Search This Blog

Thursday, May 20, 2010

This doesn't come as a surprise

This doesn't come as a surprise to me, but I expect that it does come as a surprise to most persons who want to believe that some day a computer program will be exactly equivalent to a human mind.

And, I expect that it does come as a surprise to most persons who want to believe that the 'Avida' computer program really does prove Darwinism. [The link I have no longer works, else I could point Gentle Reader to an instance of a doctrinaire evolutionist using just this -- then hypothetical, and to him apparently almost unimaginable -- example in a desperate attempt to evade the impact of my argument against the claims being advanced for the 'Avida' computer program.]

Slate: I'll Be Bach -- A computer program is writing great, original works of classical music. Will human composers soon be obsolete?

Now, though I've long expected something like this, and while as a computer programmer applaud the skill and expertise of writing such a program, I still think this would have been better left undone.

Understand, this can't even touch upon what differentiates a mind from an algorithm; it just offends my aesthetic sensibilities (and I’m only mildly interested in classical music -- you know, I think that the reason I detest most modern classical music is that the stuff that is lionized sounds as though it were written to a very bad formula; discordant sounds, but no music).

Still, when one considers pop music, haven't we been listening to mediocre "by the numbers" music for decades? So, perhaps, I'm wrong that this is Not A Good Thing -- perhaps the concept can be adapted to the generation of pop music and the mediocre part can be dropped out.


Crude said...

I may be missing something. Can you explain in more detail why you think this article/link would come as a surprise to someone who believes a program will be 'exactly equivalent' to a human mind? Or to someone who thinks 'Avida' really proves "Darwinism"?

Ilíon said...

Easily done, Crude; mostly because there's not much detail to explain.

The reason I expect this to be a surprise to most such folk is because so few of them *really believe* the core assertions they make ... and certainly not the implications of those assertions -- for, when all is said and done, there is one inescapable reason to believe that they don't believe what they assert: if a person really does believe an assertion he makes, then he will believe the logical implications of the assertion when he becomes aware of them.

And, if he cannot (or will not) believe the logical implications, then, if he is intellectually honest, he will try to show that the (alleged) logical implications do not follow, and, failing that, he will stop making the assertion(s). Most of these people, however, refuse to accept the logical implications of their assertions, fail to show that the person pointing out the implications made an error of logic (i.e. fail to show that the alleged implications do not follow), yet continue to assert the challenged assertions.

Ilíon said...

Hey, what do you know! I was able to find a link to the beginning of the thread about 'Avida' ... It's a very long thread, and ARN has changed their software in the meantime (which is why I didn't think I could find it), such that quoted statements don't display properly, making some posts very difficult to read.

Scientists prove evolution works

Here is the start of the particular exchange to which I alluded in the OP. Of course, to get the full effect of the exchange, one would have to follow from than post on with my response and his counter-response, etc.

But, for brevity ...
One line of my argument was to deny (on the basis of mathematics, logic, and computer science) a key assertion of the 'Avida' proponents; namely, that 'Avida' generates novelty when it is run. My argument was that every result (every possible result!) of any run of 'Avida' is fully specified by the program and its inputs.

"faded_Glory" tries to object to that using an example of comparing Beethoven and a hypothetical algorithm called "Beathoven."