From an idea to an industry. Interview with Piotr Wozniak, the author of SuperMemo

Translated and corrected: April 1997

JM: We speak English which might surprise our Polish readers. What is the history of SuperMemo and your personal life? The program is considered something extraordinary. I also believe in its revolutionary nature. Could you start with a few words about your roots and the family? Is SuperMemo somehow related to the fact that you prefer to speak English?

PW: Indeed there are strong connections [between the idea of a universal language and developing SuperMemo] but ... this is a long story. I will just tell you that initially I had lots of difficulties with promoting SuperMemo. I had no money, I could not find buyers and I was also wary of profit-hungry investors. I did not have a father. My sister was 17 years older and brother 16 years older. My siblings acted as an extra pair of parents. My brother was tutoring me a lot on geography, math and many other subjects. This is very good for a young man's development. When I went to school I was just exploding with desire to show what I have learned. Knowledge and the possibility to demonstrate it were the highest gratification. I was mad about chemistry and biology. I wanted to study biology to understand the way the human brain works. At home I kept dozens of animals. I was collecting all knowledge in piles of notebooks. Quickly I realized that it all cannot be kept in memory with traditional methods. There was more and more material to learn. Rehearsing this all verged on an impossibility. I needed more tricks and intuition to cope with it

JM: Which year was it?

PW: This pile-up of unmanageable knowledge was most painful in 1982. In 1983, I started systematic research on how repetitions affect the ability to recall information. I was then the third-year student of biology. I knew that intervals between successive repetitions should increase. I wanted to find how much they should increase exactly

JM: Were you a good student then?

PW: No. Just the opposite. First of all I would not attend lectures. I found studying books of more benefit. Secondly, I would not spend my time on stuff that was not within my focus of interest. I would not care about the taxonomy of leeches or bryophytes. On the other hand, I was dying to know more about humans, physiology, brain, health, etc. I was all the time in conflict with teachers [about not attending lectures]. At the same time, I worked hard to gain new knowledge. That was most important. Initially, my favorite subject was ... how to prevent aging. I sensed that with some little tricks at the molecular level, we could greatly enhance longevity. At the same time, I continued my interest in memory and learning. I started even thinking about a computer program that would help this idea. When I was graduating from molecular biology, the outline of SuperMemo was ready on paper

JM: Which year was it?

PW: 1985. I graduated from molecular biology and immediately went on to study computer science

JM: From scratch?

PW: Yes. I enrolled on a full five-year program

JM: You must have been a sort of eternal student looking for a formula to keep young by staying at the university?

PW: Not a all. I could have started working back in 1985. This was not a formula for having good time. I thought that armed with computer science, I will be able to get a job in the Institute of Biocybernetics [then newly created in Warsaw]. I wanted to do Big Science. I thought about biochemistry. My work in the laboratory showed me the gap between Polish science and the best labs in the west. I would not think of emigrating. I was too patriotic then. I could not have left the system that had paid for my education. Because of miserable status of experimental science, I decided to drift into theoretical biology. I wanted to study mathematics, physics or computer science. I had lots of ideas, and I knew that with a computer you can find things that would require years of labwork

JM: Did you graduate from computer science?

PW: Yes. Five years later

JM: What happened to SuperMemo?

PW: In May 1985, before graduation, I started my repetitions. The general outline of the learning algorithm was ready. Repetitions were made on paper but after a year I started realizing that this is an extremely effective method

JM: What did you use the method for? For learning biology or English?

PW: Both. Later, when I started studying computer science, I really started getting crazy about English

JM: Was it the time when you started speaking English in Poland?

PW: Not yet

JM: So when did you get this habit?

PW: First of all, I started from being an enthusiast of one language for the world and Esperanto. I even used to write my personal diary in Esperanto. However I was disappointed with lack of scientific literature in the language. Then I entered a loop of positive feedback between English and new knowledge. I read too little scientific literature due to a poor command of the language. The more I read, though, the more I realized what an ocean of information is taken away from me. I was isolated from the world's culture by knowing only my native Polish. Add to this communist propaganda in Polish media and you will see how I was made stupid and unaware. I knew biology and some other stuff, but I knew nothing about economics, politics, sociology, etc. I could not listen to the BBC or Voice of America because I was limited to a single language. However, when I could hear the "bad boy" Reagan speaking his own words, the world seemed to have been all turned around [bad boy was not that bad after all]. I understood that the faster I and others learn English, the faster we will truly understand the world. I wasn't then much focused on SuperMemo. I was generally interested in computer modeling of biological processes. However, learning English and my growing paper "databases" continually brought the subject of learning into my attention. Ultimately, I concluded that I desperately needed a computer program to handle the SuperMemo method. I was then making repetitions of whole pages that contained material of different difficulty. Computer could manage processing each single item separately. This would greatly increase the speed of learning. In late 1987, I finally wrote the first version of SuperMemo in which all elements were classified for their degree of difficulty

JM: What language did you use to write the program?

PW: Turbo Pascal 3.0

JM: What computer did you use?

PW: I had Amstrad PC 1512. My sister helped me greatly by sponsoring such an expensive toy. In 1987, there were very few such machines in Poland! You could not buy it in a shop. It was smuggled from Germany. Nevertheless, in December that year, I had my first SuperMemo and started my first repetitions on the computer

JM: What was the size of the hard disk?

PW: There was none!

JM: To sum it up: you studied computer science in order to use a PC without a hard disk to implement your ideas?

PW: More or less. The best point was that the program quickly surprised everyone with the results you could get. I did not expect you could learn that much in such a short period of time. I liked the paper-based method but ... the computer made a world of difference. Learning English was a real hit then. You could memorized 2000 words in less than a month with minimum effort. My colleagues could see that it is only a fraction of what can be accomplished

JM: And when did it all become a business? I know you can be a genius but you cannot do everything all by yourself. You have to have money, employ people, etc. How did you try to make it into an industry?

PW: I only had the program and a couple of colleagues who quickly measured its potential. One of them was Krzysztof Biedalak [cofounder of SuperMemo World] ...

JM: Did you test SuperMemo on your colleagues?

PW: Honestly, I did not need testing to prove the value of the program. This was evident. Those who got a copy got really crazy about learning more and more

JM: In other words, you have set up a group of SuperMemo maniacs?

PW: Yes, in a sense

JM: So you optimized your technology in a closed circle of guinea pigs?

PW: Not exactly. This circle was getting wider and wider quite quickly

JM: How long did this last?

PW: By the end of 1988 the program was ready to be sold. I started looking for someone who would want to buy it. I set the price at $10,000 [at that time in Poland, you could live comfortably for years just of the interest of this sum]

JM: Ten thousand dollars for one copy?

PW: No. I wanted to sell the program and all rights to the method. The one who would buy it could do anything he wanted with the program

JM: Have you found a buyer?

PW: No. If I did, I would not be selling SuperMemo now. I was trying to find a buyer in Holland, Krzysztof Biedalak tried in England, etc. No success. I realized that there was only one option: trying to sell it by myself. One of my colleagues, Tomek Kuehn, managed to sell about 20 first copies. In 1990, I wrote my Masters Thesis about SuperMemo. Prof. Kierzkowski from the University of Technology in Poznan was the supervisor

JM: What grade did you get for your work?

PW: Best possible. I turned into a model student by then

JM: Did you start work after the university?

PW: No. Upon graduation, me and Krzysztof Biedalak wanted to study in the US (neurophysiology and artificial intelligence respectively). At the same time we were still trying to sell SuperMemo. We spoke a lot about studying in the US and about SuperMemo. Ultimately we changed our mind. We would not go to the US. We would set up a company and try to sell SuperMemo. That was 1991 and SuperMemo World was set up in July

JM: Where did you get a programming team from?

PW: There was no team. Just me, Krzysztof and a 286 PC AT. We managed to finish SuperMemo 6 but we could not find buyers anywhere. Luckily, we registered SuperMemo for Software for Europe competition and ... we managed to get to the finals. This was the beginning of success. We got a stand at CeBIT '92 in Hanover and international promotion for free

JM: What happened next?

PW: Marczello Georgiew, Tomasz Kuehn and Janusz Murakowski joined the company, but ... this was still a very hard time. We were on the verge of bankruptcy. Still it was very hard to sell a novel product

JM: So when did the real success knock at your door?

PW: There was no single breakthrough. It all came gradually. In 1991 we have not sold a copy. Today the counter stands at 7,000 copies sold and is still accelerating [May 1994]

JM: How many people are in the company now?

PW: Five partners, a secretary and a bookkeeper

JM: How many programmers do you have?

PW: There are two people working over the Windows version (including myself). All other platforms are handled outside SuperMemo World: there is one person working over the Amiga version [Twin Spark Soft], one is programming the Mac version [AS Project] and one continues the version for DOS [Zoom Design]

JM: What programming language do you use?

PW: Borland Pascal 7.0 for Windows and DOS version, HiSoft Pascal for Amiga and Think Pascal for Mac

JM: I understand that the versions for other platforms differ from the version for Windows

PW: Yes. All authors implemented their own ideas. There is a lot of cross-inspiration

JM: What are the version numbers for particular platforms?

PW: Version 7.4 for Windows is in beta, 6.8 for DOS is also in beta, 2.0 for Amiga and 1.0 beta for Mac

JM: What about shareware version?

PW: SuperMemo 5.6 for DOS

JM: And how many databases do you have in sales?

PW: About 100 databases grouped in 10 so-called database mixes

JM: What are your plans for the future?

PW: Infinite! [see: SMC Business Plan]

JM: I guess this would take an infinite amount of time to discuss! Let us move it then to the next interview. Thanks for your time

Interviewer: Jaroslaw Mlodzki, Editor-in-chief, Bajtek, Poland

P.S: Many thanks to the staff of the restaurant of Park Hotel placed over the scenic Malta lake in Poznan. Only for their kindness were we able to continue talking long after the business hours