Meeting for Chapter 5 - The Ultimate Machine

[Joel Chippindale]

At the next meeting we'll be discussing Chapter 5 - The Ultimate Machine.

Continuing our every third Tuesday schedule it'll be 6:30-8:30pm on Tue Aug 6th at the Go Free Range HQ 2.0 (thanks Tom), unless of course this is inconvenient for everyone in which case we'll move it.

I've put it up on Lanyrd http://lanyrd.com/2013/understanding-computation-book-club-2/ if you want to say whether you'll make it or not.

See you there,

J.

[Chris Lowis]

Sounds great!

Any ideas for questions or activities for this meeting? Shall we make
one of these?

http://www.youtube.com/watch?v=cYw2ewoO6c4

Cheers,

Chris

> --
> You received this message because you are subscribed to the Google Groups
> "Understanding Computation discussion" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to computationbo...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[Tom Stuart]

On Thursday, August 1, 2013 10:32:14 AM UTC+1, Chris Lowis wrote:

Any ideas for questions or activities for this meeting?

In case you missed them, James Adam made some interesting comments on Twitter:

* The more I think about Turing machines, the further down the rabbit hole I go. (https://twitter.com/lazyatom/status/364133739237879809)

* Is it really the Turing machine that's universal, or is it the combination of the machine and the specific input? (https://twitter.com/lazyatom/status/364133977625333762)

* Does computation have any meaning without interpretation of input and output? (https://twitter.com/lazyatom/status/364134178981285888)

* Is space-time algorithmic? (https://twitter.com/lazyatom/status/364134424402591746)

Some of these have more relevance to the second half of the book (and y'all are just finishing the first half), but I thought I'd call them out here while they're fresh.

See you at the pub tomorrow.

Cheers,

-Tom

[James Adam]

Hey folks,

I just finished the book, which is what inspired those tweets. Here's a bit more context in case it's useful for discussion, etc

On Monday, 5 August 2013 at 10:10, Tom Stuart wrote:
> On Thursday, August 1, 2013 10:32:14 AM UTC+1, Chris Lowis wrote:
> > Any ideas for questions or activities for this meeting?
>
>
>
> In case you missed them, James Adam made some interesting comments on Twitter:

> * Is it really the Turing machine that's universal, or is it the combination of the machine and the specific input? (https://twitter.com/lazyatom/status/364133977625333762)

Later in the book, it's shown that lots of simple systems can in fact be universal, but it becomes clearer and clearer that the specific input to those systems is vital for them to be able to perform any significant computation (i.e. for them to be able to simulate a Turing machine). It's not every instance of the Game of Life that can simulate a UTM, right? It all depends on a very delicate initial condition. In Chapter 5 it's easy to thinking about the Turing machine as that set of rules, and the tape as just arbitrary input to be operated on, but later you might start to wonder whether or not there really is a hard distinction between the mechanism and the input of a machine when it comes to the actual computation performed.

> * Does computation have any meaning without interpretation of input and output? (https://twitter.com/lazyatom/status/364134178981285888)

I'm a bit obsessed with "meaning" and "patterns"; this is what studying emergence will do to your brain.
One way of thinking about my question here is: if you can't perceive the computation that's happening, is it really there? If the behaviour that you care about (i.e. reversing a string, adding some numbers) is encoded using a mechanism that you can't understand, or even recognise as information, then it's impossible to distinguish "computation" (the performance of an algorithm) from just random behaviour, right?

But if we then accept that computation *is* happening (imagine, for instance, we built the mechanism ourselves but then took a Forget-Me-Now and lost the memory of our encoding scheme and rules), what's to say that there aren't other systems performing genuine computation, but we just don't recognise the inputs, outputs or rules? Could computation be potentially happening *everywhere* that we observe patterns? And what about patterns we can't observe, because they're too vast, or too small, or otherwise alien to our senses and primate-born recognition abilities? How many infinities of possible patterns are there in the universe, and who's to say which might involved in computation, even on enormous and agonisingly slow scales?

> * Is space-time algorithmic? (https://twitter.com/lazyatom/status/364134424402591746)

And so, about halfway down the infinite rabbit hole, you might start to think: maybe computation *is* happening everywhere, and we are ourselves just intermediate results of some vast sub-quantum-level Turing machine operating under the universe, diligently popping us from state to state. And maybe it's Turing machines all the way down, our universe just being run on another, vaster, faster Turing machine, and within our universe, all these other unseen computations simulating their own independent vast, exponentially-more-agonisingly-slow universes, and all the way up too, Turing turtles stretching forever in every direction…


I'm normally about 4 pints in before I start jabbering about this, so I'll stop now and spare you all my insane ramblings.

-- James

[Joel Chippindale]

Plenty of food for thought, you should fly over for the next one James.

See the rest of you who are coming with your burning queries, wide eyed wonder and/or big boxes of lego in about an hour

J.

[Joel Chippindale]

Just a quick reminder of James's questions from last month which I think are particularly relevant to chapter 7 'Universality is everywhere' which we'll be discussing tomorrow.

J.