You say:

This is the part I disagree with. Variables, functions, comparisons,
computational complexity, stacks, memory, byte code, loops, iteration,
control flow, data structures, data types... These are all mathematical by
nature, and the list goes on.

 

I believe that the implementation of your whole list is CS and not Math.
Variables in CS are memory locations that appropriate functions can work on.
Is that the definition of variables in Math?  The concept of a stack might
originally have been Mathematical (I have no information on this) but the
implementation of stacks in computers is specifically a CS variation.  Most
stacks are used for function return and local storage, does that have
anything to do with Math?  A data structure to me is an ordered group of
variables that have a set offset from the start of the memory location they
are stored at.  That doesn't sound like Math to me.

 

You did say that Math is what we do which is a conclusion and I didn't see
any supporting evidence.  There is a difference between boring everybody
with 10 supporting arguments and providing none.

 

You DID ignore my points and if you agreed with some of my points, you could
have said so.  You didn't.

 

but the end result is always something that can be described using a
mathematical construct

 

This comment can also be said for describing a function in English.  Does
that mean that CS is English because it can be described in that language?

 

I appreciate your ending remarks BUT I don't believe computers were designed
based on a Turing machine and if it was, Math didn't create what constitutes
a Turing machine.  I haven't researched it recently but I know that many
techniques were tried before our current binary computers were created that
had one memory that contained both programs and data.  There were machines
made that were decimal based.  Others had memory for data and the programs
were made using a plug board, like the old telephone exchanges. Other
computers were actually analog.  Even the difference in design between CISC
CPUs and RISC chips is evidence based rather than theoretical.

 

Our current CPU architectures exist because they survived the Darwinian
evolution of what works best in the real world rather than some theoretical
model.  CS is ultimately based on what our CPUs will execute and our
inventiveness in translating ideas into a form the CPUs can execute.

 

David Clark

 

From: Aaron Hosford [mailto:[email protected]] 
Sent: January-08-13 1:42 PM
To: AGI
Subject: Re: [agi] Why Logic & Maths Have Sweet FA to do with Real world
reasoning

 

Obviously you didn't get my point about the English at all.

 

I'm pretty sure I did, actually. Maybe you missed mine.

 

  I extensively used Mathematical formulas in my 2 x third year Econometrics
courses at University but not many people would call Economics, Math.

 

Nor would I, anymore than I would call CS math.

 

If you don't understand my argument by now, I am sure any more examples by
me won't let you see the light.

 

I don't need more examples, because I understand what you're saying and
agree with you, for the most part. 

 

 Mathematics in some ways is like a language that describes things that
belong to other areas of study.  Computer Science is one of those areas.
Some Math has been used to describe some things in CS but that doesn't mean
that CS is a branch of Mathematics.

 

This specifically is what I was agreeing with, in its entirety.

 

 I have given some examples already that show that CS has many techniques,
organizations and priorities that are not Math.

 

I must have missed these.

 

 If you wish to win this argument, then you must show why my points are
incorrect.  It is not legitimate to just ignore my points.

 

It wasn't an argument from my perspective, and I didn't ignore your points.
I added a point of my own.

 

Mathematical constructs are NOT the basis for what we do! 

 

This is the part I disagree with. Variables, functions, comparisons,
computational complexity, stacks, memory, byte code, loops, iteration,
control flow, data structures, data types... These are all mathematical by
nature, and the list goes on. Even many of their names provide clues to
this. Computers are designed based on the Turing machine, which is also a
mathematical construct. I would consider these things together to be the
basis of what we do. Basis != totality, of course. When I decide to write a
particular function, method, or class to do a certain thing, my decision is
non-mathematical, and the act requires non-mathematical expertise, but the
end result is always something that can be described using a mathematical
construct, if necessary. This doesn't make me a mathematician, nor does it
make a mathematician a programmer.

 

There are rules to arguments.  Just stating conclusions is not an argument.
Intelligent people have a duty to each other to debate issues using facts
(points that are not in dispute) and logic and if you want to argue with me,
you actually do need to follow the rules.  If normal argument rules aren't
followed in normal or average discourse, that doesn't mean that is
acceptable between intelligent people.

 

 PS This is supposed to be a list of people who exchange ideas about
creating Artificial General Intelligence, how can intelligence be created
when the simple rules of intelligent debate aren't followed

 

Sorry, I'm not going to try to dump every shred of evidence or support I
have into a single email. No one would read it, and it wouldn't be worth the
effort. I made points, you disagreed with them. I've provided additional
support for my points as a consequence, and attempted to clarify in the
process, and this exchange can be repeated. That's the normal flow of
conversation, AFAIK. If you disagree with what I've said, you're welcome to
support your own points, as well. But just to be clear, I'm not trying to
negate the points you've already made, I'm trying to refine them.




-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com

Reply via email to