RE: SUO: Re: Ballot Comment - 3D versus 4D.
>Pat and Adam,
> . I agree there is a need for a formalisation for this
>principle. What's more, you are correct in implying that this is a very
>important aspect.
>
> . Unfortunately, I am not up on the logic practices well
>enough to be able to suggest which formalisations comply with current
>standard approaches,
The problem with 'approximation' is that there really arent any
standard approaches. Many people have commented that we need to be
able to formalise this cluster of ideas, but there are very, very few
ideas on how to actually do it.
> and would need to put quite a few hours into studying
>the area in order to come up with one. I'm afraid I just don't have the time
>to do this, much as I would, in an ideal world, love to take it on and be
>employed in such an area.
>
> . All I can volunteer at this stage is that I see this as
>another of the principles I see as part of a sound approach, along with my
>recent contributions about metaphor and breaking categorisations down into
>orthogonal conceptual structures. (I'll try to find the contributions, and
>attach them at the bottom, separated by double rows of +++++++++++++s.)
>
> . Basically, my thoughts about information management over
>the decades have come to a view that we can usually make most progress by
>exploiting the principles and devices human- and animal-kind have devised
>and evolved over the past few million years. (Actually it's much longer than
>that, as some of them I observe in reptiles and even arthropods.) These have
>evidently already withstood considerable tests of time, and so are
>demonstrably quite robust.
Quite. But we don't know what those principles and devices ARE, is
the problem. Obviously people are able to think reasonably clearly
about approximations and tolerances and so on (although in fact they
are not very good at it without training, and tend to make systematic
errors). BUt we have no idea how we are able to do that, and as far
as I can see, no way to find out. Certainly introspection is not the
slightest use.
>
> . The ones I mention above are some of the more useful ones I
>see for this stage of development. Others will be appropriate in the more
>esoteric areas I hinted at below, but also include the evolution of emotions
>and emotional responses.
>
> . So, I suggest the best approach is to mimic how the
>principles I suggested are most consistently, logically and successfully
>implemented among various societies of the genus of Homo Sapiens Sapiens.
If we knew how to implement H.S.S., AI would be a breeze. It isn't, however.
> . Incidentally, with a bit of thought, you should be able to
>see from this why I was such a strong advocate of a modified and restricted
>version of English for the project. That's also part of the same principle.
I fail to see the connection. Why do you think that using English
will provide any insight into how our minds work?
Pat
---------------------------------------------------------------------
(650)859 6569 w
(650)494 3973 h (until September)
phayes@ai.uwf.edu
http://www.coginst.uwf.edu/~phayes