Week 4 Media Technology & Society

Winston readings – Drew

The first computers

At the beginning of the 20th century there was a real need for accurate firing tables for the ranging of artillery. And the onset of the First World War there was still a lot of guesswork involved in shooting a large gun. Creating a firing table involved tabulations of dozens of factors across thousands of possible trajectories, any one of which represented half a day’s work for a human computer with a desk calculator. This was the so-called firing table crisis. This dilemma was the supervening necessity for a fully electronic complex difference analyzer, what would eventually become the computer.

In addition to the laborious nature of computing firing tables, the British were keenly interested in breaking the codes used by the German U-boat fleet. Advanced electronic calculators were needed for cryptanalysis to overcome the German submarine dominance in the Battle of the Atlantic. British intelligence knew the Germans were using an encoding machine called Enigma to encrypt messages sent to and from the fleet. Polish intelligence had been aware of the encryption system since 1928. Eventually the Poles produced a contrivance called the Bombas to reverse the Enigma’s encryption key, unlocking the messages. While this worked with early German codes, over time the Germans added more and more variables to the schema. This required more and more computing power in the British Bombas to break the code. The supervening necessity to break the codes locked the English and Germans in an ever escalating dance, where tougher ciphers required more computations to read the encrypted messages. There were refinements to the technology, a natural evolution from an army of clerks to electrical relays to paper tape to crude internal memory systems. Eventually, I am sure because everyone must have realized that Bombas sounds more like a dance than a secret weapon, they gave their contraption the designated name, Colossus.

Like many of the inventions we have reviewed in this class, there were competing claims on who was responsible for what innovation in the evolution of the computer. For the development of the American variation of Colossus, Iowa physics professor John Atanasoff started it all when he built a binary digital device for solving linear algebraic equations. In 1940 he shared his concepts and discoveries with John Mauchly from the Moore School. Atanasoff failed to patent his machine before moving on to a new position at the Naval Ordinance Lab. Building on what he had learned, Mauchly wrote a paper in August of 1942 on “The Use of High Speed Vacuum Tube Devices for Calculating”, which led directly to the “Electronic Numerical Integrator and Computer”, or ENIAC, project. One of the significant differences with the ENIAC compared to comparable computers of the time, was the sheer number of tubes used as valves, or switches, in the computer. Over 17,000 tubes operating at 100,000 pulses per second. Due to wartime shortages it took over three years to build.

The immediate post-war period, from 1946 until the first years of the 1950’s, was subject to an iron rule called in America the ‘von Neuman Constant’. In Britain it was called the ‘Hartree Constant’.  This constant, a variation of the law of suppression of radical potential, would slow the development of the computer because of institutional inertia. This draconian obstruction of development by the science bureaucrats would retard development for over half a decade.

Compared to the Colossus, the vision of the ENIAC was more advanced with its electronically stored program and extensive memory. But with the end of the Second World War, the need for computing seemed less urgent. Without firing tables to compute or codes to break the need for massive computing power was reduced to the interests of a largely academic community. Throughout the development of American computing there was a close relationship between the academic, industrial and military communities. Like in any academic environment, the scientists working on pushing the power of these nascent machines were always looking for funding. By 1945 when ENIAC was finally working, scientists from Los Alamos were the first users, exploring the computation of some problems in thermonuclear ignition. Their calculations were even used to debug the machine. But even though there was interest from the a-bomb scientists, the government’s interest in building computers was waning. That is until 1949, when the Soviets exploded their atomic bomb. Suddenly the Cold War rekindled interest in building bigger and faster machines.

What followed was an alphabet soup of acronyms; ORACLE, MADM, EDVAC, SEAC, EDSAC, ORDVAC, ILLIAC, AVIDAC, JOHNNIAC and MANIAC. I was disappointed no one chose sacroiliac. All these computers were advancing variations on a theme; all had greater computational power, greater memory, and the rudimentary ability to execute programmed commands.

Suppressing the Main Frames

Winston points out that outside of the military industrial complex, the “law” of suppression of radical potential was fueled by three factors: indifference to computers, no programming language and an infatuation with ever larger hardware.

There were of course all the bureaucratic obfuscation we have come to know and love as computing tried to make inroads into non-governmental arenas. In one instance a private firm created by pioneers Eckert and Mauchly were looking for funding for their foundering company, EEC. As they were searching for investors the National Bureau of Standards forwarded their proposal to Bell Labs for review. Bell Labs made mechanical switches, not valves (tube based switches.) As you can guess, the Bell Labs review of the proposal was less than glowing. They told the governmental agency that their mechanically switched calculators were just as good at computation as the computer proposed by EEC. With limited funds, Eckert and Mauchly were forced to spend more time fundraising than they did advancing the technology. While there was uneven advancements in computers, the law of suppression was well entrenched.

IBM was late to the party. EEC was acquired by Remington Rand, maker of typewriters and adding machines, and not incidentally, IBM’s chief competitor. They were delivering large UNIVAC main frame computers to government and aviation customers. IBM seemed satisfied initially to be selling office equipment. That is until the US census bureau ordered its third UNIVAC computer. They had been dragging their corporate feet for six years before they made a decision to build a “defence calculator”. They saw specific potential for sales in the computational research being done on the Hydrogen bomb. Once again, the dollars being spent around the military industrial complex in support of the cold war accounted for their initial revenue stream. IBM proceeded to build a succession of ever larger computers, eventually putting Remington Rand out of the computer business.

If the advancements in hardware were slow in coming, the development of a useful computer language was nearly at a standstill. This glacial development pace of programming languages was definitely an obstacle to widespread adoption of the computer. They were so difficult to use that few business people could see cost-effective applications for them. Collectively, this fledgling industry had little interest in discovering how to sell computers, consequently they devoted next to no attention to programming. One of the reasons for this resistance was computer engineers rather liked the secret and arcane nature of the craft. Not just anybody could understand binary code. In fact, very few could. If you could read the code it was as if you were a member of an elite secret society. And if the members of the club don’t want new members, they probably won’t make it easier by creating a language that is more accessible. The computer industry has its share of straightforwardly hostile obstructionists, and the first generation of programmers would be exhibit “A”.  Obviously, market forces swayed the engineers to write a language that was more accessible. FORTRAN was developed by IBM only after Boeing developed their own language for running their IBM701. 

The final obstacle to widespread adoption of the computer was size. There was a prevailing belief that computers had to be growing exponentially in both computational power and the real estate they occupied. Remember, these were boys who felt they were doing big work and they needed big computers. You couldn’t do any significant work on a little wimpy thing. So IBM was building a series of ever larger computers. The 701 begat the 702 which led to the 704 which ultimately led to the 7030, also known as “Stretch”. Computing remained a big unwieldy institutional thing, totally inaccessible to the general public, wrapped in a shroud of mystery. Come to think of it, considering how difficult it is to get my laptop to work, you could argue that it still is.

Trackbacks are closed, but you can post a comment.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: