Computer as Material: Messing About with Time

By Seymour Papert

This article was published in the Teachers College Record in Spring 1988 (Volume 89, Number 3). The project reported in this article was carried out at The Computer School, New York City Board of Education, 100 West 77th Street, New York, NY 10024. The work presented here was aided by a grant from the Apple Education Foundation. Seymour Papert is affiliated with the Massachusetts Institute of Technology. George Franz is affiliated with The Computer School and the New York City Board of Education.

Computers began in education with a charismatic aura that cannot remain characteristic of their long-term presence. As they become part of the everyday toolbox that kids can dig into, will they have any special value?

Computers, with their power and technological sophistication, fascinate just about everyone. Teachers, parents, administrators, and students agree that computers have added an important presence and dimension to educational settings. However, within the history of education, computers represent a very recent arrival on the scene and their role has not yet begun to be explored.

An examination of computer use in schools today reveals that students' interactions with computers are largely teacher-directed, workbook-oriented, for limited periods of time, and confined to learning about the machines themselves or about programming languages. Further, computers are located in separate labs and are not integrated into the standard curriculum. "Doing computer" in school is thought of as an exciting activity in and of itself. This separation is reflected in the often asked question: "Does what children learn with the computer transfer to other work?" The present separation of computers from other curricular areas is reflected too in arguments about whether computers might even be bad for children.(1)

The project described in this article approaches the computer in quite a different sense. Instead of the familiar uses of the computer, which Robert Taylor has christened "Tutor, Tutee, Tool," (2) the computers in this project are employed in a new way, which we call "Computer as Material."

The setting for the project was a junior high school science classroom in the New York City public schools. The classroom was well supplied with various materials from test tubes, pulleys, and microscopes to scrap wood, broken electronic devices, marbles, and the like. Also present in the room were computers and LOGO. In this project, the students built devices for measuring time using any materials they wished. Some used string and a metal weight to make a pendulum, some used plastic containers to dribble sand -- and some used computers. Our central focus is this use of the computer as just another type of material.

We mention one other closely related point of interest. The phrase "messing about" in our title is, of course, taken from a well-known paper by David Hawkins.(3) Marvelously entitled "Messing About in Science," it describes how he and Eleanor Duckworth introduced children to the study of pendulums by encouraging the students to "mess about" with them. This would have horrified teachers or administrators who measure the efficiency of education by how quickly students get to "know" the "right" answers. Hawkins, however, was interested in more than right answers. He had realized that the pendulum is a brilliant choice of an "object to think with," to use the language of Papert's Mindstorms (4), one that can build a sense of science as inquiry, exploration, and investigation rather than as answers.

Just as pendulums, paints, clay, and so forth, can be "messed around with," so can computers. Many people associate computers with a rigid style of work, but this need not be the case. Just as a pencil drawing reflects each artist's individual intellectual style, so too does work on the computer.


Step one of the project was to bring the students to understand the need for the measurement of time. The teacher began by putting an empty glass jar over a lit candle and having the class watch the flame go out.(5) This was repeated several times, then the students' wristwatches were collected and the classroom clocks were disconnected.

Repeatedly the candle was lit and the jar was placed over it, and the students were asked to predict when the candle would go out. They quickly realized the need to develop a timing procedure -- such as counting their heartbeats or breaths, or counting one-Mississippi two-Mississippi, and so forth. It might take fifty-seven heartbeats or fourteen breaths from the time the jar was placed over the candle until the flame was extinguished.

After perfecting their own "body timers," the students covered their eyes and raised their hands to indicate when they thought the candle had gone out. There was much discussion concerning different methods of timing and the accuracy of each, prompting the students to defend, evaluate, compare, and evolve their individual timing systems.

The need for something more objective than body clocks was sharpened by introducing environmental influences. Without any explanation, phonograph records (first a fast rock 'n' roll, later a slow Brahms) were played as the students carried out their timing methods. Further, students reentering the room on successive days were asked to predict from memory when the candle would extinguish. While some students' predictions remained quite accurate, most were significantly off. Discussion led to the conclusion that the rate of their body timers varied from one day to the next and that they were also affected by what went on in the class, such as the type of music that was played. It thus became obvious that the next step was to create a timer that was much more accurate and consistent from day to day.

What to do? How to proceed? Different suggestions were offered, most of them fantastic and impossible, often ideas centering around the creation of complex timers such as the gear-driven type on the wall or on their wrists. After a period of discussion, the teacher suggested that this was enough talk. "Let's get to work and make some clocks," was the challenge.


The room was well stocked with materials -- in part because students were encouraged to bring in what a casual observer (and even the children) might call "junk," such as egg cartons, soda bottles, tin cans, and so forth. (The project would have been very different in a room full of only "sterile," store-bought equipment.)

The students set to work constructing timers, which took many different forms. Plastic cups taped together after having been filled with just the right amount of sand (determined experimentally) became crude egg timers. Water dripped out of a small hole in the bottom of a tin can and loudly plopped on a tin plate; the number of drops was carefully counted to tell elapsed time. A metal marble rolled through grooves chiseled out of pieces of wood; its speed, and therefore the time involved, could be controlled by varying the angle of the wood. Water slowly flowed into a cup on the up-end of a seesaw, and when the proper amount of time had elapsed, enough water filled the cup to make the seesaw tilt, at which point a piece of metal was tripped to complete a circuit that rang a bell. For students who found it hard to imagine a timer, there were library books with drawings, descriptions, and model plans of many different timing devices.

The room was also well supplied with material of a very different kind -- computers and LOGO -- which some of the students chose to use in constructing their clocks. The first LOGO timers were not really very different from the. other clocks being built. Some examples included a regularly blinking screen, or a turtle alternately moving forward and then pausing, or the computer beeping at regular intervals.

While some students were speaking the language of LOGO in order to achieve their goal of making a timer, others were speaking the language of a chisel, or of a battery and electric motor, or of a ball rolling down an inclined plane. Most of these languages were new to the children. What was important was that the students were learning to speak the languages of many different materials in the classroom in an attempt to create their clocks from ideas in their minds. When the students let their imaginations go, they found a variety of odds and ends for different explorations and investigations. The emphasis was on inquiry and learning, not on the type of material used. The computer was just one more material, alongside candles, crayons, ammeters, and rulers. The computer did, however, add dimensions not present in other materials, allowing students to go beyond the capabilities of the clocks constructed with the more commonly found materials.


While computers were used like wood, string, and electricity as material to mess about with, they evolved into something else as the LOGO timers became more differentiated and sophisticated. The students began constructing LOGO clocks that were highly accurate and precise, a goal not easily attained with the other materials. Many of the LOGO clocks became as accurate as the students' wristwatches. These clocks ranged from a second hand that moved clockwise around the face of a square clock each minute to a digital readout of hours, minutes, and seconds. Some students added a beep for each second; others printed out on the computer monitor the number of seconds as they ticked by.


The original problem of predicting when the candle would go out could be solved without using standardized units of time, that is, seconds, in the handmade clocks. As long as a clock beat with a fixed rhythm, it could be used to find out that the candle burned for, say 47 units, while another "slower" clock might have used only 27 units. Now a new question was posed to the class: "How do the units of our clocks compare with the standard unit of seconds?" This is the problem of calibration, and it gave rise to a new phase of the project. The goal this time was to relate the unit of their clocks to the standard time unit of seconds.

The precision of the LOGO clocks did not come automatically from the precision of the computers themselves. Like the other clocks, these also needed to be calibrated. It is easy to write a LOGO program that will repeat an action with a fixed period. However, calibration was still needed to make the period precisely match normal time units. The computer clocks were just like the non-computer clocks in this respect, so calibration is discussed in its general form.

Suppose you have a process that repeats about once a second and you want to adjust it to repeat exactly once a second. How would you proceed? Some of the students began by trying to adjust each individual unit to once per second. Discrepancies were easy to see when the intervals between their clocks and actual seconds were very far apart (say once per two seconds or twice per second), but when the intervals came close together, judgment could not be made by eye. Students tried using intermediate processes -- for example, clicking their fingers in time with a watch to signal one beat per second and then comparing this with the period of their own clocks. Such tricks improved their estimation but were still rather limited.

A suggestion that spurred more fruitful directions of inquiry was the idea of thinking in terms of series of cycles rather than individual cycles. Instead of trying to time a single event that took one second, students could make a test run by timing twenty of these events-which should take twenty seconds. This was quite an improvement.

Students also incorporated the concept of averaging numbers to their data. They realized that in order to ensure the accuracy of their clocks, they had to make several test runs and then average the results of the runs. Obtaining an average had never been quite so effortless in math class! This was the first time any of the students had come across the concept of statistical averaging -- using more than one trial run since it was possible that they had made a gross error on just one test try.

The connection with statistics was only one of many ways in which the work on clocks led into analytic reasoning. We saw another example in the LOGO clocks. If the clocks used WAIT 20, the program counted out the seconds too slowly. (The LOGO manual states that the command WAIT 20 pauses for one second.) Interestingly, when they analyzed the problem, the students often thought that their clocks ran too slowly because "twenty was too small," so they changed it to twenty-five. This made their timers pause for a longer period of time and therefore go even more slowly. Trial and error mingled with ample quantities of thought and discussion led them to realize that the smaller the number following WAIT, the shorter the wait. Then it was simply a matter of finding the right number by further trial and error. In the case of the non-computer clocks, too, it was not always obvious to the students which direction of change would increase the period. Should one lengthen or shorten the pendulum's string or maybe increase the weight of the pendulum bob?

In the work on calibration, careful measurement revealed that a large graduated cylinder with a small funnel in its mouth became an accurate timer as it filled up with one milliliter of water every two seconds; a battery-driven LEGO car moved precisely one centimeter in three seconds, so the distance it traveled also represented the elapsed time; a pendulum was carefully constructed with a swing time of precisely one second and was ingeniously electrically wired to blink a light bulb on each swing. Regardless of the material selected to construct their timers and clocks, the students were dealing with many of the same types of issues, such as accuracy and calibration.


A significant difference of the computer clocks became apparent in the area of extensibility. While many of the sand or water clocks were excellent timers, their use could not be extended beyond that. The computer clocks, however, were put to a variety of uses.


In one instance, the students obtained a photoelectric eye, similar to the type used by stores to signal entrance and exit. Using an electronic interface box, they plugged it into the game port of a computer, and used a LOGO command to measure the amount of light the eye was sensing. In another project, students were messing about with motion by using LEGO blocks to build cars to go down an eight-foot ramp. Wanting the cars to be fast, the students experimented with design variables, such as the size and weight of the car, the diameter of the tires, whether more weight should be in the front of the car ("front wheel drive") or the rear ("rear wheel drive").

At some point, the students who were building LOGO clocks realized that they could use their clocks in conjunction with the electric eye as timers for these LEGOmobile races. Now, different groups of students were working together, combining and expanding upon each others' projects. They placed the electric eye opposite a light bulb at the bottom of the ramp, and wrote LOGO programs to measure the amount of time it took the cars to travel down the ramp. They placed a LEGOmobile at the top of the ramp, and let it go simultaneously as they began their timer programs. When the car reached the bottom of the ramp, it passed between the electric eye and its light source. The amount of light hitting the eye momentarily decreased, causing the timer program to stop. The last number printed was the time the car took to run down the ramp.

Realizing the inadequacy of whole seconds, the students expanded the computer clocks by improving their timer programs so that tenths of a second were printed out on their monitors (by changing the number following the WAIT command). This was probably the first time in their lives they had used decimals in a real and useful way. One boy, inspired by the Olympics, even tried to print out hundredths of a second. Some of the students were encouraged to calculate the speed of the cars in miles per hour. They did this by knowing the length of the ramp and the amount of time it took the LEGOmobile to move that distance, and then converting to miles per hour. After many calculations (off the computer!) and spurred on by their own excitement and curiosity, they determined that their little cars were traveling at a rate of six to eight miles per hour.

The point of these explorations is that different groups of students had come together to solve problems in which they were interested. Some students had created a highly sophisticated timing device. Some had built the ramp, others the cars, two were experts on the electric eye, while others had written the functional LOGO programs. Some had perfected the clocks to an accuracy of tenths of a second while others had calculated the speed of the cars in miles per hour. They had all joined in a rather informal way and had worked toward a common goal. To say that the computer was the central focus of the project is to miss the point, but it is clear that the extensibility of the computer to other objects was fundamental to the project's success.


Another area of computer extensibility was seen in some interesting work done with long-term LOGO clocks used to measure aspects of the environment over periods of time. Some students, and the teacher as well, had long been concerned about the well-being of the class animals (hamsters, mice, snakes, turtles, and fish) during the cold winter nights, weekends, and vacations. Rumor had it that the schools saved money by shutting down their boilers at night, and it was feared that the animals would die or become ill as a result of the cold temperatures. We obtained a temperature sensor able to interface with a computer. The students wrote LOGO programs that instructed their clocks to print out the reading of the temperature sensor every hour, and we set up the clock programs and left them running over the weekends. Students were relieved to find out that although the outdoor temperature measured in the teens Fahrenheit, the nighttime and weekend classroom temperature remained fairly warm.

Once again, the point is that while the computer was treated as another type of material in the classroom, it did possess the power to allow a link to be formed between the students' clocks and their very real concern for their animals.


It was not necessary for the students to be fluent with LOGO to set up their clock projects. Indeed, they learned a good deal about computer programming in the process of creating their clocks and timers. For example, students had been exposed several times to the idea of variables in a computer program. (A variable is a number that changes as the program proceeds.) However, only some of the students had assimilated this concept and used it in their programming. Others had not been so quick to grasp the idea of LOGO variables.

When the students were confronted with the problem of making a timer, many of those using computers needed to find a way to represent seconds by a number that increased by one as the seconds ticked by. For the first time, they needed variables to solve a problem they were interested in. When they realized that they could solve it by using "one of those words with the two dots in front of it" (i.e., "seconds"), they understood the previously learned but not fully comprehended idea.

For most of the students, creating LOGO timers was the first time they had used computers to make programs that made connections with the physical, tangible, non-computer world. The insight that LOGO could be used to solve real-world problems was further amplified when they used their LOGO timers to determine the speed of their homemade cars and when they interfaced their clocks with the temperature sensor.



The computer clocks, as compared with those made from other materials, were unique in several aspects. First, these clocks could be extremely accurate. As we have noted, some students calibrated them to tenths of a second. This high degree of accuracy is clearly unparalleled when compared with the other types of clocks the students made, and the students appreciated this accuracy when they wanted to time their LEGOmobile's speed precisely.

Second, it was quite simple to adjust the speed of the LOGO clocks by changing the number following the WAIT command. Thus, it was relatively easy to match the handmade LOGO clock to the clock on the wall or the wrist. By contrast, calibrating the sand clocks meant ripping them apart and changing the size of the hole through which the sand flowed. Only painstaking trial and error showed exactly how much larger or smaller the hole had to be. This was true for the other non-computer clocks as well.

Further, the LOGO clocks were more adaptable and could be easily connected to other ongoing projects in the classroom. We have described the timer/ LEGOmobile connection and the clock/temperature-sensor experiment. The students took quite naturally to the integration of the LOGO clocks with these other projects.


None of this should be taken to mean that the computer and LOGO are the be-all and end-all of this type of exploration. Certainly the LOGO clocks were accurate, adaptable, and easily adjustable -- but the other clocks were wonderfully inventive, creative, and fun. They were also far more accurate than we would have predicted at the beginning of our study of time. While not as accurate as the LOGO clocks, almost all of them certainly did the job, within a few seconds, of telling their inventors when the candle would go out.

It is important to note that none of the clock media (computers, sand, wood, etc.) stood out for the students as more desirable or valuable than the others. Some of the students were attracted to wood and so made their clocks out of wood. Similarly for water, or electrical devices, or pendulums. There was no competition for who could make the "best" clock (whatever that would mean) or even the most accurate one. There was simply a classroom filled with various types of clocks being constructed, some on the computer, some out of wood, some with water, and so on. All of the students were involved with their own, individual clocks, trying to perfect them to the best of their ability and interest.


We have described what we consider to be an example of truly educational computing -- active, exploratory, student-directed learning involving the use of the computer. Through programming languages such as LOGO, computers allow our students, within certain limits, to perform tasks that are difficult or even impossible to achieve with other materials. We emphasize that it is possible to create activities that connect many different students' interests to various curricular areas, and to connect "separate" disciplines to each other. Our goal was, and continues to be, to create learning situations in which connections are allowed to develop freely and to move in any direction, albeit many or even most of them unpredicted. A certain degree of openness and flexibility on the part of both teachers and students is obviously necessary to keep the inquiry interesting, stimulating, and exciting.

Some important guidelines, then, for the placement and use of computers in schools include the following:

  1. Seek out open-ended projects that foster students' involvement with a variety of materials, treating computers as just one more material, alongside rulers, wire, paper, sand, and so forth.
  2. Encourage activities in which students use computers to solve real problems.
  3. Connect the work done on the computer with what goes on during the rest of the school day, and also with the students' interests outside of school.
  4. Recognize the unique qualities of computers, taking advantage of their precision, adaptability, extensibility, and ability to mirror individual students' ideas and constructions of reality.
  5. Take advantage of such new, low-cost technological advances as temperature and light sensors, which promote integration of the computer with aspects of the students' physical environment.

While the theme of this article has been the role of the computer in the educational process, let us clearly state that the ideas underlying our teaching strategies were formulated by educators and philosophers whose lives long predated the invention of the computer, and whose ideas can be applied to any learning situation and to any material. Our emphasis, as was that of Piaget, Dewey, Susan and Nathan Isaacs, and others, is clearly on the inquiry and the learner, not on the specific curriculum or facts to be learned. In this undertaking, all materials are created equal, although admittedly the computer did add unique and powerful aspects to the learning process.


(1) "The Computer in Education in Critical Perspective." Teachers College Record 85, no. 4 (1984): 539-639.
(2) Robert Taylor, ed., The Computer in the School. Tutor, Tool, Tutee (New York: Teachers College Press, 1980).
(3) David Hawkins, "Messing About in Science," Science and Children 2, no. 5 (1965): 5-9.
(4) Seymour Papert, Mindstorms: Children, Computers and Powerful Ideas (New York: Basic Books, 1980).
(5) See Hubert Dyasi, African Primary Science Program, Measuring Time: Part I- Making Many Simple Clocks (New York: The Workshop Center, n.d., NAC 4/220 City College of New York).