> On 19 Jul 2019, at 10:51, Telmo Menezes <te...@telmomenezes.net> wrote: > > > > On Thu, Jul 18, 2019, at 20:09, John Clark wrote: >> On Tue, Jul 16, 2019 at 11:31 AM Bruno Marchal <marc...@ulb.ac.be >> <mailto:marc...@ulb.ac.be>> wrote: >> >> >> A Turing Machine is compatible with both pure mathematics and pure >> >> physics, but Lambda Calculus is compatible only with pure mathematics. >> >> Why? >> >> Ask Alonzo Church the inventor of Lambda Calculus who admitted it's true, >> and so did Godel. >> >> > The so called LISP machine implements combinators and lambda expression >> >> LISP machines were just Turing Machines that incorporated common subroutines >> used in the LISP language in HARDWARE to enabled them to run faster, but by >> the early 1990's microprocessors had gotten so fast that cheap home >> computers ran faster than any dedicated LISP machine and that's why nobody >> makes them anymore. > > Nobody ever used the Turing Machine as an architecture for computation, > outside of theoretical domains. Not even Turing himself, for the simple > reason that it would be terribly inefficient. Even though it is possible to > build a finite Turing Machine, it was always meant to be a model of > computation that made it easy to talk about its theoretical aspects. > > Computers to this day mostly follow the Von Neumann architecture, with its > familiar CPU, memory unit, I/O bus, etc. Due to economic effects related to > production at scale, it is very hard to change the underlying architecture of > general computation once one takes hold. The fact that we us Von Neumann > machines instead of LISP machines or connection machines is probably just a > historical accident more than anything else. > > Things are now starting to change, due to the adoption of Graphics Processing > Units (GPUs), first for real time 3D rendering (computer games) and then for > Machine Learning tasks. GPUs are in fact a different architecture, highly > parallelizable, while Von Neumann is ultimately sequential (the illusion of > parallelism is created by the Operating System). Modern computers, including > the one you are probably using, are already hybrid-architecture, with > parallel Von Neumann machines (multiple CPUs) + GPUs. If you play a modern > computer game, you will be enjoying the collaborative efforts of these two > (three?) architectures. > > It seems clear to me that Turing Machines, Van Neumann Machines and GPUs are > just implementations of something which is purely abstract -- computation. > > You insist that nobody has been able to produce a computer without using > matter. I agree. What you refuse to consider is the possibility that matter > is the dream of computations, and not the other way around. Whatever we are, > it seems clear that we are bound to perceive reality as made of matter, but > it doesn't follow that matter is the ultimate reality. This is just Plato's > Cave with modern language.
You are right. I would say, contra Clark, that today a lot of work is concentrated to implementing the lambda calculus into graph reduction machine, making them directly related to natural physical implementations. This allows to keep the functional approach to computation, respect a big amount of the so-called “functional abstraction” which makes easier to specify the denotational semantics of the programs. The Lisp machine was born with such idea at its roots. The Turing machine can be said more theoretical than Lambda calculus, and their close cousins: the combinators. But the Turing machine are pedagogically very simple, although, as I have illustrated with the combinators, much less tedious to be used to prove the Turing universality, although it has become the most used in textbook, not because they are more physical or natural, but more close to the human, and *that* is what convince Gödel of the Turing-Church thesis. Bruno > > Telmo. > >> >> >> >> >>I am telling you that matter is needed to make that happen, in this case >> >>the matter in the microprocessor of the computer that is running the video >> >>game that is using Bitcoins as money. >> >> > But why? >> >> Why what? >> >> > Why to make that assumption, >> >> What assumption? >> >> >> Consciousness? What the hell does that have to do with the price of >> eggs? >> >> > You are the one saying that we need matter for a computation to happen >> >> Because every computation ever observed in the history of the world has >> required matter. >> >> > (and I infer “to support genuine consciousness”). >> >> And every time in the history of the world a change in consciousness >> resulted in a change in the physical state of a brain and a change in the >> physical state of a brain resulted in a change in consciousness. >> >> > If not, then it is even more weird why you want for matter, given that the >> > computation are realised in arithmetic, >> >> And not once in the history of the world has anyone observed a computation >> being made in nothing but a change in arithmetic. In fact nobody has ever >> observed a change in arithmetic period. >> >> >>> >>> the whole video game is executed through pure number relation >>> >> >> Incorrect. The whole video game is executed through voltage differences >> >> in the microprocessor. >> >> >> > You can implement it, >> >> You've got it backwards. The numbers don't emulate the voltages in the >> microprocessor, the voltages in the microprocessor emulate the numbers. >> >> >> >> We can use the language of mathematics to help us understand how those >> >> voltage differences effect each other, and we can if we wish interpret >> >> those voltage differences as numbers. >> >> > In your theory which assumes a physical universe. >> >> The only thing I assume is that if something works then it works and if >> something doesn't work then it doesn't work. Making calculations with the >> help of matter works, making calculations without matter doesn't work. >> >> And that is your cue to refute what I just said by referring to a textbook >> that will never be able to calculate 2+2. >>> > See the combinator thread for a precise disproof of this. >> >> Ah yes, that legendary post >> >> >Ad hominem. Boring. >> >> What's boring is your referring to posts that don't exist, your constant >> whining and using that incredibly pompous Latin phrase. >> >> >> post of yours that plugs all the holes in your theory and proves that >> >> everything I've said is wrong, the post that you've been talking about >> >> for the better part of a decade, the post that NOBODY HAS EVER SEEN. >> >> > I just said that I have proven that the giving of the lambda expressions >> > [x][y]x (which does the same job as K) and [x][y][z]xz(yz) >> >> I agree, "[x][y]x" does indeed *do* the same job as "K) and [x][y][z]xz(yz)" >> because both ASCII sequences *do* precisely NOTHING and 0=0 so they both >> *do* exactly the same thing. Nothing. >> >> >>The logical operation of every computer ever made can be reduced to a >> >>Turing Machine. >> >> >True but irrelevant. >> >> How in the world is that fact irrelevant?! >> >> >> > Actually it makes my point, but usually, thanks to our physical laws (and >> > transistors) the boolean operation will be used to simulate a Turing >> > machines. >> >> Boolean operations don't simulate Turing Machines, Turing Machines simulate >> Boolean operations. >> >> >> Ironically to rebut my accusation that you keep changing the meaning of >> >> "Aristotle theology" you introduced the concept of "Aristotle's second >> >> God"; I've never heard anybody mention that before, but I admit you know >> >> more about Greek silly ideas than I do. >> >> > The first God is Aristotle first mover it is [...] >> >> Bruno, I did ask you not to tell me, I've given up keeping track of your >> constantly mutating definitions of common words and invented phrases and >> acronyms used by nobody but you. >> >> John K Clark >> >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Everything List" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to everything-list+unsubscr...@googlegroups.com >> <mailto:everything-list+unsubscr...@googlegroups.com>. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/everything-list/CAJPayv2EDrpjXzYoKKGjJkFytDAy2oRRPUktDMmgH2ZRs6WW_A%40mail.gmail.com >> >> <https://groups.google.com/d/msgid/everything-list/CAJPayv2EDrpjXzYoKKGjJkFytDAy2oRRPUktDMmgH2ZRs6WW_A%40mail.gmail.com?utm_medium=email&utm_source=footer>. > > > -- > You received this message because you are subscribed to the Google Groups > "Everything List" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to everything-list+unsubscr...@googlegroups.com > <mailto:everything-list+unsubscr...@googlegroups.com>. > To view this discussion on the web visit > https://groups.google.com/d/msgid/everything-list/2265f50b-f02e-445c-ba40-981a7fadac1e%40www.fastmail.com > > <https://groups.google.com/d/msgid/everything-list/2265f50b-f02e-445c-ba40-981a7fadac1e%40www.fastmail.com?utm_medium=email&utm_source=footer>. -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/6A0AFF80-59A0-4E99-AFEE-DC21E0D74939%40ulb.ac.be.
