I don't think so. Humans adapt to anything. If it wasn't the von Neumann (VN), let's say, if the dataflow machines were invented before VN, we would adapt to it and build everything around it and correct its imperfectness via the peripheral components of the architecture, such as, different memory designs, IO, and other external components ... VN was the first very-practical architecture that was formulated, and it was so practical and GENERIC-USE at the time that, an incredible amount of support built around it exponentially fast. INTEL has its designs completely around VN, our Computer Architecture education system was heavily built around VN, and many other things ... For another architecture to swoop in and take over, it has to be a DISRUPTIVE TECHNOLOGY. I do not see one happening any time soon.

There have been many other proposals, dataflow machines being one of them. They just didn't have the DISRUPTIVE strength to take over this exponential growth that built around VN. Anything else became a specialty architecture (or, paradigm) and had its little small 0.1% or whatever share.

In the end, this is just the LOWEST LAYER of a computer, the ARCHITECTURE. You can correct for a lot of NOT-SO-PERFECT-nesses of this layer with the upper layers (software, etc ...). I think the GENERALITY of the VN architecture is what made it caught on so fast. Anything else is a SPECIALTY platform, which doesn't have as much GENERALITY. Also, remember. Nothing is really holding our education system back from teaching non-VN architectures.

Thanks Tolga!

I appreciate your answer, but it seems you are a technical optimist. I tend to fall into the field of social construction of technology (SCOT) where technology evolves according to the human needs more than it is exemplary for our thinking and environment.

I have tried to follow the technological evolution from analogue computing machines to the digital computing machines and marvelled about how a whole school system adopts in "what" and "how" computing is taught. Teachers of today heavily depend on what technology can offer. Therefore I do not agree on your statement "nothing is really holding our education system back from teaching non-VN architecture".

To enhance a reflection about our thinking and teaching we need to understand the historical variation of computing machines and our dependence on VN-architecture. It would be great if you, Tolga, could give me some direction of where to read about the generality of VN-architecture.

Hmmm. This is though :) My inclination would be that: "nothing can hold us back but US." So, I am skeptical about BLAMING anything, let alone the von Neumann (VN) architecture. Your "concern" seems to be broader than VN, but, rather, almost like "do COMPUTERS constrain our education." right ?

As far as the GENERALITY of VN, I didn't mean it the way you are describing. so, I won't be able to give you an answer ... I meant it in a very technical way like : "it is a GENERIC PROGRAMMING ENVIRONMENT." But, your question is approaching the issue from a different angle. I think you would ask the same question, even if there was a totally different dominant computer architecture in the world right now. Your problem seems to be at a "much higher level," like, THE COMPUTER ? right ?

True, but as far as I understand modern computers in school is built mainly on VN-architecture. I'm in search for how the architecture of VN could possibly constrain our ability to learn and teach "computational thinking". Technology itself varies, but somewhere under the technology itself, you have some principals of importance for computing literacy education.

Hmm. You might want to look at the research on HCI (Human-Computer Interface). I have a sneaky suspicion, you definitely are onto something, but, the wrong thing ! The von Neumann architecture is a small piece of the puzzle. The MOUSe , the KEYBOARD, the way we interact with the computer ... all of it is a part of the big picture.

Think about this: You are saying "I wonder if the way the cars are designed is constraining our transportation." And, you are strictly focusing on the way the ENGINE IS BUILT.

What I am saying is, you have to look at the glass, seats, steering wheel, the wheels themselves, gas tank , EVERYTHING.

apologies ... HCI=Human Computer Interaction

Amazing, I like this!

In your parallel, between a car interface and a computer interface, you are suggesting the experience of using the human-artefact-interface as a basis for understanding the constraints in education, while I suggest the experience of concepts, underpinning the evolution of artefacts, as a basis for what school can offer. Interesting ,,, Such an dualism/dichotomy seems like a research in itself; to what extent do artefact come to the fore instead of the principals in question.

To me it seems like we are holding two different approaches - the engineering approach and the philosophical approach. The question then matures into what actually matters in school and on what basis do we offer education - on artefacts or on underlying principals?

Yes, my approach is engineering, but, in the end we can philosophically argue that, there is little difference in engineering and philosophy. All of the great mathematicians 3000 years ago (Euclid, Archimedes) were philosophers, which meant THINKERS. There was no such thing as mathematics. They built the mathematics, and then Archimedes built a pump to pump the water out of a well, he was a THINKER, i.e., PHILOSOPHER.

One thing in 3000 years never changed : THE HUMAN ! So, ERGONOMICS never changed. Anything that got BUILT had to work well with the human (Ok, call it engineering) . So, you had to THINK BEFORE YOU BUILT it.

Your question is really asking : How does this computer thing we built INTERFACE with the human ? If there is any issue with its education, the reason will be the HUMAN. If we were Marsians with, say, 3 hands, 4 ears, no VISION, just telepathy (I guess :) would you ask a different question ? or, would your von Neumann question even be relevant ? The education system for these folks with different senses would be so different that, would there even be a Marsian von Neumann that came up with the initial idea ? Education is the way we absorb the knowledge from outside into our brain through our senses (vision, smell, etc ...) So, how does VN constrain it ?

If we were Marsians as you described, I guess computing and automation, still would be of our concern for the educational system. The crucial question therefore becomes, would it be possible to teach ALL Marsians the underpinning principals of computers? And if so, is there other ways to do it, instead of using the artefacts built on the discoveries of Church, Turing and von Neumann?

The actual interface (found in interaction with computers) do not always reflect the crucial principals of concern in an educational setting, as they could be hidden by the interface. I believe the architect of an artefact always embraces some non-obvious values for the students to tackle and work around, Therefore it is a challenge to teach about it ...

I'm aware of the ancient philosophers and how natural science and technology were under the same domain - natural philosophy which remained until the 1600th/1700th. The philosophy I referred to, is mainly found in genetic epistemology (Piaget) where students' experiences of concepts evolves into a conception (Sfard, 1991), which is more or less an image of different artefacts in the world. I'm glad you challenge me to embrace the engineers' view as the process of decisions and different turns have transformed the product, and could be crucial to understand the artefact's construction and constraints. The problem though, is evident as it is problematic to appreciate today's computer, as much work done during the last 50 years is hidden in the artefact/interface itself.

Hmmm. Now I am following you (I think). Being an engineer, I will always use different terminology then you, but, if something is right, we should eventually mean the same thing. I definitely believe that, there is a big problem with our EDUCATIONAL SYSTEM. Here is my view:

Somebody invents something , say, MAXWELL EQUATIONS. It doesn't matter what it is from the standpoint of our discussion. This was invented 200+ years ago. The inventor started with a key question : say, DOES THE CURRENT FLOW THROUGH a capacitor ? He invented his equations to be able to answer this question. He came up with a set of 4 mathematical equations. In the following 200-300 years, we kept teaching our kids what the EQUATIONS ARE, not, what IT SOLVED. So, the education system MISSED THE POINT. The problem is that, even the educators didn't understand what they were ACTUALLY teaching. They knew the equations, but, not the underlying PROBLEM. The saddest thing is : If we could be tele-ported back in time, and talk to Maxwell, the inventor, he could explain everything to us so simply tht, it is actually very simple. Only if everybody could go back to the source like this. Every generation farther from the inventor, the WHY gets distorted more, and all we teach is WHAT.

In the case of the von Neumann architecture, I don't know how this applies :)

I like your historical examples. I really would love to listen to Newton and his religios thinking which influenced him to a high degree. What I' trying to say is that even von Neumann held some sort of values (influenced by rationality etc), that led him to the discovery as well as Maxwell suggested his delta-equations.

When I attended Physics class, learning about Maxwells equations I became frustrated, as the educator simplified it and presented it as something whole and complete.

Something in the discoverers time influenced their thinking which led them on the right track, something we can't fully appreciate today, when teaching and learning because we all live in another century with other values..

That's precise;y what I am saying ! Bingo !