Re: How Robots Will Steal Your Job

[David Longley]

A lot of the posts in this thread are interesting and entertaining <g>,
but I
have a request. The thread is cross-posted to 5 newsgroups which means
that comp.ai.philosophy is getting what seems like over a hundred posts
a
day from this thread. If you intend the material to go to
comp.ai.philosophy, that's fine - if you don't, can you leave it out of
the header?

 

[Frithiof Andreas Jensen]

Rational Rose RT (Real Time) works exactly like this - except it also deals
with threading, concurrancy etc. It is very, very good too.

product for the same reason many of these products got canceled: They
don't sell. And IMHO they don't sell because

They didn't sell because they were crap!

One had to *manually edit* the generated source code to "fill in the blanks"
thus loosing most - if not all - of the benfits of using a visual tool in
the first place; Additionally, the tools of old (i.e .3 years back) were
neither fast nor reliable and they did not integrate properly into revision
control systems.

- "Real programmers" don't like them, because they are not "macho"
enough and smell like child's play

Real programmers like/get paid to solve real problems for real people and
appreciate the productivity gained by using f.ex. Rose RT *a lot more* than
having to nit-pick through thousands of lines of crufty C++ to add some user
requirement.

Amateurs/"Language Enthusiasts" are the ones obsessing with tools and
languages; they do not last long - businesses cannot afford them!

- Non-programmers still have to learn basic things like what is a
loop, a condition, branching, structuring your data, geting the
program's logic right. So the prommis of "no programming required"
does not hold. If you are lucky you get some "no typing required"
environment.

It is still programming - you just use UML to describe the problem you want
to solve instead of coding manually - the problems are still hard, it is
just that a lot of the trivialities are handled by the tool so that one can
safely take on harder tasks than would have been practical "manually".

 

[Frithiof Andreas Jensen]

For the example of a purely imperative language that really is
designed: Algol 68, Eiffel, Oberon, etc...

Ahh - so the more obscure and 'leete the language is the "better" it is, I
See!

PS:

Has anyone actually manage to writte an application in Algol 68?
I actually do know of one in Eiffel - Market Analysis System (MAS) - but I
never used it.

 

[Programmer Dude]

Since when did understanding newsgroup posts become a foundation
requirement of intelligence? If that /were/ a requirement, I'm
guessing some people that keep posting to this thread would
probably fail to qualify :>

Here's another data point. My dog and I invented a game pretty
much out of the blue, but what's interesting is that it seems to
have distinct rules and a goal. The game might be called "Guard
the Tree" (or bush or whatever), and it consists of me playing
guard duty for a tree. Her goal is to get around me--to be between
me and the tree.

A variation involves territory. We run around an area--I try to
"herd" her into a small area, she's tries to "break the line".

The interesting part to me is how we formed the game without
discussion, but we both seem to understand the rules. Reminds
me a bit of jamming with another musician without picking a
set of chord changes.

 

[Programmer Dude]

no, or at least that's not what I meant. For that it's enough if
the animal feels pain (whose purpose is avoidance). I'm thinking
more of a 3-dimensional model of the surroundings of the animal.

I'm *pretty* sure my dog has a fairly good map of our neighborhood.
We walk twice a day, about an hour each time, and we both seem to
enjoy mixing up the route as much as possible. Point is, we cover
the neighborhood well--see every street, even the dead ends.

What I've noticed is that if we encounter something new and
interesting--a new dog or a dead animal--she'll remember it and
be looking for it when we start approaching.

 

[Programmer Dude]

The difference is the "feel". They feel like waking consciousness.
They have much more continuity than a dream. You remember just as
easily as waking consciousness, unlike dreams.

For the time being, I consider them a special form of dream, or
perhaps closer, hypnotic state.

Makes sense. Very subjective phenomena, so it's hard to evaluate.

But then most of my dreams are about debugging Java code.

Heh. I've had some *very* strange dreams after long and intense
coding sessions!

 

[Programmer Dude]

We evolved from just such creatures.

And there the key word is evolved. Our brains are a highly evolved
cognitive tool.

On the other hand, robots evolved via a totally different path and
thus there is no reason for them to share characteristics with us.

Indeed. I don't assume mechanical sentience is possible. It may
well be--and if it is, someday we'll find that out--but my opinion
right now is that it may not be.

When you postulate different mechanisms for animals and humans what
you are really expressing is your species chauvinism.

It's not so much *different* mechanisms as different places in the
same continuum (with regard to intelligence). I *do* believe that
sentience is a quantum jump. I simply don't think animals have it
(nearly all of them anyway...jury may be out on elephants and some
of the other primates).

You don't give any REASONS they should be different.

Evolution. Possibly the work of "God" (for some definition of "God";
I don't presume to have a clue about what that might be; might be as
simple as "The Force"; e.g. something that opposed entropy).

You just want them different so you can feel superior to animals
and justify exploiting and mistreating other species.

Sorry, Roedy, that's an ignorant statement wrt me. It's simply
not so. You were unhappy with people misunderstanding you; don't
make their mistake.

From a biological point of view we are very similar to chimps.

Similar indeed. And not very similar to elephants, which may be
quite intelligent. (When I see an animal species with art, history,
culture, government, law and a will and desire to communicate, I
will gladly agree they are likely sentient. Until then, I have to
see them as--on the sentience scale--below us.

That does NOT excuse maltreatment! If anything, it puts us in the
position of knowing better. That WE are sentient makes us
responsible for our actions. In fact, maybe that's the best
definition of sentience: raising above your natural conditioning.

Altruism is rare and notable in the animal kingdom. It's common
among humans.

I suspect your religious training taught that humans are not animals,
and they it is highly shameful your were born naked, in bed, with a
woman.

Again you commit an error of presumption. You are incorrect.

 

[Programmer Dude]

If anything, animals have more intense emotions (= instincts)
than we do, because they are entirely emotion (= inctinct)
driven.

Yes, but see previous post.

 

[FM]

On the other hand, there is nothing in evolution that guarantees that
any of those variables will be strictly increasing. On the contrary,
you can expect them to be bouncing all around the place in tune to
changes in the environment. I think it would be more akin to Brownian
movement than what might be called "direction".

Exactly. The point is that both the mean and the max
tend to increase with such random fluctuation, which is
the "direction" that people have been noting.

Dan.

 

[Programmer Dude]

With a new brainstorming idea, you don't stomp on it proving why
it is stupid. You look for any elements of it that can be salvaged
for a more practical solution.

Of course. But even the most outlier solutions are *in* the problem
domain. Not random. (An RNG is a bad example anyway: it's a totally
mechanical and deterministic process.)

 

[Roedy Green]

...because a large number of real-life problems were solved with Forth
despite it's lack of real class ;-p

Forth is by far the most powerful language I have ever used.

The nice thing about it is you gradually build yourself a language,
complete with custom syntax, that fits your problem domain. It then
becomes much higher level that you can hit with traditional
techniques. It is much easier to hide the plumbing.

The bad things are it is hard to understand someone else's code.
Reading code at a detailed level in very much like reading Java byte
code. You can learn to do it, but your skills go rusty if you get
away from it.

It was more important in the days when you often has to create a silk
purse out of 32K of ram. I know of nothing else that comes close for
compactness.

 

[William]

Why would they go to all the fuss of expressing pain if they were not
actually feeling it? Why do you PRESUME this elaborate deception?

More to the point, perhaps, why would they express pain (as a specific
sound, rather than an involuntary grunt) if they didn't expect that it would
do some good? That implies 1. they feel it, 2. letting the world know about
it sometimes does some good (makes you take your foot of the dog's tail,
for example). -Wm

 

[Roedy Green]

...where most modern systems are written to work around user error in
one way or another.

No. They either don't bother to verify, most common. or they simply
list a set of errors after the user has entered a formful of data.

What I was doing was informing the user on the very keystroke of any
error, and as soon as an assertion could be computed otherwise.

The user could navigate around failed assertions by rekeying erogenous
fields (the assertions have an error messing in English that explains
the problem), by changing inputs so that the flow of fields keyed
change, thus bypassing the assert.

The program guide you by the hand through only the fields it wants you
to key. What you key can be used to influence what you key next.

A simple example would be skipping over state and zip if you said
earlier you lived in Canada.

It is not like a form where you have to key everything, or skip fields
manually.

In theory a user could attempt to sneak around a programming bug that
manifested as a failed assertion by keying different data. This has
never happened to my knowledge, but in theory it is possible.
The failed assertion messages are often of the form, "If you key A you
also have to key B, or leave both blank". "No such zip code in that
state."

In pseudo java you code them like this:

assert( a.blank() == b.blank(), "If you key a, you also have to key
b");

assert( stateCrossZip( state, zip ), "No such zip code in that state."
)


Literally they would be coded:

CLEAR? a CLEAR? b = " If you key a, you also have to key b" MUST


FROM state zip STATExZIP " No such zip code in that state." MUST

They are much like Java assertions, except they have an English
explanation. When they fail, the action is to back the user up and
let him rekey the previous field, or back up even further if the user,
(or the programmer) requests, using the CULPRIT verb to indentify the
most likely culprit FIELD, not place in the code where that field was
keyed. Where you back up to could be in a routine miles away. You
don't just jump there, your restore the state you had when you were
there. You even jump back to a method YOU called, not that called YOU.
This is how jaunting differs from an exception.

 

[Roedy Green]

Has anyone actually manage to writte an application in Algol 68?

I did write a parser for it, at least for the declare statement.
It got me noticed at university. Every student failed the assignment
but one, who "cheated" by using SNOBOL. Peck, one of the designers of
Algol 68 was my prof.

The Algol 68 report must be the most snobbish document ever written in
computer science, deliberately trying to exclude the unwashed from
understanding it. It quotes from Lewis Carrol.

The designers were mathematicians. They had utterly NO concept of what
a language needs to be maintainable. Some of the things that were
legal Algol68 would make your hair stand on end.


ROSETY ROSETY ROSETY.

 

[Roedy Green]

If so, I would find it curious to learn that a dog I spend a longer
amount of time with will know what I'm thinking of.

I noticed that my dog Sheldon was more attuned to my emotional states
than people were. He had the advantage of being able to smell my
fear.

 

[Roedy Green]

Have you ever seen a shop selling Kentucky Fried Owl?

Don't they have a spotted owl franchise in Oregon?

 

[Grant Wagner]

Steve,

yes, those are the ones the system allows you to use. But when
the system becomes intelligent, it will not let you use those it
does not want you to use. This is already happening today.

The system means today's system of computers and humans, but
humans are gradually being taken out of the loop.

Humans are taking themselves out of the loop. I don't believe the danger
is from machines becoming self-aware and taking away our ability to turn
them off. The danger is from us becoming dependant on the machines and
removing our own ability to turn them off (which is what I believe you
are talking about).

However, our dependance on machines does not mean they are self-aware,
or as intelligent as us, it simply means we are dependant on them for
our survival. I don't think anyone would argue that the computers used
in apollo missions were "self-aware" or "intelligent", yet if the
astronauts had turned them off, they would have died.

All this may say something about /our/ intelligence (that our very
survival at times depends entirely upon the imperfect machines we
build), but I don't think it says anything about about the intelligence
of the machine.

 

[Airy R Bean]

A large number of real-life programs were solved by
Machine Code, by assembler, by________________ (Put
your favourite language in).

That does not mean that good practice was followed to
ease the maintenance phase when software slowly goes rusty.

If Forth was used for any significant project then the maintenance
phase will be excessively costly.

 

[Roedy Green]

the ability to predict
outcomes of new situations that cannot have been inborn or
learned, and act purposefully.

A bear cub or a lion cub without a parent never learns to hunt.

A human baby without its culture is a quite pathetic creature as well.

Attempt to preserve endangered species is zoos are doomed to fail for
some species, since the cultural learning of the species to survive in
the wild is lost.

 

[Roedy Green]

Using that criteria, a common calculator is more intelligent than your
average person, in that it can calculate the square root of a number
quicker. I do not believe that a calculator is intelligent.

The calculator is not GENERALLY intelligent, but at the problem of
computing square roots it is.


My idea is you must put the person and the calculator in a black box,
and compare their productivity. You must not allow your vanity to get
in the way of noticing that FOR THAT PROBLEM, the calculator is
smarter than you are.

To make the test fair, you need an intermediary inside the black box
who tells the problem to the person and the calculator without
revealing the judges which was which.