Re: Moving Sedna (was USA urges scientists to block out sun)

On 13 Feb, 01:06, Willie.Moo...@xxxxxxxxx wrote:
You only need weak for a VN machine.

That is true, you only have to be as smart as a genome.

In fact to achieve a VN swarm you
need to be able to produce a series of processes whose NET input is
sunlight ans space material.

This is true if you don't want something comlex like chocolate mousse
or champagne.

A VN machine BASED ON CAD/CAM will make a rage of items. In fact for a
sunschield all you need is aluminium foil - Lots and lots of it. This
is one reason I am a little bit cool on nanotechnology. A biological
type organism tends to only be able to reproduce itself.
This can be done simply by putting the
available processes into a database and thereby configuring your
swarm.

Well, that's a step toward strong AI. Wouldn't it be easier if you
said, you know swarm I think I'll have fried chicken tonight with
mashed potatoes and southern gravy along with green beans and an ice
cold beer for dinner tonight.

It is a step towards AI. I don't know about strong. If you are doing
the same thing with subroutine modules - Well that might well get
intelligent.
A flatpack assembler clearly makes many more proocesses
available. In fact the processes that would become available put you
within striking distance of your replicator.

Strong AI makes it possible to take raw materials and process it into
60,000 or so consumer goods and make them available on demand to
everyone as they desire them without a whole lot of management while
keeping general goals in mind, like, don't hurt the Earth, don't kill
anyone, and stuff like that.

That is, strong AI makes the system safe, reliable, and useful.

No, no Strong AI is very dangerous. Once of the characteristics of
organisms, and a VN machine is an inorganic organism is that they
evolve. If the genome is stored in a Reed Soloman code Evolution
effectively stops.
2) desktop hot fusion

I don't think you need that and anyway there is quite a lot of
radioactivity even if it is considerably less than fission.

Why not? You need to power your products, and aneutronic fusion on a
desktop at power levels equal to that of todays engines would be
ideally suited for conditions found on Sedna.

And you have radiation only if you are foolish enough to use
deuterium, sure. But aneutronic fuesion is very clean.

Lets assume you use He3. You get protons. There is always gama though.
Basic He3 does indeed give totally stable products. There will still
be a little though. This is in fact where I depart from the
mainstream. They are all for Deuterium anf tritium. They are not
looking at He3.

http://en.wikipedia.org/wiki/Aneutronic_fusion

And if we're wishing for things we don't have, we might as well wish
for the good stuff, not the nasty stuff, wouldn't you agree?

3) self-replicating machine systems

Weak AI is needed.

Yes, genomes are smart, I'll give them that. But, it would be nice
to solve one problem instead of having to solve thousands wouldn't
it? I mean, we're wishing again right. Why wish for weak-AI, the
bare minimum needed, when strong AI can be wished for just as easily?
And strong AI gives you the ability to set goals and have the AI unit
figure it out. In fact, figure out strong AI -ONCE - and then ask the
strong AI to solve your other problems for you.

If it was guaranteed to solve MY problems.

If you have strong AI you must have everything
weak.

True. But if you're working on weak AI, and have success, why the
hell change your focus and add a lot of other problems, when you can
just continue to make your AI strong and then ask IT to solve the
other problems for you.

That is one possible route.
It is impossible to talk about "robot rights" and have sea
levels rising.

Robots have no rights, and if we're smart, they never will. Naturally
evolved systems have manifested and unmanifested potential. We build
tools to help manifest the unmanifested potential. To the extent the
tools assist with this process, they are a benefit. To the extent
they do not, they are a problem. As creators of these tools, we have
every right to limit their function to useful functions. Since they
are engineered and not evolved beings - they will give us no argument
about this.

We are getting into philosophy here. I believe that part of free will
is indeed the fact we have evolved. In the garden of Eden we would not
have been free to eat the fruit anyway. The point I was making however
is that strong AI encompasses weak and you cannot postulate strong AI
and count the meters sea level will rise by.


- Ian Parker

.

Relevant Pages

  • Re: Some more Bull****! Whats the 1.2-5% dna Diff! (Chimp to man)
    ... Nearly all of these segments are also conserved in the chicken and dog ... You dont like rat, then try chicken! ... Preferrably one that is not found in the chimpanzee genome. ... since it's not an organism. ...
    (talk.origins)
  • Re: How can the evolutionary progress from slime to humans be linear?
    ... Uno Lapideus wrote to sci.bio.evolution on the 01/12/2005: ... > organism. ... > the genome *per year*, ...
    (sci.bio.evolution)
  • Re: Replicating What?
    ... > replicated is the blueprint for the structure which will be ... blueprint because it does not contain structural plans or any where near the ... information needed to build an organism. ... of the genome with existing cellular machinery encodes specific RNA ...
    (sci.bio.evolution)
  • Re: Characterizing complexity
    ... > complexity ultimately arise wholly from the genome, ... > time sequential development of the individual organism from ... While understanding the historical evolution of these ... the genome does not encode functionality at all. ...
    (sci.bio.evolution)
  • Re: Please keep your *RETARDS* in restraints. (invariable genome-retry)
    ... organisms don't evolve at all. ... Populations evolve. ... ignoring mutation to their genome) ... Even adaptation (different from evolution) would seem to be beyond a minimal cell. ...
    (talk.origins)