{"id":37033,"date":"2019-10-31T22:15:24","date_gmt":"2019-10-31T19:15:24","guid":{"rendered":"https:\/\/prohoster.info\/blog\/alan-kej-kak-by-ya-prepodaval-computer-science-101\/"},"modified":"2021-01-02T13:01:57","modified_gmt":"2021-01-02T11:01:57","slug":"alan-kej-kak-by-ya-prepodaval-computer-science-101","status":"publish","type":"post","link":"https:\/\/prohoster.info\/en\/blog\/news\/alan-kej-kak-by-ya-prepodaval-computer-science-101","title":{"rendered":"Alan Kay: How I Would Teach Computer Science 101","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

One reason to truly pursue a university education is to move beyond simple vocational training and instead grasp deeper ideas.<\/i><\/p><\/blockquote>\n

\"Alan<\/p>\n

Let's take a moment to reflect on this question. A few years ago, Computer Science departments invited me to give lectures at several universities. By chance, I asked my first audience, consisting of seniors, graduate students, and professors, for their definition of 'Computer Science.' They could all only provide an engineering definition. I have done this at every new place, and the results have been similar everywhere.<\/p>\n

Another question was: 'Who is Douglas Engelbart?' A few people said, 'Wasn't he somehow related to the computer mouse?' (and I found this quite disappointing since my academic community has made great efforts to ensure that the answer to this question is easily accessible after two or three clicks, and to confirm that Engelbart was indeed related to the computer mouse).<\/p>\n

Part of the issue was a lack of curiosity, partly due to narrow personal goals that were not related to learning, and partly due to a lack of understanding of what this science entails, and so on.<\/p>\n

I have been working part-time in the Department of Computer Science at the University of California for several years (essentially, I am a professor, but I do not need to attend department meetings). Occasionally, I teach classes, sometimes to freshmen. Over the years, the already low level of curiosity about Computer Science has significantly decreased (though the level of popularity has increased since computer science is seen as a pathway to a well-paying job if you can program and have a certificate from one of the top 10 schools). Accordingly, no student has yet complained that the first language at the University of California is C++!
\n<\/a><\/noindex>
\nI feel that we have encountered a situation where both the terms 'Computer' and 'Science' have been diluted by weak massive concepts to create a new term \u2014 a kind of label on jeans \u2014 that sounds good but is quite empty. A related term that has similarly been diluted is 'software engineering', which again did not utilize the most brilliant ideas of 'programming' and 'engineering' but simply combined them (this was intentionally done in the sixties when the term was conceived).<\/p>\n

One reason to actually attend university is to move beyond simple vocational training and instead grasp deeper ideas. It seems quite sensible for an introduction to the field to try \u2014 if possible with examples \u2014 to engage learners with real problems and help them understand what is genuinely interesting, important, and central in this area.<\/p>\n

First graders are thrilled when shown how a ruler over another ruler becomes a calculating tool with which they can outdo fifth graders in adding fractions. And then they will be happy to partake in the development of improved calculating tools. They have touched a real computer \u2014 a physical and mental instrument that helps us think. They have learned a truly effective way to represent numbers \u2014 more effective than what is taught in schools!<\/p>\n

They managed to connect their sound idea of 'adding' as 'accumulation' with something similar that has new powerful properties. They programmed it so that it is capable of solving diverse problems.<\/p>\n

They also expanded it. And so on. This is not a digital computer. And it is not a stored-program computer. But it is the essence of a computer. Just like the Antikythera mechanism<\/a><\/noindex> \u2014 this is basically the essence of computers and computation.<\/p>\n

\"Alan<\/p>\n

The Antikythera mechanism<\/i><\/p>\n

How far can we go and how much can we accomplish before everything spirals out of control and we get lost in abstractions? I have always been somewhat partial to the characterization of Alan Perlis<\/a><\/noindex> \u2014 the first recipient of the Turing Award, who may have coined the term 'Computer Science' \u2014 stated in the 1960s: 'Computer Science is the study of processes.' All processes.<\/p>\n

For the sake of Quora, let\u2019s not try to push this further or turn it into a religious doctrine. Let\u2019s just happily embrace the idea Alan Perlis<\/a><\/noindex>, to think better about our field. And especially about how to teach it. Now we need to look at the modern meaning of 'science,' and Perlis was quite certain that it should not be diluted with old meanings (such as 'knowledge collection') and uses (for example, 'library science' or even 'social sciences'). By 'science,' he sought to understand the phenomenon by creating models\/maps that attempt to show, 'track,' and predict phenomena.<\/p>\n

\"Alan<\/p>\n

I\u2019ve given several interviews about how the best maps and models can often fit a T-shirt, as Maxwell's equations and others do. The analogy is that there is a 'science of bridges,' even though most bridges are man-made. But once a bridge is built, it reflects phenomena, and scientists can study them; from bridges, it\u2019s possible to make models of many kinds and formulate comprehensive and useful 'bridge theories.' The fun part is that then one can design and build new bridges (I\u2019ve mentioned that there\u2019s probably nothing more enjoyable than collaboration between scientists and engineers to solve big and important problems!)<\/p>\n

\"Alan<\/p>\n

Herbert Simon \u2014 Turing Award and Nobel Prize winner \u2014 referred to all of this as 'the sciences of the artificial' (and wrote a magnificent book with the same title).<\/p>\n

\"Alan<\/p>\n

Let me provide an example. In the 1950s, companies and universities built computers with stored programs and began programming them \u2014 and there was a particular moment when Fortran came out in 1956 \u2014 which was not the first high-level language, but perhaps the first made so well that it was used in many different areas, including many that had previously been done only in machine language.<\/p>\n

All this generated 'phenomena.'<\/p>\n

\"Alan<\/p>\n

John McCarthy<\/i><\/p>\n

The history of Lisp is more complex, but John McCarthy was interested in the attempt to find a 'mathematical theory of computation' and was determined to make everything work perfectly. The eval function, which interprets Lisp, would easily fit on a t-shirt! Compared to a 'programming system' \u2014 it's insignificant. More importantly, this 'theory of computation' had a more powerful concept than Fortran! It was the best bridge idea!<\/p>\n

The compactness of Lisp allows the entire idea of programming to be grasped with just a few keystrokes on a deeper level, and contemplated at a level that seems simply impossible when looking at massive artifacts (this is one reason scientists prefer mathematics to be compact and powerful). The mathematics used here is new mathematics, as it permits concepts like 'before' and 'after', leading to 'variable magnitude logic', which allows both functional dependence and logical reasoning, while also accounting for position and the passage of time. (This is still not understood in our harsh world of situational programming.)<\/p>\n

Lisp, acting as a powerful programming language and metalanguage capable of presenting its own theory, is an example of true computer science. If you learn it along with other similar things, you will be able to think more deeply and be more responsible for your own fate than if you simply learned to program in Fortran or its modern equivalents (\u2026 this way you can get closer to programmers!).<\/p>\n

You will learn much more about the special types of designs required in computations (for example, it's usually not appreciated when computations often require going beyond the computational environment: one of the special features of stored program computations is that it is not just material for a program, but material for a completely new computer).<\/p>\n

Another reason for choosing Perl's definition is that computation is generally much more related to the creation of systems of many kinds than to algorithms, 'data structures,' or even programming as such. For example, a computer is a system, computations are systems, local networks and the Internet are systems, and most programs should represent better systems than those they are (the old style of programming from the 1950s has lasted to the point where it seems that programming should be like that \u2014 nothing could be further from the truth).<\/p>\n

The Internet is a good example \u2014 unlike most programs today, the Internet does not need to be stopped to fix or improve anything \u2014 it is more like a biological system \u2014 by our intent, than what most people think of as a computational system. And it is much more scalable and reliable than almost all software systems that exist today. This is something to seriously consider before teaching beginners in programming such less powerful concepts!<\/p>\n

So, what we need to do in the first course on Computer Science is to take into account what exactly students can do at the very beginning, and then try to stay within their 'cognitive load' to help them reach what truly matters. It is crucial to 'stay real' and find ways that are intellectually honest and suitable for those just starting their education. (Please do not teach bad ideas just because they seem a bit easier \u2014 many bad ideas are actually simpler!)<\/p>\n

Students should start by creating something that incorporates numerous important characteristics I've discussed here. It should be a system made up of several dynamically interacting parts and so on. A good way to decide which programming language to use is to simply create something that has thousands of interacting components! If not, you should find such a project. The worst thing one can do is set students on a path of too little fluency, which would severely limit their expansive ideas. This simply stifles them \u2014 and we want to nurture them, not hinder their growth.<\/p>\n

About the GoTo School<\/h3>\n

\"Alan<\/p>\n