On 2015-09-17 06:56, Klaus Krippendorff wrote:
> carlos,
> i agree: " The very definition of random is that you can't replicate
> it or predict it exactly".
>
> but note:
> who is the you? there are a lot of behaviors that you, i, or anyone
> can't predict exactly. an inability can hardly serve as a definition
> of how a machine operates. see ross ashby's black box conception.
>
> random generators in a computer are always pseudo-random generators --
> unless they respond to atmospheric noise, johnson-myquist noise, or
> random processes from nature that we haven't figured out and may well
> never be able to. nature is what it is. it's not a machine.
>
> the randomness of throwing a dice stems from the human inability to
> replicate the act of throwing precisely and the complexity of
> accounting for the airflow that interacts with the dice during its
> fall.
>
> to accept your definition of a non-deterministic machine as one of
> which nobody can predict its exact behavior asks me to conflate how a
> machine operates and whether it can be understood by an observer.
>
> to go a bit deeper into that difference, i suggest you consult ross
> ashby's notion of a black box or read about heinz von foerster's
> concept of a non-trivial machine. the latter is quite primitive. it
> can be build but not be understood by observation.
> even the example of google's search algorithm would by your definition
> be non-deterministic. yet google folks have designed the algorithm and
> know how it operates, but nobody else has been able to predict
> precisely how it does what it does or reverse-engineer it. i would say
> most contemporary information technology has the property of escaping
> predictability while being deterministic (state-determined or
> history-determined). there is not much more to it.
>
> i am glad you like science fiction. but there is also a place on earth.
>
> klaus
Dear Klaus,
I understand your line of reasoning.
I think what you are saying is that machines can never produce
randomness by themselves, only pseudo-randomness. This assertion leads
me to 2 questions:
1) Is this true?
2) Does it mater?
1) Is it true that machines can never produce randomness by themselves?
Not entirely. That thinking is based on the flawed assumption that
machines are outside of the domain of nature. Machines are still made of
matter and still have to obey the laws of physics and quantum mechanics.
There are already applications of quantum random number generators out
there, some place on Earth.
2) Does it mater if machines can produce randomness by themselves?
I don't think so. For two reasons:
2.1) Randomness is not a necessary condition for design;
2.2) Humans are unable to produce randomness by themselves;
2.1) We were discussing if machines are theoretically capable of doing
design. The best model for an entity able to design is (currently) a
human designer. So we have to ask ourselves if the ability to produce
randomness is a necessary condition for humans to be able to design. If
it is not, then it would be beside the point to require that ability
from a machine.
2.2) In spite of the answer you might give to the above question, you
also have to ask if humans are able to produce randomness by themselves.
Though a human's biological processes possess inherent randomness,
humans are not able to produce randomness by themselves. That's why you
throw dice and flip coins: you yourself need an external source of
entropy when you want to generate randomness. Again, it follows that it
would be beside the point to require such ability from a machine.
--
Best regards,
==================================
Carlos Pires
[log in to unmask]
[log in to unmask]
-------------------------------------------------------------
Design & New Media MFA // Communication Design PhD Student @ FBA-UL
Check the project blog:
http://thegolemproject.com
-----------------------------------------------------------------
PhD-Design mailing list <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------
|