Man + Machine: Understanding the development of real-time interactive computing

Module by: eric leshinsky

Summary: Real-time interactive computing- or the prevalence of the personal computer- may well be the hallmark of contemporary society. But how did we get here? This module will investigate various challenges, both intellectually and technologically, that were engaged in the development of personal computing.

Figure 1:

human evolution

“Real-time interactive computing” may not be the most eloquent way of describing what we do when we use personal computers but it distills what is easily the hallmark of contemporary society. As the overwhelming basis for contemporary economic productivity, lifestyle and general communication, the meaning and geneology of “real-time interactive computing” deserves to be better understood. In other words, if this type of computing is at the heart of what we do in a 21st century post-industrial society, how did we get here?

Figure 2: (left to right)the first analog computer- the abacus; textile looms operating on punch card patterns; extreme development of the analog computer: the Differential Analyzer, an intricate system of gears, switches and levers that required manual readjusting for each new problem;

human + machine evolution

To answer this question, the following module is structured around some of the various intellectual and technological hurdles that have been negotiated over the last 100 years to achieve true real-time interactive computing. Whether explicit or not, all of these hurdles represented some type development towards a more efficient and seamless relationship between humans and machines. And eventually, around 1960, this relationship came to be known as a “symbiosis” in the words of the prolific computer scientist J.C.R. Licklider. This term is crucial to understanding real-time interactive computing because it underscores that at some point the development of computing-machines came to be inseparable from an interrelated human development, that is, how humans interact with machines. And thus the importance of computing machines that demand real-time interaction with humans to achieve maximum value.

This will hopefully become more clear by the end of the module. If not, you might consider visiting an excellent historical overview at the Museum of Computer History

Shifting from Analog Thinking to Digital Thinking

This shift, which may seem obvious to some, was critical to just about all subsequent developments in computing, especially regarding memory and dynamics (see below). In essence, analog process represented a relationship between human and machine where the machine depended on constant supervision by the human to do anything. The power of digital process was in its fundamental efficiency, that a computer could rely on a set of numbered instructions for complex sequential operations thereby removing the human from constant computer babysitting. While the size of the computers in no way reflected this new efficiency- they remained huge machines for decades- they would fill their bulk with increasingly more detailed instructions, or programs, for how to operate. All based on a digital language. And while this important development may seem to move away from the future of real-time interactive computing, it actually enabled it by building up the capacity of the computer to respond to increasingly complex demands of human inputs.

Figure 3: (left to right) Convential IBM personal punch card tabulator; The Mark I: the first stored program computer; S.A.G.E./Whirlwind: commissioned by the U.S. Defense Deptartment for Cold War air-defense, it was the first digital and truly real-time interactive computer allowing human programmers to track the changing trajectories of missiles and planes.

human + machine evolution (continued)

The Problem of Memory

In terms of computing, memory is of the utmost importance to the actual enabling of real-time computing. Or, to say it another way, the alternative to not having memory in the computer is the tedium of continuously reminding the computer of the problem it just solved simply to move onto the next problem in a sequence. So, memory became essential to allowing humans to detach themselves from the computer. And while theory about human and computer interraction would eventually move toward symbiosis, at this point theory would say that it was better for computers to be more independent and memory was the way to achieve this. In essence, memory represents the foundations for software and the programmable computer we know today where instructions for complex sequences of problem solving can be embedded in the machine. In the early 1940's, Howard Aiken's Mark I computer - or the Automatic Sequence Controlled Calculator- became the first stored-memory computer which was quickly followed, and dwarfed, by the S.A.G.E./Whirlwind computer.

Figure 4: (left to right)S.A.G.E. in action: this Cold War air-defense linked hundreds of human radar monitors to the same computers-- considered the first real-time interactive computing;The TX-2 computer: the kernel of an idea about computers being used on a personal level; More real-time interactive computing: scientists play the first video game on the PDP-1 computer in the first case of 'fun' with computers;

human + machine evolution (cont'd)

The Problem of Dynamics

If the S.A.G.E. project (Semi-Automated Ground Environment) became the first example of real-time interactive computing, it has much to do with a very pressing need to compute information that was changing quickly. Like the majority of devlopments in 20th century computing, S.A.G.E. was designed for military use, specifically the ongoing Cold War fear of a Soviet missile or bomb strike. Thus, there existed an urgency to develop a computer that could effectively respond to the dynamics of a military attack, i.e. the need for a computer to process the quickly changing radar information of incoming missiles and enemy planes. Prior to this point, there simply did not exist any urgency to make computing respond to real-time changes in information. In other words, devices such as punch card machines-where information to be computed was fixed on cards- were adequate for the problems at hand.

From Electromechanical to Vaccuum Tubes to Transistors

On a purely practical level, computers would not be able to achieve real-time interactive computing unless they could run continuously for extended amounts of time without failing. And to do this meant to have a stable power distributor. Without explaining the intricacies of each technology, the computer evolved through essentially three types of distributors with electromechanical relays representing the analog phase, vaccuum tubes representing the early digital phase and transistors arriving in the latter stages. With each technology came advancements in information processing speed and/or reliability with transistors maximizing in both areas.

Figure 5: (left to right) IBM's System/360, one of many computers developed in the late 1960's and early 1970's for general office use; the standard office as dictated by the rise of the personal computer; The PalmPilot PDA makes real-time interactive computing pocket-size at the end of the 20th century.

human + machine evolution (cont'd)

Computers Become 'Fun' and 'Personal'

Again, what may seem obvious in light of our current fully-integrated-computer-lifestyle, but notions of the computer being both fun and operated on a personal level signify major philosophical turning points. This had something to do both with the early rationale for developing better computers and on a practical level, it took decades for scientists to reduce the hardware of the computer down to a personally manageable and affordable level.

The fact that the 20th century development of computers served as the hand-maiden to a constantly evolving need for help with increasingly complex scientific computation, largely related to warfare, was a strong reason for computers to remain outisde the purview of the average person. Put another way, the types of efficiencies that computers were asked to create, had nothing to do with the general types of commication-related and organization-related efficiencies to which we relate today.

In the mid-1950's, the sheer stodginess of the current computers and the overwhelming demand for them to achieve efficiencies simply for government bureaucracies began to unnerve certain scientists, namely Kenneth Olsen, Wes Clark and Harlan Andersen. They simply wanted a computer with which they could experiment freely without being bothered to devote the computer's time to bureaucratic number-crunching. "We believed computers should be fun. They were exciting. They could do so many things. The opportunities were just without bounds.", said Olsen in a later oral history project (NMAH Oral History). So in a very real sense, scientists had to shift their thinking about how computers should be used. But of course the parallel development was to enable computers to be smaller and more manageable which is heavily related to the shift from vaccuum tubes to transistors mentioned above. When scientists began recognizing that there was indeed a need to make computers "personal-sized",and to allow people to interract with computers on their own terms whenever they wanted and for whatever reason, the notion of symbiosis between human and machine that J.C.R. Licklider popularized simply fell into place.

This module has attempted to boil down the derivation of what is commonplace in contemporary societies: the prevalence of real-time interactive computing and the need for this to be so. But the reader should be encouraged to ask what's missing from this brief account. How does software fit into this equation? Or any of the peripheral devices on which we rely for real-time interactive computing such as the mouse, the printer, the jump-drive and other USB accessories? Clearly the story doesn't end here.

Comments, questions, feedback, criticisms?

Send feedback

• E-mail the author of the module, Man + Machine: Understanding the development of real-time interactive computing

Footer

More about this content: Metadata | Version History

• How to reuse and attribute this content
• How to cite and attribute this content

This work is licensed by eric leshinsky under a Creative Commons Attribution License (CC-BY 2.0).

Last edited by eric leshinsky on Oct 18, 2005 10:05 am GMT-5.