UPLifting Octopi Again

Twink (neptune@mars.superlink.net)
Fri, 21 Nov 1997 18:59:16 -0500 (EST)


At 11:40 PM 11/20/97 +0100, Anders Sandberg <asa@nada.kth.se> wrote:
>> But their development cycles are shorter than many other candidates.
>> This would mean more wetware experiments can be done in the same
>> amount of time. (Of course, if we come to a better understanding of
>> the genetics involved, computer simulations might be the way to go...)
>
>The problem is that a short development cycle also may limit
>intelligence. Humans have the longest childhood of all animals
>(measured as the fraction of normal adult lifespan in the wild),
>likely because we need a lot of time to leanr and absorb our culture's
>memes. Neotony may be the way to go for uplift, but that would slow it
>down a lot.

I know, BUT octopi are already very smart without neotony AND initially,
short development cycles might be better. We can make changes and
get results quickly in the early phases of the project. This would speed
up progress for later phases. (I am assuming technology in this area will
remain roughly as is. I make this conservative assessment because I
want the experiment to happen -- not remain an armchair speculation.)

>> I agree, but do not want to get bogged down in a we-must-have-them-
>> talking-English-with-a-Brooklyn-accent-or-we've-failed attitude. Also,
>> it might be easier to increase intelligence overall and worry about
>> language later. Or the two might be related in such a way that
>> upgrading one ability feeds into the other.
>
>I never said we needed *human* language.

I was engaging in hyperbole to underline my point -- not to attack yours.

>If we could make the octopi communicate better with skin color, it
>would be a great step.

Which sounds like the route to take, since Octopi alread appear to use
color displays for expressive purposes -- in mating and territoriality.

>The idea
>is that if different individuals can communicate, then they can also
>learn more from each other and we get more help in the uplift process
>by the uplifted species itself, since it will begin to expand its
>memetic niche.

An obvious point, but nonetheless one that should be kept in mind.
And, of course, this means, hopefully, interspecies communication.

We should collect data on octopus development and genetics. We
should also design tests to measure intelligence so that we can
monitor progress. There should not only be many tests, but they
should progressively get more elaborate.. It would be ineffecient
to spend years developing octopi which were not sentient but
merely good at passing some simple test. We would learn
something, but at high cost -- part of the cost being what we
could have learned.

I would like to set up some scenarios and preliminaries for the
above. Any takers? We will need minds and money. Hopefully
enough of the former will mean less of the later is needed.

Also, because I am not a moral relativist, I would also like to
hear anyone who believes this is wrong or isn't sure speak
up.

Daniel Ust