boys so they had to sit on the hot seat," thereby neatly wrapping up decades of controversy about the case. (This was of course well before Gov. Dukakis's posthumous pardon.) At the end of the tour, Web siteholder's mother let the driver have a piece of her mind for that remark and received what was no doubt the standard reply: "Jeez, lady, I just work here, I never meant no harm."

__________________________________

*with apologies to Samuel Eliot Morison (One Boy's Boston: 1887-1901, Boston 1962).

**This was a vast improvement over the system of submitting batch jobs in trays of punched cards and waiting overnight to get comments like "Bad Compilation" or "Illegal Hollerith Character", or to be able to gloat over the program´s successful execution.

Time-sharers were connected to the mainframes by teletypes fitted out with rolls of paper that recorded protocols of online sessions. Shortly before the midday service break was over at one PM, computer nerds with sandwiches and cokes gathered around the available consoles in the terminal room at Tech Square to be on hand when the system came up, signalled by the simultaneous clatter of maybe two dozen typewriting machines. Late lunchers were sometimes left out in the cold.

The CTSS and Project MAC time-sharing systems were also widely used for writing memoranda and other messages and exchanging them within the user community (though not yet between the two IBM mainframes fitted out for time-sharing). While these relatively banal applications were thought by some to be a waste of the high-powered new technology at a time when most people with access to computers actually wrote programs and ran them, the memo-writing and mailbox procedures may now be seen as forerunners of modern text-processing and electronic mail systems.

Last modified: 12 December 2009