| layout | title |
|---|---|
default |
Bio |
My name is Christoph Reif and I am (private) researcher in computer science. The word of a research als includes the word search.
I am just searching for peaces that fit together to build a picture about the truth and future of computing. Like in a jigsaw puzzle. You just put the right parts that fits together until you see the whole image.
Truth never get killed. It is like a seed in the dessert. If the rain comes, the flowers with grow. A desert bloom is a climatic phenomenon that occurs in various deserts around the world.
"What you already know, hinders you, from what I want you to know."
Some people does not seek for truth, because they believe they already know the truth about a subject. So they do not look behind stones (like Paul Graham said). They will often react, if someone came with the truth, because their mind says, this can not be true, because what I believe is the truth. If some reacts he mostly does not have the truth, but mostly protecting a lie.
Also Douglas Crockford said in his talk "The Power of the Paradigm"1 about the problem of the conformation bias. If new data comes into your brain and it is incompatible with paradigm in your head, you will reject it.
In the 70s they did a lot of invention, but did not have the hardware we have today. So computer where slow and you have to things you would not do today. These are things that can change. Because now we have the computer power and the reason why we have a lot of things is gone. So todays computer can be 10 times simpler made. The question is why you does not want to remove things that have fallen out of time. We are now 2020 folks and heading 2030 oder 2040. IT industry had to start from scretch otherwise it can not go into a sustainable direction and things get more worse.
Signs tell you if you are on the right or wrong path. Many do not believe in signs. They are still heading north despite the sign saying "Stop! Please turn around. You are going the wrong way south!" I am sure you've ever heard the saying "blind lead the blind".
Paul Graham wrote in his book "Hacker & Painter" and in his article "Revenge of the Nerds about this ongoing problem of the pointy-haired bosses.
Niklas Wirth wrote 1995 in his article "A Plea for Lean Software" about the signs of software bloat.
John Backus wrote in his article "Can programming be liberated from the von Neumann style? A functional style and its algebra of programs" about the signs.
Paul Graham wrote is his remarkable article "The Roots of Lisp" that all languages heading towards Lisp. This is also a sign that functional programming is the future.
We all have a neuronal network (a real AI) between our ears. This brain is a rule extraction machine. If there are rules in the input, the brain will find the rules. We should be careful with which data (=information) we feed our brain (=neural network). For instance this is how propaganda works. They feed you half-truths that you can't verify. Then you start believing these things as if they're true, but they're not.
Moore's law is the observation that the number of transistors in a dense integrated circuit (IC) doubles about every two years. Also CPUs get more and more cores on every generation.
John Backus already described the problem with the Neumann's bottleneck in 1977.
I believe we need real parallel and concurrent computing2. Which implies functional programming and for hardware something like a dataflow processor architecture3.