Re: How to Cool a Planet (far fetched ideas)


Eric Chomko wrote:

How is natural language going to assist with software code reuse? Natural
language research in computers, which is at least 50 years old, is based
upon having a computer understand human language which is often
heavily context-sensitive (i.e. 'which witch is which', is the classic
spoken phrase).

Software code reuse is based upon open systems where source code sharing
is paramount. Why should someone rewrite a conversion alogrithm when many
already exist? The whole idea of no reinventing the wheel is the basis for
software reuse.

I have 2 points.

1) In software engineering terms the descriptions of a program. You
have in any software project the following documents.

Operational/Business requirement.
Software specification
Module descriptions
Code

Everything is reusable. You need NL to fetch a given document.

2) Look at the Internet. If you want a program to do anything all you
need to do is simply look and string some modules together. In

http://ipai.blogspot.com

I have developed a concept called Hilbert Space Description. This
provides for module compatibility. Take an example. We have a
rectangle. We can specify iot in terms of 2 opposite points, one points
and the length of lines, of center and one point etc etc. The arguments
of a module are described in NL which may translate into a Hilbert
description.

: > So it is not truly clear that we will be able to build robot space
: > probes that can build copies of themselves and then turn the asteroid
: > belt into space habitat or space mirrors or harvest the Oort Cloud to
: > feed humanity.
: >
: > We may have to send people up there to do those things after all.
: >
: > Fast only wins the race when slow doesn't have a huge head start. We
: > don't really know just how big a computer will have to be to power a von
: > Neumann machine.
: >
: I think there is a little bit of confusion here. Chlorella is not
: intelligent yet it is a Von Neumann machine. I think we do in fact
: understand the basic AI requirements for VN. VN based on nanotechnology
: need not have any AI. VN based on CAD/CAM needs to have enough
: understanding to assemble from a drawing. In other words assembling a
: flatpack would be a major step forward. Now everything - Furnaces,
: digging machines etc. all have a CAD specification.

: Someone in management once said to me that the distance you are "up the
: ladder", depends on how long the organization can last without your
: input. The lesson was that the lower level tasks require constant
: intervention, which top level managers should delegate. A similar

I view it as how many people can do the job you are doing? If the answer
is everyone then you're not that valuable. Face it, people complain about
baseball salaries, but if you think you're a better player than say, Derek
Jeter, then go take his job as shortstop of the Ynakees.

In space the corresponding question is can a robot do an astronauts
job, or what parts of it can a robot do?
: situation holds for space robotics. There is a double journey of light.
: At LEO all we need is telepresence. If we ever make an interstellar
: journey decision making will have to be done on the spot. Destinations
: in the solar system come between these two extremes. Something that
: could be looked at is ProEngineer with Jlink. This would enable
: commands at a range of levels to be made.


Obviously, the c factor (speed of light) changes things WRT to
telepresence. The closer we are the more interactive we can be with a
telepresence robot. In the case of things being beyond the moon, then we'd
have to store a program, that would have to interrupt itself in the case
of required human input. Interrupt programming has been around for
decades. Also, self-monitoring systems with redundancy, fault-tolerant and
failover capability all exist now and have for at least two decades.

I am very surprised that we haven't had a lunar prospector on the moon for
at least 25 years now like we have on Mars right now.

Eric

.

Relevant Pages

  • Re: How to Cool a Planet (far fetched ideas)
    ... :> language research in computers, which is at least 50 years old, is based ... :> Software code reuse is based upon open systems where source code sharing ... I have developed a concept called Hilbert Space Description. ... In space the corresponding question is can a robot do an astronauts ...
    (sci.space.policy)
  • Re: How to Cool a Planet (far fetched ideas)
    ... :> How is natural language going to assist with software code reuse? ... :> language research in computers, which is at least 50 years old, is based ... :> Software code reuse is based upon open systems where source code sharing ...
    (sci.space.policy)
  • Re: Language of a web site in ASP.NET
    ... How do you perform language selection? ... based on the user agent language preference a robot will always see ... URL rewriting allows to expose the same page as both /mysite/fr/ ... site indexers like the ones of Google don't try ...
    (microsoft.public.dotnet.framework.aspnet)
  • Possbile anyone willing to trade multi-lingual word/term dictionary databases?
    ... I am a Windows Automated Robot Script Programmer ... search engine request list of search terms in their ... native language font or characters. ...
    (sci.lang.translation)
  • Anyone willing to trade multi-lingual dictionary databases?
    ... Because some language has position each word completely different from ... I doubt how can nowadays Google foreign language do? ... I am a Windows Automated Robot Script Programmer ... search engine request list of search terms in their ...
    (alt.internet.search-engines)