[ExI] fun outsider's view on ai

[Brent Allsop]

Hi John,

I can agree with everything you are saying, even when you say "we do 
know that a program with a million lines of code can manufacture the 
qualia 'red'".  I must admit that this is a very testable scientific 
theory that could be proven correct by demonstration.  I just happen to 
currently favor the theory that it is something much simpler, like a 
particular neuro transmitter that is responsible for an elemental qualia 
like red.  But let's go with your theory in this conversation.


OK, so something less than a million lines of code can "manufacture" the 
elemental qualia red.  I assume you will agree that a different set of 
code can "manufacture" the qualia green, and that eventually we will be 
able to know, recognize, and detect each of these and their differences 
in each of our minds.  Then we will be able to see each of these in our 
brains, and be able to tell things like if my code "manufacturing" red 
is more like your code "manufacturing" green.  In other words we will be 
able to "eff the ineffable" and know how our minds differ, 
qualitatively, or not - at least to some degree.


It is still a fact that both the words "red" and "green" do not have 
either of these "manufactured" qualities, but only are representing 
such.  So the question is, which do you interpret them as, my 
manufactured red or your manufactured red (possibly my green)?  The 
point being, that without knowing how to properly interpret them, they 
are just that: qualia absent representations that must be properly 
interpreted.  Similarly, a million lines of code can surely represent 
either my red or your red, if you interpret them in the right (or wrong) 
way.


I can agree with you that simulating neurons and dendrites is "exactly 
the [most important] point".  But you are still being blind to the 
difference between an abstract representation that represents what is 
"manufactured" and the real quality being "manufactured".  And that is 
at least a little important, too.


Brent Allsop




OK, let's go with the assumption that everything you say is correct.


On 5/10/2016 4:09 PM, John Clark wrote:
> On Mon, May 9, 2016 , Brent Allsop <brent.allsop at canonizer.com 
> <mailto:brent.allsop at canonizer.com>>wrote:
>
>     ​> ​
>     simulated neurons and dendrite synapses are surely possible, but
>     not the point.
>
>
> ​I think that's exactly the point.​
>
>     ​> ​
>     Sure a word like "red" can represent, and thereby simulate a
>     redness quality, but it clearly does not have the quality it can
>     represent.
>
>
> ​We don't yet know what all the steps in the recipe to produce the 
> subjective sensation of red are but we know the maximum size of the 
> entire cookbook. ​The human genome is about 750 million bytes but has 
> massive redundancy, run it through a loss-less compression program 
> like ZIP and it's down to 50 million bytes. About half the genome 
> deals with the brain so that's 25 million bytes or about a million 
> lines of code. So although we don't know exactly what it is yet we do 
> know that a program with a million lines of code can manufacture the 
> qualia "red".
>
> By comparison MAC OS X has 85 million lines of code.
>
>  John K Clark
>
>
>
>
>
>
>
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20160512/0099c3b4/attachment.html>