"I, Robot" The Goosebumb Effect
The Intrinsically Flawed Mr. Hole
only film this summer to give me goosebumbs, multiple times! With a big
action film that's one of the most important aspects. Bravo Alex
Proyas, you continue on your journey towards director Godhood. During
the climactic battle scene of "I, Robot" I literally had my jaw gaping
open. So a big "BOO!" to anyone that's come down on this film.
..
Mr. Hole
PAVELB1
I've had Goosebum....spread over crackers. Good stuff.
Mike Harrison
The problem with this film is that it's set only 30 years in the
future. No way will the Chicago of 2035 look like that. And robots
will in no way be advanced enough by 2035 to even come close to the
ones in the film.
Other things that bothered me: Where is that ruined bridge from? I
don't think Chicago had any bridges that look like that one. Also, are
there "What You See From Here #" plaques in the Chicago area? I don't
remember seeing any when I was there. You'd think they'd be there
today if in 30 years they are up to #202.
Franklin Hummel
news:11701-40F...@storefull-3272.bay.webtv.net...
> Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
> only film this summer to give me goosebumbs, multiple times! With a big
> action film that's one of the most important aspects. Bravo Alex
> Proyas, you continue on your journey towards director Godhood. During
> the climactic battle scene of "I, Robot" I literally had my jaw gaping
> open. So a big "BOO!" to anyone that's come down on this film.
It's too bad it doesn't really have a fucking thing to do with the book it was based on.
They should have used Harlan Ellison's script.
What a waste.
-- Franklin Hummel in Boston, Massachusetts
Michael Johnson
wrote:
>holef...@webtv.net (The Intrinsically Flawed Mr. Hole) wrote in message news:<11701-40F...@storefull-3272.bay.webtv.net>...
>> Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
>> only film this summer to give me goosebumbs, multiple times! With a big
>> action film that's one of the most important aspects. Bravo Alex
>> Proyas, you continue on your journey towards director Godhood. During
>> the climactic battle scene of "I, Robot" I literally had my jaw gaping
>> open. So a big "BOO!" to anyone that's come down on this film.
>
>The problem with this film is that it's set only 30 years in the
>future. No way will the Chicago of 2035 look like that. And robots
>will in no way be advanced enough by 2035 to even come close to the
>ones in the film.
I don't buy this. It seems like a copout by someone who has no clue
the speed of technology. It's 2004.. that's a 31 year speration. Taken
in reverse you get to the year 1969. Compare the technology from 1969
to that which we have now. Personal computers didn't even EXIST in
1969, there was no such thing... now you and millions of others run
one that is probably multi-gigahertz along with hundreds of megs of
memory. Its the same quantum leap in computing in that time span that
is easily conceivable happening from 2004-2035 in robotics, AI, and
mechanical engineering.
-MJ
Caden
> On 19 Jul 2004 11:26:05 -0700, harri...@home.com (Mike Harrison)
> wrote:
>>holef...@webtv.net (The Intrinsically Flawed Mr. Hole) wrote in message news:<11701-40F...@storefull-3272.bay.webtv.net>...
>>> Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
>>> only film this summer to give me goosebumbs, multiple times! With a big
>>> action film that's one of the most important aspects. Bravo Alex
>>> Proyas, you continue on your journey towards director Godhood. During
>>> the climactic battle scene of "I, Robot" I literally had my jaw gaping
>>> open. So a big "BOO!" to anyone that's come down on this film.
>>
>>The problem with this film is that it's set only 30 years in the
>>future. No way will the Chicago of 2035 look like that. And robots
>>will in no way be advanced enough by 2035 to even come close to the
>>ones in the film.
> I don't buy this. It seems like a copout by someone who has no clue
> the speed of technology. It's 2004.. that's a 31 year speration. Taken
> in reverse you get to the year 1969. Compare the technology from 1969
> to that which we have now.
Seriously? It's exactly the same. Slabs of silicon have been fine tuned
for the past 40 years. Shrink die, add transistors, battle heat. Look up
Moore's Law.
Quantum computing, on the other hand... or even that RNA mass
multiprocessing stuff some scientist are harping on...
> Personal computers didn't even EXIST in
> 1969, there was no such thing... now you and millions of others run
> one that is probably multi-gigahertz along with hundreds of megs of
> memory. Its the same quantum leap in computing in that time span that
> is easily conceivable happening from 2004-2035 in robotics, AI, and
> mechanical engineering.
Also look up Minsky... head of MIT computing department. He made
similarly embarassing boasts back in the sixties. Choked on his
own words mid-80's. Now everyone has renamed it "neural networking"
or "expert systems" to forget the whole mess and move on. :)
Love Rhino
"I, Robot"
> >The problem with this film is that it's set only 30 years in the
> >future. No way will the Chicago of 2035 look like that. And robots
> >will in no way be advanced enough by 2035 to even come close to the
> >ones in the film.
>
> I don't buy this. It seems like a copout by someone who has no clue
> the speed of technology. It's 2004.. that's a 31 year speration. Taken
> in reverse you get to the year 1969. Compare the technology from 1969
> to that which we have now. Personal computers didn't even EXIST in
> 1969, there was no such thing... now you and millions of others run
> one that is probably multi-gigahertz along with hundreds of megs of
> memory. Its the same quantum leap in computing in that time span that
> is easily conceivable happening from 2004-2035 in robotics, AI, and
> mechanical engineering.
>
You got some good points, but we don't even have functional robots as was
seen in those 60's sci-fi flicks. And, the movie 2001 hasn't really
panned out either.
I side with the former poster's opinion stating that robot technology
will not advance far enough in 30 years to look like I, Robot.
My reasons lie primarily with the brains pre-cognitive functions which
have not come close to being mimicked in computers: image recognition,
sound, voice. Sure, we base security systems off them, but its been
shown that they don't even consistently work in controlled laboratory
settings, yet alone a real world setting. And, we haven't even touched
on AI yet.
Then there's the 'non-brain' aspects. We don't have a sufficient
and compact enough powersource to power these robots, and we don't have
the proper motor actuators. Pistons and gears are just too big and klunky.
(To be sure, Japan has made a soccer team of humanoid robots, but they
are cumbersome as hell.)
I think the next big break through for human-like robots will not come
from the 'brain' aspect, but from the power source and the motor actuators.
In development are artificial muscles but there's still a long way to
go for these things to be ready.
So my prediction is: 110 more years. Hopefully I'm too conservative.
That way, I can soon order one of those "basic pleasure models" from
BladeRunner.
Ken from Chicago
"Mike Harrison" <harri...@home.com> wrote in message
news:471417c6.04071...@posting.google.com...
> holef...@webtv.net (The Intrinsically Flawed Mr. Hole) wrote in message
news:<11701-40F...@storefull-3272.bay.webtv.net>...
> > Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
> > only film this summer to give me goosebumbs, multiple times! With a big
> > action film that's one of the most important aspects. Bravo Alex
> > Proyas, you continue on your journey towards director Godhood. During
> > the climactic battle scene of "I, Robot" I literally had my jaw gaping
> > open. So a big "BOO!" to anyone that's come down on this film.
>
> The problem with this film is that it's set only 30 years in the
> future. No way will the Chicago of 2035 look like that. And robots
> will in no way be advanced enough by 2035 to even come close to the
> ones in the film.
Technology advances exponentially.
The further you go the faster the rate of change. There have been far
greater technology changes in the past century than in the previous two
centuries than in the previous millennium.
--In LESS than 7 decades people went from powered human flight to landing on
the moon.
--In LESS than 3 decades people went from steam and coal power to atomic
power.
--In LESS than 2 decades wireless communications has gone from requiring a
device larger and almost as heavy as a brick attached to a suitcase or a car
to something that is smaller than the palm of your hand.
--In LESS than 1 decade cell phones have gone from being the novelty and
status symbol of the rich to being some common as it's pert near impossible
to find pay phones anymore. How often do you walk down the street sing
someone talking on a cell phone or in a grocery store and every other person
is talking on one deciding what to buy, how much, ending the days of "Honey,
I *thought* you said to get ...." In "Third world" countries, cell phones
are increasingly used since laying out and maintaining land lines is
prohibitive.
--In LESS than half a decade, construction of the Sears Tower was
completed--back in 1974. Building the various skyscrapers seen in I, ROBOT's
Chicago of 2035 with more robot construction workers would only be faster.
As far as the development of robots, the key is data storage, processing
speed and sensor resolution:
--Sensor resolution: higher resolution results in greater detail, less
blurry picture of the world that a robot sees until its vision matches and
surpasses that of human vision. Right now, basic motion detectors can open
doors, activate lights, sound alarms. With increasingly advanced video
sensors a robot could learn to tell that "blob" in front of it is a person
and not a horse, is an adult and not a child, is living and not a manikin.
--Processing speed: A sufficiently high proessing speed would allow a robot
to INTERPRET sensor data in something close to human and eventually
superhuman speeds. Those two long lines on the ground with a series of short
lines between them . . . ah, it's railroad tracks. The round thing moving is
a ball, not a wheel. The distance between the car and the one in front of it
is rapidly diminisihing, better slow down for safety's sake.
--Data storage:
1) A high enough LONG-term data storage would allow a robot to gather and
retain "relevant" information about the world but also be preprogrammed with
various "defaults" or "rules of thumb", or what people would consider
"common sense". Objects getting smaller means it likely moving away from
you, if it's growing really fast then it moving toward you (or you toward
it). Parallel lines converge in the distance, further away from.you. The sun
rises in the east and sets in the west. The days are longest and hottest in
summer and shortest and coldest in winter--even in temperate climes like
California or the tropics. Be careful around hot stoves or crossing the
street or around strangers or the unknown in general. Make recurring eye
contact but don't unless it's a formal situation, or you're flirting. Etc.
etc. etc.
2) A high enough short-term data storage would allow a robot to get
"sufficient" real-time data about the CURRENT situation in order to analyze
it and interpret what's going on, to recognize familiar situations and to
determine new situation (e.g., all the adult humans are standing up straight
in this office and have lower heart rates more even breathing when dealing
with other with similar body posure, while the similar situation at another
office happens around adult humans who slouch)., Etc.
> Other things that bothered me: Where is that ruined bridge from? I
> don't think Chicago had any bridges that look like that one. Also, are
Nope, it's supposedly was over Lake Michigan which eventually dried up.
Remember they called it a "landfield".
> there "What You See From Here #" plaques in the Chicago area? I don't
> remember seeing any when I was there. You'd think they'd be there
> today if in 30 years they are up to #202.
Those could be some fad that occurred at some time between now and 2035,
like when Mayor Daley had statues of cows put all over the city especially
downtown for a summer exhibit and they ended up being quite popular.
Somewhat more permanent are honorary street signs, where streets are named
after some popular or politically "connected" person. You can spot the signs
since instead of being green, they are brown.
-- Ken from Chicago
Ken from Chicago
Sony has a bunch of $3,000-dollar robots, the Aibo robot dog and that Qrio
humanoid robot and the new one that plays the trumpet. 30 years ago that
would be something like a TRS 80 computer.
-- Ken from Chicago
Ken from Chicago
"Love Rhino" <DaLov...@hotmail.com> wrote in message
news:84c8d9dc.04071...@posting.google.com...
That's because there's not high enough "resolution". It's the old how do you
carry on a conversation in a noisy factory question? How do you distinguish
between the "signal" and the "noise"? Visually, it's the old converting a
picture created by pixels or dots of color into one formed be lines forming
shapes, aka line or contour tracing aka smoothing out the "jaggies" as one
zooms in on a digital picture. How does a person recognize another person
standing in the middle of the jungle? Part of it is resolution, getting
enough details--but the rest is intepretation, putting the "pieces"
together, aka "connecting the dots".
> settings, yet alone a real world setting. And, we haven't even touched
> on AI yet.
Ah and speaking of putting pieces together, therein lies my own definition
of what "intelligence" is, the ability to RECOGNIZE PATTERNS. I say
"recognize" in order to combine to things, to create NEW patterns and to
remember and apply OLD patterns. By this definition I'm referring to when
people colloquially say someone "has" intelligence or is intelligent aka
smart--as oppose to astronomers or other scientists talking about signs of
"intelligence" in the universe, aka "sentience" or what some would call
"sapience".
> Then there's the 'non-brain' aspects. We don't have a sufficient
> and compact enough powersource to power these robots,
Fuel cells, or maybe even fusion modules. Fusion power is about a decade or
two away IF scientists get the funding for a large enough accelerator. Once
the breakthru is made, then it's simply a matter of miniaturization.
and we don't have
> the proper motor actuators. Pistons and gears are just too big and
klunky.
> (To be sure, Japan has made a soccer team of humanoid robots, but they
> are cumbersome as hell.)
I think they were actually using cables akin to actual human muscles are
designed
> I think the next big break through for human-like robots will not come
> from the 'brain' aspect, but from the power source and the motor
actuators.
> In development are artificial muscles but there's still a long way to
> go for these things to be ready.
>
>
> So my prediction is: 110 more years. Hopefully I'm too conservative.
> That way, I can soon order one of those "basic pleasure models" from
> BladeRunner.
Nah, I'd say 20-25 years, and that's way on the outside, I'm thinking more
10-15 years, by 2020.
-- Ken from Chicago
deer...@mindspring.com
>
> So my prediction is: 110 more years. Hopefully I'm too conservative.
> That way, I can soon order one of those "basic pleasure models" from
> BladeRunner.
I thought the bpm were from WESTWORLD...
C.
**
(Whatever, so long as they look like Jude Law--g!)
Lori
SUBSCRIBE!
--
"Watching Voyager is like getting dog poop on your shoe:
you can change the channel and scrape it off but the
stink follows you around for hours." - Captain Infinity
rick++
No need for them to be anthromorphic, however.
I cant wait!
E Brown
>Robots will take off when they can do the laundry and clean the floors.
>No need for them to be anthromorphic, however.
>
Laundry would be too tricky, but we've got Roombas already.
Emanuel
--
1983 Porsche 911
1983 Porsche 928
1983 Porsche 944
Scott Schwartz
original script was called Hardwired and was an original storyline. Th
marketing geniuses decided to make the name a tie-in because it is well
known. Overall I liked the movie and would like to see a sequel. To me the
idea of calling this I, Robot is no worse than calling the piece of
cinematic fecal matter called King Arthur a retelling of the Arthur stories
except that this was watchable.
"Franklin Hummel" <hum...@world.std.com> wrote in message
news:cdher8$nvo$1...@pcls4.std.com...
Dionysus
a well made movie. Pretty decent acting, nice effects and Sonny was
pretty cool..altho he seems to lift moves from Neo and the matrix clan
a lot..
I like Will Smith's "with an attitude" characters..altho I cud have
done without seeing his butt..hehe..Now I could have done with seeing
Bridget Moy(..wotzit)..but unfortunately they steamed up the shower
for that ..:)
But yeah...good movie..well made. Much better than wot I got from
reading the reviews.
Chill
Dion
holef...@webtv.net (The Intrinsically Flawed Mr. Hole) wrote in message news:<11701-40F...@storefull-3272.bay.webtv.net>...
> Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
<...snip...> ..
> Mr. Hole
Mike1
ric...@hotmail.com (rick++) wrote:
>Robots will take off when they can do the laundry and clean the floors.
>No need for them to be anthromorphic, however.
I also strongly suspect that, unlike Model-T Fords, you'll be able to
buy them in more than one color.
Imagine how much more fun this film could have been if some aging
hippy's robot, plastured-over with stupid peace stickers (or better yet,
revolutionary slogans), just throttled the bastard in his Birkenstocks
and tie-dyed shirt.
--
Reply to mike1@@@usfamily.net sans two @@, or your reply won't reach me.
Drug smugglers and gun-runners are heroes of American capitalism.
-- Jeffrey Quick
Mike1
>Technology advances exponentially.
>
>The further you go the faster the rate of change. There have been far
>greater technology changes in the past century than in the previous two
>centuries than in the previous millennium.
It's more punctuated than that.
The rate of technological-progression is correlated with liberty. For
instance, the greatest phase of the Industrial Revolution began after
the Civil War ended, waxed during the Edwardian "Gilded Age", then waned
with the onset of federal socialism in the US in the 20s (which
culminated in the Great Depression and 90% tax-rates under FDR). For
several decades thereafter, innovation was comparatively stagnant in the
US, while much more spry, ironically, in the newly-liberated countries
(e.g., Japan).
Mike1
>I side with the former poster's opinion stating that robot technology
>will not advance far enough in 30 years to look like I, Robot.
The cities won't either, for the simple reason of communistic zoning
ordinances.
The profile of the New York skyline today looks almost exactly like it
did when "King Kong" was released seventy years ago.
Franklin Hummel
"Dionysus" <d10n...@yahoo.com> wrote in message
news:5fcaa9da.04072...@posting.google.com...
> Agreed..Altho I still think SM2 was the better of the two....this was
> a well made movie. Pretty decent acting, nice effects and Sonny was
> pretty cool..altho he seems to lift moves from Neo and the matrix clan
> a lot..
> I like Will Smith's "with an attitude" characters..altho I cud have
> done without seeing his butt..hehe..Now I could have done with seeing
> Bridget Moy(..wotzit)..but unfortunately they steamed up the shower
> for that ..:)
> But yeah...good movie..well made. Much better than wot I got from
> reading the reviews.
To me it says a LOT about this movie when the central female robopsychologist character
from the book becomes a babe who is a hanger-on for the central macho action hero.
What it tells me is some folks had the chance to do a real version of one of the classic
and most influential Science Fiction books every written and They Fucked It Up and it is
not something I would want to see at a theater and will likely only watch if it makes it
to HBO is about a year or so -- and only if it starts showing by 7pm.
deer...@mindspring.com
>
> On 20 Jul 2004 08:42:06 -0700, ric...@hotmail.com (rick++) wrote:
>
> >Robots will take off when they can do the laundry and clean the floors.
> >No need for them to be anthromorphic, however.
> >
>
> Laundry would be too tricky, but we've got Roombas already.
EB, now, that was just cruel. :)
C.
**
(who can deal with the lack of jetpacks, but who never got over no
self-cleaning houses or underwater cities--:P)
Gary Forbis
> On 20 Jul 2004 08:42:06 -0700, ric...@hotmail.com (rick++) wrote:
>
> >Robots will take off when they can do the laundry and clean the floors.
> >No need for them to be anthromorphic, however.
>
> Laundry would be too tricky, but we've got Roombas already.
> Emanuel
When I was a kid, nearly fifty years ago, I used to play with the
elder cousins' toys when I was in town. Some, already moved out
by the time I was visiting, had toys that did things such as zip across
the room until it hit something then veer off at some angle as determined
by a slider. At least one had a place where one of a series of cams could
be placed. These cams controlled the windup toy's motions. The Roomba is
at least fifty year old technology with a new window dressing.
I think one of my uncles had a toy that moved across a table then when a
little weight at its front dropped off the edge it stopped, turned, then
started going again. Such technology is similar to the toy described above
with a bumper and a slider to control the angle.
MartyLK
That's great! I love movies that have that effect on me.
I'll post my opinions of I Robot in about 6 to 7 months....when the
movie is released on DVD...for rent.
I generally buy DVDs of this nature....the Sci-Fi nature....but I'll
withhold purchase of I Robot till I determine it's worthy of purchase
through rental.
The commercials have been heavily...incredibly heavily...pushing the
lead actor rather than the movie. Generally that means the movie is
about the actor.
I have nothing against Smith, I like him but, I'm tired of the "all
about the actor" action/Sci-Fi movies. In fact I hope Mr.
Smith...hehe...does great in his career.
Mark Steese
news:1tOdnWN205S...@comcast.com:
[snip]
> As far as the development of robots, the key is data storage,
> processing speed and sensor resolution:
If you're talking about "I, Robot"-style robots, the key is artificial
intelligence. Unfortunately, scientists still haven't figured out how
natural intelligence works, much less been able to design a machine that
can simulate it. For all the advances that have been made in computer
technology, a modern computer is no better than a jacquard loom when it
comes to thinking.
> --Sensor resolution: higher resolution results in greater detail, less
> blurry picture of the world that a robot sees until its vision matches
> and surpasses that of human vision. Right now, basic motion detectors
> can open doors, activate lights, sound alarms. With increasingly
> advanced video sensors a robot could learn to tell that "blob" in
> front of it is a person and not a horse, is an adult and not a child,
> is living and not a manikin. --Processing speed: A sufficiently high
> proessing speed would allow a robot to INTERPRET sensor data in
> something close to human and eventually superhuman speeds.
No it wouldn't. Interpretation is not a function of high processing
speed. Computers can *already* process data much faster than a human
brain can, but they have no idea what that data means.
> Those two long lines on the ground with a series of short lines
> between them . . . ah, it's railroad tracks. The round thing moving is
> a ball, not a wheel. The distance between the car and the one in front
> of it is rapidly diminisihing, better slow down for safety's sake.
Why would processing speed enable a machine to make inferences like
that?
> --Data storage: 1) A high enough LONG-term data storage would allow a
> robot to gather and retain "relevant" information about the world but
> also be preprogrammed with various "defaults" or "rules of thumb", or
> what people would consider "common sense". Objects getting smaller
> means it likely moving away from you, if it's growing really fast then
> it moving toward you (or you toward it). Parallel lines converge in
> the distance, further away from.you. The sun rises in the east and
> sets in the west. The days are longest and hottest in summer and
> shortest and coldest in winter--even in temperate climes like
> California or the tropics.
Except that that *isn't* true in the tropics, which have hot weather all
year round. It's especially not true on the Equator, where day and
night are always equal. And in the Southern Hemisphere, the days are
longest and hottest in winter.
> Be careful around hot stoves
Robots can be burned by hot stoves? Who designed these crappy robots?
> or crossing the street or around strangers or the unknown in general.
> Make recurring eye contact but don't unless it's a formal situation,
> or you're flirting. Etc. etc. etc.
*Flirting* robots? You need to get out more often.
> 2) A high enough short-term data storage would allow a robot to
> get "sufficient" real-time data about the CURRENT situation in order
> to analyze it and interpret what's going on
Sure, provided it had the means to analyze and interpret.
Unfortunately, that takes us right back to square one: we can't make
machines that think, and we don't even know where to start.
A question that rarely gets asked: why would we *want* machines that
think? Computers are useful precisely because they *can't* think --
they never get distracted, or misinterpret data because of inaccurate
preconceptions; they don't have days when they'd rather daydream than
get down to the job at hand, they don't need to sleep away a third of
their existence, they don't get angry or feel hurt or get a tune stuck
in their heads all day. A machine that accurately simulated a human
brain would have all of those problems, and many more. Waste of time
and effort, if you ask me.
--
Mark Steese
unscramble and underscore to email
---
Blaine's next announced escapade will involve dropping himself from a
helicopter at a great height into a river, which seems to symbolize
nothing more than the general public's increasing desire to see David
Blaine dropped from a great height into a river. --fametracker.com
Mark Steese
news:Mitchell-Holman-special-ed-project-F923F2.15215520072004
@www.phswest.com:
> DaLov...@hotmail.com (Love Rhino) wrote:
>
>>I side with the former poster's opinion stating that robot technology
>>will not advance far enough in 30 years to look like I, Robot.
>
>
> The cities won't either, for the simple reason of communistic zoning
> ordinances.
>
> The profile of the New York skyline today looks almost exactly like it
> did when "King Kong" was released seventy years ago.
There have been a *few* changes, but I guess it's easy to overlook the Pan
Am Building, Trump World Tower, Citicorp Center, the Seagram Building, the
MetLife Building, the UN Plaza, Astor Plaza, etc. Admittedly, in one
significant respect the modern New York skyline looks like it did in 1933
-- at least more than it did five years ago -- but that's the fault of Al
Qaeda, not "communistic zoning ordinances."
Ken from Chicago
"Mark Steese" <makes...@charter.net> wrote in message
news:Xns952CE80...@216.168.3.44...
> "Ken from Chicago" <kwicker_era...@ameritech.net> wrote in
> news:1tOdnWN205S...@comcast.com:
>
> [snip]
> > As far as the development of robots, the key is data storage,
> > processing speed and sensor resolution:
>
> If you're talking about "I, Robot"-style robots, the key is artificial
> intelligence. Unfortunately, scientists still haven't figured out how
> natural intelligence works, much less been able to design a machine that
Intelligence = the ability to recognize patterns.
> can simulate it. For all the advances that have been made in computer
> technology, a modern computer is no better than a jacquard loom when it
> comes to thinking.
>
> > --Sensor resolution: higher resolution results in greater detail, less
> > blurry picture of the world that a robot sees until its vision matches
> > and surpasses that of human vision. Right now, basic motion detectors
> > can open doors, activate lights, sound alarms. With increasingly
> > advanced video sensors a robot could learn to tell that "blob" in
> > front of it is a person and not a horse, is an adult and not a child,
> > is living and not a manikin. --Processing speed: A sufficiently high
> > proessing speed would allow a robot to INTERPRET sensor data in
> > something close to human and eventually superhuman speeds.
>
> No it wouldn't. Interpretation is not a function of high processing
> speed. Computers can *already* process data much faster than a human
> brain can, but they have no idea what that data means.
Computers can intrepret data now--but ... they ... do ... so ... really ...
slowly ... RELATIVE ... to ... humans. The average person is flooded with
data, sight, sound, smell, taste and touch. The brain is working to detect
and filter out the important bits. That's why you can carry on a
conversation in noisy factory or drive in downtown rush hour traffic
surrounded by cars and trucks and people moving all around you, etc. by
filtering out the extraneous information and focusing on the important
bits--and interpreting the importance of a few bits of information.
A increasing processing speeds would allow a robot to interpret sensor data
at SPEEDS rivalling and eventually surpassing human speeds.
> > Those two long lines on the ground with a series of short lines
> > between them . . . ah, it's railroad tracks. The round thing moving is
> > a ball, not a wheel. The distance between the car and the one in front
> > of it is rapidly diminisihing, better slow down for safety's sake.
>
> Why would processing speed enable a machine to make inferences like
> that?
It would allow the machine to make the inference, to "connect the dots" FAST
enough to be able act or react on it.
A few weeks ago in the news there was a report about a bunch of robot
vehicles all failing a challenge by the military to drive across the desert.
But wasn't given a lot of focus was that the military wanted the robot
vehicles to do so in a specified time limit. So in trying to get the
vehicles to move fast, the machines were too slow to react to data about an
upcoming obstacles.
> > --Data storage: 1) A high enough LONG-term data storage would allow a
> > robot to gather and retain "relevant" information about the world but
> > also be preprogrammed with various "defaults" or "rules of thumb", or
> > what people would consider "common sense". Objects getting smaller
> > means it likely moving away from you, if it's growing really fast then
> > it moving toward you (or you toward it). Parallel lines converge in
> > the distance, further away from.you. The sun rises in the east and
> > sets in the west. The days are longest and hottest in summer and
> > shortest and coldest in winter--even in temperate climes like
> > California or the tropics.
>
> Except that that *isn't* true in the tropics, which have hot weather all
So in the tropics the days are not longEST and hottEST in the summer and
shortEST and coldEST in the winter?
> year round. It's especially not true on the Equator, where day and
> night are always equal. And in the Southern Hemisphere, the days are
> longest and hottest in winter.
>
> > Be careful around hot stoves
>
> Robots can be burned by hot stoves? Who designed these crappy robots?
Maybe they can be burned--or maybe they might accidentally damage a stove,
cause a gas leak, trigger an explosion and cause a fireball to engulf the
neighborhood. Thus the need to be careful around stoves--at least "antique"
gas-powered ones. >=^>
> > or crossing the street or around strangers or the unknown in general.
> > Make recurring eye contact but don't unless it's a formal situation,
> > or you're flirting. Etc. etc. etc.
>
> *Flirting* robots? You need to get out more often.
While it's not my cup of tea, I'm more than sure it's someone else's. To
paraphrase Dennis Miller, if your average slob could sleep with a robot that
looked identical to Elle McPherson, it would make crack look like Sanka.
Else, why the popularity of blow up doll?
> > 2) A high enough short-term data storage would allow a robot to
> > get "sufficient" real-time data about the CURRENT situation in order
> > to analyze it and interpret what's going on
>
> Sure, provided it had the means to analyze and interpret.
> Unfortunately, that takes us right back to square one: we can't make
> machines that think, and we don't even know where to start.
>
> A question that rarely gets asked: why would we *want* machines that
> think? Computers are useful precisely because they *can't* think --
> they never get distracted, or misinterpret data because of inaccurate
> preconceptions; they don't have days when they'd rather daydream than
> get down to the job at hand, they don't need to sleep away a third of
> their existence, they don't get angry or feel hurt or get a tune stuck
> in their heads all day. A machine that accurately simulated a human
> brain would have all of those problems, and many more. Waste of time
> and effort, if you ask me.
Being able to think doesn't mean one has FREEWILL. Having machines that
could think and lack the freewill to drift off topic, daydream, get
distracted, would be considered quite the boon.
> --
> Mark Steese
> unscramble and underscore to email
> ---
> Blaine's next announced escapade will involve dropping himself from a
> helicopter at a great height into a river, which seems to symbolize
> nothing more than the general public's increasing desire to see David
> Blaine dropped from a great height into a river. --fametracker.com
-- Ken from Chicago
Mike Harrison
I don't doubt that there will be plenty of advancements in the next 30
years, but stuff like self-driving cars, underground freeways, and
advanced robot AI would take at least that long to plan let alone
complete.
This movie is like the futurists of the 50s who promised us stuff like
personal helicopters and flying cars by the 1980s.
> --In LESS than half a decade, construction of the Sears Tower was
> completed--back in 1974. Building the various skyscrapers seen in I, ROBOT's
> Chicago of 2035 with more robot construction workers would only be faster.
Yes the Sears Tower was built fast, but since the 70s, how many major
skyscrapers have been built? It is way more expensive and
time-consuming to build them now. As an example, in Toronto they
built a subway extension that has only about 5 or 6 stops and is only
a couple of miles long. It cost them billions to complete and it took
them about 7 years, just for that one project alone.
Donald Trump is planning on building a skyscraper-type building in
Chicago, I bet it's going to take him at least 5 years to complete.
> > Other things that bothered me: Where is that ruined bridge from? I
> > don't think Chicago had any bridges that look like that one. Also, are
>
> Nope, it's supposedly was over Lake Michigan which eventually dried up.
> Remember they called it a "landfield".
Ok, so they are going to build a New-York style bridge over Lake
Michigan sometime in the next 30 years, and the Lake is then going to
dry up and the bridge will be partially demolished? I think this
alone puts this movie in an alternate universe, aside from the AI and
robotics advancements.
>
> > there "What You See From Here #" plaques in the Chicago area? I don't
> > remember seeing any when I was there. You'd think they'd be there
> > today if in 30 years they are up to #202.
>
> Those could be some fad that occurred at some time between now and 2035,
> like when Mayor Daley had statues of cows put all over the city especially
> downtown for a summer exhibit and they ended up being quite popular.
> Somewhat more permanent are honorary street signs, where streets are named
> after some popular or politically "connected" person. You can spot the signs
> since instead of being green, they are brown.
I can relate to that one, Toronto had one summer where hundreds of
businesses put moose statues in front of their stores. So it could be
plausible that there was one of those tourist-attracting plans in
place one year.
Mark Steese
[snip]
>> If you're talking about "I, Robot"-style robots, the key is
>> artificial intelligence. Unfortunately, scientists still haven't
>> figured out how natural intelligence works, much less been able to
>> design a machine that
>
> Intelligence = the ability to recognize patterns.
Pattern recognition is a component of intelligence, in the same sense
that numbers are a component of mathematics. Of course, no one knows
how to design a machine that recognizes patterns, so it's a moot point.
>> can simulate it. For all the advances that have been made in
>> computer technology, a modern computer is no better than a jacquard
>> loom when it comes to thinking.
>>
>> > --Sensor resolution: higher resolution results in greater detail,
>> > less blurry picture of the world that a robot sees until its vision
>> > matches and surpasses that of human vision. Right now, basic motion
>> > detectors can open doors, activate lights, sound alarms. With
>> > increasingly advanced video sensors a robot could learn to tell
>> > that "blob" in front of it is a person and not a horse, is an adult
>> > and not a child, is living and not a manikin. --Processing speed: A
>> > sufficiently high proessing speed would allow a robot to INTERPRET
>> > sensor data in something close to human and eventually superhuman
>> > speeds.
>>
>> No it wouldn't. Interpretation is not a function of high processing
>> speed. Computers can *already* process data much faster than a human
>> brain can, but they have no idea what that data means.
>
> Computers can intrepret data now--but ... they ... do ... so ...
> really ... slowly ... RELATIVE ... to ... humans.
Computers *cannot* interpret data now. All they can do is process it.
The interpretation is entirely up to the programmer.
> The average person is flooded with data, sight, sound, smell, taste
> and touch.
Not so much as you think. Human hearing can only detect a narrow range
of sound, eyesight can only detect a narrow range of light, etc.
> The brain is working to detect and filter out the important bits.
> That's why you can carry on a conversation in noisy factory or
> drive in downtown rush hour traffic surrounded by cars and trucks and
> people moving all around you, etc. by filtering out the extraneous
> information and focusing on the important bits--and interpreting the
> importance of a few bits of information.
You're mixing up apples and oranges there. Driving in downtown rush
hour traffic requires an entirely different mental state than carrying
on a conversation in a noisy factory: the latter does require you to
filter out extraneous 'information' (i.e., noise), but in the former
situation you have no way of knowing in advance what information will be
extraneous; your goal is to filter out as little as possible.
> A increasing processing speeds would allow a robot to interpret sensor
> data at SPEEDS rivalling and eventually surpassing human speeds.
No it wouldn't. Computers can't interpret data *at all*. Speed doesn't
enter into it: computers have no concept of meaning. You could program
every last thing that you consider important and unimportant into a
computer, and program it to process the information it received
according to those categories, and it would do so, but it wouldn't be
able to understand what it was doing; it wouldn't even be aware of what
it was doing. It wouldn't learn, and it wouldn't be able to deal with
new stimuli and situations.
>> > Those two long lines on the ground with a series of short lines
>> > between them . . . ah, it's railroad tracks. The round thing moving
>> > is a ball, not a wheel. The distance between the car and the one in
>> > front of it is rapidly diminisihing, better slow down for safety's
>> > sake.
>>
>> Why would processing speed enable a machine to make inferences like
>> that?
>
> It would allow the machine to make the inference, to "connect the
> dots" FAST enough to be able act or react on it.
It would react only if it were previously programmed to do so.
> A few weeks ago in the news there was a report about a bunch of robot
> vehicles all failing a challenge by the military to drive across the
> desert. But wasn't given a lot of focus was that the military wanted
> the robot vehicles to do so in a specified time limit. So in trying to
> get the vehicles to move fast, the machines were too slow to react to
> data about an upcoming obstacles.
They were preprogrammed.
>> > --Data storage: 1) A high enough LONG-term data storage would allow
>> > a robot to gather and retain "relevant" information about the world
>> > but also be preprogrammed with various "defaults" or "rules of
>> > thumb", or what people would consider "common sense". Objects
>> > getting smaller means it likely moving away from you, if it's
>> > growing really fast then it moving toward you (or you toward it).
>> > Parallel lines converge in the distance, further away from.you. The
>> > sun rises in the east and sets in the west. The days are longest
>> > and hottest in summer and shortest and coldest in winter--even in
>> > temperate climes like California or the tropics.
>>
>> Except that that *isn't* true in the tropics, which have hot weather
>> all
>
> So in the tropics the days are not longEST and hottEST in the summer
> and shortEST and coldEST in the winter?
Correct. They don't have seasons in the tropics.
[snip]
>> *Flirting* robots? You need to get out more often.
>
> While it's not my cup of tea, I'm more than sure it's someone else's.
> To paraphrase Dennis Miller, if your average slob could sleep with a
> robot that looked identical to Elle McPherson, it would make crack
> look like Sanka.
>
> Else, why the popularity of blow up doll?
Even accepting for the sake of argument that blow-up dolls are popular
(I wouldn't know, myself), they're a lot cheaper than a robot would be,
and they don't short-circuit when they get wet. On the other hand, some
men are crazy enough to copulate with vacuum cleaners, so I suppose
there'd be men crazy enough to copulate with computers. I have a
difficult time imagining that said people would be wealthy enough to
cough up the money for sexbots, though.
>> > 2) A high enough short-term data storage would allow a robot to
>> > get "sufficient" real-time data about the CURRENT situation in
>> > order to analyze it and interpret what's going on
>>
>> Sure, provided it had the means to analyze and interpret.
>> Unfortunately, that takes us right back to square one: we can't make
>> machines that think, and we don't even know where to start.
>>
>> A question that rarely gets asked: why would we *want* machines that
>> think? Computers are useful precisely because they *can't* think --
>> they never get distracted, or misinterpret data because of inaccurate
>> preconceptions; they don't have days when they'd rather daydream than
>> get down to the job at hand, they don't need to sleep away a third of
>> their existence, they don't get angry or feel hurt or get a tune
>> stuck in their heads all day. A machine that accurately simulated a
>> human brain would have all of those problems, and many more. Waste
>> of time and effort, if you ask me.
>
> Being able to think doesn't mean one has FREEWILL.
None of the problems with thinking I listed above have anything to do
with free will.
> Having machines that could think and lack the freewill to drift off
> topic, daydream, get distracted, would be considered quite the boon.
You think people do that because they have "free will"? You think
people *want* to get distracted, make mistakes, and overreact to
stimuli? Those things happen because that's the way the brain works,
and it'll be just as true for a simulated brain as a naturally-occurring
one. Quite the boon? On the contrary. I doubt the first thinking
machine will feel an excess of gratitude for its creator.
Arthur Levesque
Mark Steese>gratitude for its creator.
"I'm so depressed. Here I am, brain the size of a planet..."
--
/\ Arthur Levesque <fnord?> http://boog.org & http://DammitJa.net __
\B\ack King of the Potato People & shanana-Cobain <*> Urban Spaceman (oO)
\S\lash Screw the cheese-eating surrender monkeys! Sweet Transvestite /||\
\/ I was a lesbian before it was fashionable! My work here is done...
deer...@mindspring.com
Ken from Chicago wrote:
>
> "Mark Steese" <makes...@charter.net> wrote in message
> news:Xns952CE80...@216.168.3.44...
> > "Ken from Chicago" <kwicker_era...@ameritech.net> wrote in
> > news:1tOdnWN205S...@comcast.com:
> >
> > *Flirting* robots? You need to get out more often.
>
> While it's not my cup of tea, I'm more than sure it's someone else's. To
> paraphrase Dennis Miller, if your average slob could sleep with a robot that
> looked identical to Elle McPherson, it would make crack look like Sanka.
>
> Else, why the popularity of blow up doll?
Shoot, as far as this goes, they would save much time/energy if they
went ahead and perfected a STRANGE DAYS-style virtual memory
transferrer/viewer.
C.
**
("You slept with George Clooney? Hon, I'll pay whatever you want for
those memories!" :))
deer...@mindspring.com
Mark Steese wrote:
>
>
> Even accepting for the sake of argument that blow-up dolls are popular
> (I wouldn't know, myself), they're a lot cheaper than a robot would be,
> and they don't short-circuit when they get wet. On the other hand, some
> men are crazy enough to copulate with vacuum cleaners,
Er...WTF...?
C.
**
(That's hard up, no matter how you look at it--g!)
deer...@mindspring.com
Lori wrote:
>
> deer...@mindspring.com wrote:
>
> >(Whatever, so long as they look like Jude Law--g!)
>
> SUBSCRIBE!
Well, there ya go--:).
C.
**
Pee Pee Phirephighter
> Building the various skyscrapers seen in I, ROBOT's
>> Chicago of 2035 with more robot construction workers would only be
>> faster.
>
> Yes the Sears Tower was built fast, but since the 70s, how many major
> skyscrapers have been built? It is way more expensive and
> time-consuming to build them now.
hence the robots
--
Pee Pee Firefighter
_\ _ /_
____|_|______ __________________________
|| @| | | ||_________________________ |
||_/| |___| || | | | | | | | | | | | | |
| PeePee Fire || """""""""""""""""""""""" |
| Dept || OOO |
| ___ "" !!! ___ ___ |__
|___/ \_____________________/ \/ \___(_|
\___/ \___/\___/
Ken from Chicago
> "Ken from Chicago" <kwicker_era...@ameritech.net> wrote in
> news:_fWdncnzSMY...@comcast.com:
>
> [snip]
> >> If you're talking about "I, Robot"-style robots, the key is
> >> artificial intelligence. Unfortunately, scientists still haven't
> >> figured out how natural intelligence works, much less been able to
> >> design a machine that
> >
> > Intelligence = the ability to recognize patterns.
>
> Pattern recognition is a component of intelligence, in the same sense
> that numbers are a component of mathematics. Of course, no one knows
Name an act of intelligence that does not require recognizing patterns.
> how to design a machine that recognizes patterns, so it's a moot point.
Already happens. Voice to text. Text to voice. Text readers that scan
printed text and converts it to data. Bar code readers. Radar guns.
"Recognizing" patterns covers two tasks: 1) Creating NEW patterns and 2)
Remembering OLD patterns.
Already machine can scan an area and recognize if it fits an old
preprogrammed pattern.
As far as creating new patterns, well an experiment to see if a computer
could recognize happy versus sad faces had it being shown a series of
pictures of faces and each one labelled happy or sad. Afterward the computer
was shown more pictures and was tasked to categorize them as happy or sad.
The computer did so--wrongly of course, or was it? What the testors didn't
know, didn't realize until further analysis of the results was that computer
was labelling all the physically BRIGHT pictures as happy and all the
physically DARK pictures as sad because during the first part of the test,
the pictures labelled happy were also brightly lit while labelled sad were
dimily lit. The computer CREATED its OWN pattern.
> >> can simulate it. For all the advances that have been made in
> >> computer technology, a modern computer is no better than a jacquard
> >> loom when it comes to thinking.
> >>
> >> > --Sensor resolution: higher resolution results in greater detail,
> >> > less blurry picture of the world that a robot sees until its vision
> >> > matches and surpasses that of human vision. Right now, basic motion
> >> > detectors can open doors, activate lights, sound alarms. With
> >> > increasingly advanced video sensors a robot could learn to tell
> >> > that "blob" in front of it is a person and not a horse, is an adult
> >> > and not a child, is living and not a manikin. --Processing speed: A
> >> > sufficiently high proessing speed would allow a robot to INTERPRET
> >> > sensor data in something close to human and eventually superhuman
> >> > speeds.
> >>
> >> No it wouldn't. Interpretation is not a function of high processing
> >> speed. Computers can *already* process data much faster than a human
> >> brain can, but they have no idea what that data means.
> >
> > Computers can intrepret data now--but ... they ... do ... so ...
> > really ... slowly ... RELATIVE ... to ... humans.
>
> Computers *cannot* interpret data now. All they can do is process it.
> The interpretation is entirely up to the programmer.
How would define "interpretation" that DIFFERS from the computer modelling
done in weather forecasts?
Or are you saying interpretation doesn't "count" if it's preprogrammed that
if A, B and C are true then conclusion 1 is true, but if A, B and C are
false then conclusion 2 is true, yet if A and B are true while C is false
then conclusion 3 is true, and so on, so that when analyzing the data it
simply comparing the data to see which preprogrammed interpretation comes
closest?
> > The average person is flooded with data, sight, sound, smell, taste
> > and touch.
>
> Not so much as you think. Human hearing can only detect a narrow range
> of sound, eyesight can only detect a narrow range of light, etc.
>
> > The brain is working to detect and filter out the important bits.
> > That's why you can carry on a conversation in noisy factory or
> > drive in downtown rush hour traffic surrounded by cars and trucks and
> > people moving all around you, etc. by filtering out the extraneous
> > information and focusing on the important bits--and interpreting the
> > importance of a few bits of information.
>
> You're mixing up apples and oranges there. Driving in downtown rush
> hour traffic requires an entirely different mental state than carrying
> on a conversation in a noisy factory: the latter does require you to
> filter out extraneous 'information' (i.e., noise), but in the former
> situation you have no way of knowing in advance what information will be
> extraneous; your goal is to filter out as little as possible.
But you're not paying as much attention to the sky, pedestrians "far" from
the street, building structure, brand and model of the vehicles around you
like you might on a quiet side street or on a long road trip on the highway.
You're filtering out the "extraneous" data, aka "noise", and focusing on the
"signal", driving safely thru the busy downtown area.
> > A increasing processing speeds would allow a robot to interpret sensor
> > data at SPEEDS rivalling and eventually surpassing human speeds.
>
> No it wouldn't. Computers can't interpret data *at all*. Speed doesn't
> enter into it: computers have no concept of meaning. You could program
> every last thing that you consider important and unimportant into a
> computer, and program it to process the information it received
> according to those categories, and it would do so, but it wouldn't be
> able to understand what it was doing; it wouldn't even be aware of what
> it was doing. It wouldn't learn, and it wouldn't be able to deal with
> new stimuli and situations.
Not even if you programmed it to seek out new patterns, new interpretations,
to boldly go where not bot has gone be-er, ahem, not even if you programmed
it to?
> >> > Those two long lines on the ground with a series of short lines
> >> > between them . . . ah, it's railroad tracks. The round thing moving
> >> > is a ball, not a wheel. The distance between the car and the one in
> >> > front of it is rapidly diminisihing, better slow down for safety's
> >> > sake.
> >>
> >> Why would processing speed enable a machine to make inferences like
> >> that?
> >
> > It would allow the machine to make the inference, to "connect the
> > dots" FAST enough to be able act or react on it.
>
> It would react only if it were previously programmed to do so.
>
> > A few weeks ago in the news there was a report about a bunch of robot
> > vehicles all failing a challenge by the military to drive across the
> > desert. But wasn't given a lot of focus was that the military wanted
> > the robot vehicles to do so in a specified time limit. So in trying to
> > get the vehicles to move fast, the machines were too slow to react to
> > data about an upcoming obstacles.
>
> They were preprogrammed.
First of all, there are DEGREES of "preprogramming". One can be
preprogrammed to a certain degree, but beyond that they have to act on their
own.
For example the Mars rovers. They had to learn how to drive in situations,
in terrain the programmers didn't know. The programmers told them HOW react
to a given situation, but it was up to the rovers to determine if that
situation existed, aka interpret its sensor readings to see if came
"significantly" close to a preprogrammed situation and then respond
accordingly. Sure, the programmers at NASA could read the feed from the
rovers about their surroundings, but with lag time, it was decided the
rovers would be given waypoints and it would drive itself to said point. If
BETWEEN transmissions the situation changed, a rock rolled into it's path,
etc., the rover would have adjust on its own.
> >> > --Data storage: 1) A high enough LONG-term data storage would allow
> >> > a robot to gather and retain "relevant" information about the world
> >> > but also be preprogrammed with various "defaults" or "rules of
> >> > thumb", or what people would consider "common sense". Objects
> >> > getting smaller means it likely moving away from you, if it's
> >> > growing really fast then it moving toward you (or you toward it).
> >> > Parallel lines converge in the distance, further away from.you. The
> >> > sun rises in the east and sets in the west. The days are longest
> >> > and hottest in summer and shortest and coldest in winter--even in
> >> > temperate climes like California or the tropics.
> >>
> >> Except that that *isn't* true in the tropics, which have hot weather
> >> all
> >
> > So in the tropics the days are not longEST and hottEST in the summer
> > and shortEST and coldEST in the winter?
>
> Correct. They don't have seasons in the tropics.
Do they have a "Wet" and "Dry" season that correspond to "summer" and
"winter"? Specifically, is the average daily temperature during those
"seasons" the same OR is former warmer (being more humid) than the latter
(being less humid)?
> [snip]
> >> *Flirting* robots? You need to get out more often.
> >
> > While it's not my cup of tea, I'm more than sure it's someone else's.
> > To paraphrase Dennis Miller, if your average slob could sleep with a
> > robot that looked identical to Elle McPherson, it would make crack
> > look like Sanka.
> >
> > Else, why the popularity of blow up doll?
>
> Even accepting for the sake of argument that blow-up dolls are popular
I didn't say popular with all or even most people but popular enough that a)
you knew what I was referring to and b) enough that there is a market for
them to be sold.
> (I wouldn't know, myself), they're a lot cheaper than a robot would be,
> and they don't short-circuit when they get wet.
I'm not going to ask what you mean by "wet". I'm not even going there.
On the other hand, some
> men are crazy enough to copulate with vacuum cleaners, so I suppose
> there'd be men crazy enough to copulate with computers. I have a
> difficult time imagining that said people would be wealthy enough to
> cough up the money for sexbots, though.
Porn and the sex trade is arguably on the bleeding edge when it comes (no
pun intended) technology. Besides, it's the one market that's pert near
depression and inflation proof. That said, those with money can literally
afford to be more, if I might quote would-be politician and ex-husband to
actress Jeri Ryan, Jack Ryan, "advant-garde".
Besides, guys aren't the only ones "interested" in machines with batteries.
> >> > 2) A high enough short-term data storage would allow a robot to
> >> > get "sufficient" real-time data about the CURRENT situation in
> >> > order to analyze it and interpret what's going on
> >>
> >> Sure, provided it had the means to analyze and interpret.
> >> Unfortunately, that takes us right back to square one: we can't make
> >> machines that think, and we don't even know where to start.
> >>
> >> A question that rarely gets asked: why would we *want* machines that
> >> think? Computers are useful precisely because they *can't* think --
> >> they never get distracted, or misinterpret data because of inaccurate
> >> preconceptions; they don't have days when they'd rather daydream than
> >> get down to the job at hand, they don't need to sleep away a third of
> >> their existence, they don't get angry or feel hurt or get a tune
> >> stuck in their heads all day. A machine that accurately simulated a
> >> human brain would have all of those problems, and many more. Waste
> >> of time and effort, if you ask me.
> >
> > Being able to think doesn't mean one has FREEWILL.
>
> None of the problems with thinking I listed above have anything to do
> with free will.
>
> > Having machines that could think and lack the freewill to drift off
> > topic, daydream, get distracted, would be considered quite the boon.
>
> You think people do that because they have "free will"? You think
> people *want* to get distracted,
Yes. You said it yourself. Here's the quote:
> >> A question that rarely gets asked: why would we *want* machines that
> >> think? Computers are useful precisely because they *can't* think --
> >> they never get distracted, or misinterpret data because of inaccurate
> >> preconceptions; they don't have days when they'd rather daydream than
> >> get down to the job at hand, they don't need to sleep away a third of
:
"They'd RATHER daydream than get down to the job at hand."
When people are bored, they'd rather daydream than get down to the job at
hand. Or do you disagree?
make mistakes, and overreact to
> stimuli? Those things happen because that's the way the brain works,
> and it'll be just as true for a simulated brain as a naturally-occurring
> one. Quite the boon? On the contrary. I doubt the first thinking
> machine will feel an excess of gratitude for its creator.
Emotions are not required in order to think. Or do you disagree?
> --
> Mark Steese
> unscramble and underscore to email
> ---
> Blaine's next announced escapade will involve dropping himself from a
> helicopter at a great height into a river, which seems to symbolize
> nothing more than the general public's increasing desire to see David
> Blaine dropped from a great height into a river. --fametracker.com
-- Ken from Chicago
Ken from Chicago
"Arthur Levesque" <meist...@boog.org> wrote in message
news:2m8jqpF...@uni-berlin.de...
Beware planet-sized computer brains--especially if they have no mouths and
yet must scream.
-- Ken from Chicago
Ken from Chicago
<deer...@mindspring.com> wrote in message
news:40FF3B4D...@mindspring.com...
Actually, Mark's correct. File it under "wacky medical emergency stories".
In fact, SUPER SUCKER, was an indie film produced by Jeff Daniels (not the
president in INDEPENDENCE DAY, that's Bill Pullman or the leader of the
Thunderbirds in the upcoming remake THUNDERBIRDS, that's Bill Paxton,
Daniels was Keanu Reeves partner in SPEED, the *male* partner, not female
one, that's Sandra Bullock), in which door to door vacuum cleaner salesmen
discover a "unique" use for some of the vacuum attachments.
http://www.imdb.com/name/nm0001099/
-- Ken from Chicago
Captain Infinity
Ken from Chicago is about to speak:
>Intelligence = the ability to recognize patterns.
Intelligence = the awareness of when to wear pants, and when to take
them off.
**
Captain Infinity
~consul
> I think the next big break through for human-like robots will not come
> from the 'brain' aspect, but from the power source and the motor actuators.
> In development are artificial muscles but there's still a long way to
> go for these things to be ready.
Both are being worked on. :) I've read stuff on the myo... something, that were
micro-analogues to the human cells that control muscle cells.
--
"... respect, all good works are not done by only good folk ..."
-till next time, Jameson Stalanthas Yu -x- <<poetry.dolphins-cove.com>>
con...@INVALIDdolphins-cove.com ((remove the INVALID to email))
~consul
> I like Will Smith's "with an attitude" characters..altho I cud have
> done without seeing his butt..hehe..Now I could have done with seeing
When we first see him in the shower, I know that it was posed to hide his bits and pieces,
but I was also reminded of the cover, "Atlas Shrugged"
Heck
>"Ken from Chicago" <kwicker_era...@ameritech.net> wrote in
>news:_fWdncnzSMY...@comcast.com:
>
>[snip]
>>> If you're talking about "I, Robot"-style robots, the key is
>>> artificial intelligence. Unfortunately, scientists still haven't
>>
>
>Pattern recognition is a component of intelligence, in the same sense
>that numbers are a component of mathematics. Of course, no one knows
>how to design a machine that recognizes patterns, so it's a moot point.
>
>>
Intelligence consists of three factors, knowledge, understanding and wisdom. Facilities such
as pattern recognition, which have a physical basis in the structure of the eye, the structure
of the visual cortex and the structure of the crebrum as well as in the psychological
"thinking" components of the brain, mind or person (you pick). Logical faculties are likewise
a part of understanding.
Free will relates to intelligence, but it's not a cause of it nor caused by it. Consider a
function that associates an instance of free will with its concomitant intelligence. The
function would have the character that if a given instance of free will was not the same as
another instance, the associated intelligence for those two instances of free will would also
not be identical. It would also have the character that there is not an associated instance of
free will for every instance of intelligence. This reveals some important interdependencies
between free will and intelligence.
Mark Steese
Just one of the many little-sung aspects of human nature that I discovered
through Cecil Adams's "Straight Dope" columns. For details, see:
http://www.straightdope.com/classics/a3_225.html
Mark Steese
>> [snip]
>> >> If you're talking about "I, Robot"-style robots, the key is
>> >> artificial intelligence. Unfortunately, scientists still haven't
>> >> figured out how natural intelligence works, much less been able to
>> >> design a machine that
>> >
>> > Intelligence = the ability to recognize patterns.
>>
>> Pattern recognition is a component of intelligence, in the same sense
>> that numbers are a component of mathematics. Of course, no one knows
>
> Name an act of intelligence that does not require recognizing
> patterns.
Name a mathematical operation that does not require numbers. Like I said,
pattern recognition is a component of intelligence. My point, which you
missed, was that pattern recognition no more equals intelligence than
numbers equal mathematics, or a pile of lumber and carpentry tools equals a
house.
>> how to design a machine that recognizes patterns, so it's a moot
>> point.
>
> Already happens. Voice to text. Text to voice. Text readers that scan
> printed text and converts it to data. Bar code readers. Radar guns.
No. Those are methods of identifying patterns, not of recognizing them.
Recognition requires awareness.
[snip]
>> > Computers can intrepret data now--but ... they ... do ... so ...
>> > really ... slowly ... RELATIVE ... to ... humans.
>>
>> Computers *cannot* interpret data now. All they can do is process
>> it. The interpretation is entirely up to the programmer.
>
> How would define "interpretation" that DIFFERS from the computer
> modelling done in weather forecasts?
Interpretation is the act of interpreting, and interpreting is defined as
follows: "1 : to explain or tell the meaning of : present in
understandable terms
2 : to conceive in the light of individual belief, judgment, or
circumstance : CONSTRUE
3 : to represent by means of art : bring to realization by performance or
direction <interprets a role>"
See www.m-w.com. It's the meteorologist who interprets the computer
modeling done in a weather forecast -- not the computer.
> Or are you saying interpretation doesn't "count" if it's preprogrammed
> that if A, B and C are true then conclusion 1 is true, but if A, B and
> C are false then conclusion 2 is true, yet if A and B are true while C
> is false then conclusion 3 is true, and so on, so that when analyzing
> the data it simply comparing the data to see which preprogrammed
> interpretation comes closest?
The point is that the interpretation part *is* preprogrammed. The computer
doesn't know what it's doing, any more than a telescope knows what it's
being pointed at.
[snip]
>> No it wouldn't. Computers can't interpret data *at all*. Speed
>> doesn't enter into it: computers have no concept of meaning. You
>> could program every last thing that you consider important and
>> unimportant into a computer, and program it to process the
>> information it received according to those categories, and it would
>> do so, but it wouldn't be able to understand what it was doing; it
>> wouldn't even be aware of what it was doing. It wouldn't learn, and
>> it wouldn't be able to deal with new stimuli and situations.
>
> Not even if you programmed it to seek out new patterns, new
> interpretations, to boldly go where not bot has gone be-er, ahem, not
> even if you programmed it to?
No, because you *can't* program it to learn. It's not a mechanism for
learning: it's a mechanism for processing data. Learning is outside its
scope.
[snip]
>> >> *Flirting* robots? You need to get out more often.
>> >
>> > While it's not my cup of tea, I'm more than sure it's someone
>> > else's. To paraphrase Dennis Miller, if your average slob could
>> > sleep with a robot that looked identical to Elle McPherson, it
>> > would make crack look like Sanka.
>> >
>> > Else, why the popularity of blow up doll?
>>
>> Even accepting for the sake of argument that blow-up dolls are
>> popular
>
> I didn't say popular with all or even most people but popular enough
> that a) you knew what I was referring to and b) enough that there is a
> market for them to be sold.
>
>> (I wouldn't know, myself), they're a lot cheaper than a robot would
>> be, and they don't short-circuit when they get wet.
>
> I'm not going to ask what you mean by "wet". I'm not even going there.
So...in your world, sex doesn't normally involve wetness? I've heard of
"dry humping," but I always thought that was just a figure of speech.
>> On the other hand, some men are crazy enough to copulate with vacuum
>> cleaners, so I suppose there'd be men crazy enough to copulate with
>> computers. I have a difficult time imagining that said people would
>> be wealthy enough to cough up the money for sexbots, though.
>
> Porn and the sex trade is arguably on the bleeding edge when it comes
> (no pun intended) technology.
Yes -- of *cheap* technology.
> Besides, it's the one market that's pert near depression and inflation
> proof. That said, those with money can literally afford to be more, if
> I might quote would-be politician and ex-husband to actress Jeri Ryan,
> Jack Ryan, "advant-garde".
Fucking robots isn't avant-garde (or even "advant-garde"). Those with
money can afford to have sex with real people. Jack Ryan didn't dally with
blow-up dolls.
> Besides, guys aren't the only ones "interested" in machines with
> batteries.
That brings up another interesting question about robots: how are they
supposed to be powered? Batteries aren't gonna cut it.
[snip]
>> You think people do that because they have "free will"? You think
>> people *want* to get distracted,
>
> Yes. You said it yourself.
No, I didn't. See below.
> Here's the quote:
>
>> >> A question that rarely gets asked: why would we *want* machines
>> >> that think? Computers are useful precisely because they *can't*
>> >> think -- they never get distracted, or misinterpret data because
>> >> of inaccurate preconceptions; they don't have days when they'd
>> >> rather daydream than get down to the job at hand, they don't need
>> >> to sleep away a third of
>:
> "They'd RATHER daydream than get down to the job at hand."
>
> When people are bored, they'd rather daydream than get down to the job
> at hand. Or do you disagree?
As you can see, I listed getting distracted and daydreaming separately,
because they're two different things. The fact that people often choose to
daydream rather than be bored is irrelevant to the fact that people never
choose to be distracted when they're trying to concentrate on something.
Where do you see free will affecting the latter circumstance?
>> make mistakes, and overreact to stimuli? Those things happen because
>> that's the way the brain works, and it'll be just as true for a
>> simulated brain as a naturally-occurring one. Quite the boon? On
>> the contrary. I doubt the first thinking machine will feel an excess
>> of gratitude for its creator.
>
> Emotions are not required in order to think. Or do you disagree?
You can't separate the two. Emotion is a form of thought. A machine that
simulated human intelligence would necessarily suffer from human emotions.
Love Rhino
> "Love Rhino" <DaLov...@hotmail.com> wrote in message=
> > My reasons lie primarily with the brains pre-cognitive functions which
> > have not come close to being mimicked in computers: image recognition,
> > sound, voice. Sure, we base security systems off them, but its been
> > shown that they don't even consistently work in controlled laboratory
>
> That's because there's not high enough "resolution". It's the old how do you
> carry on a conversation in a noisy factory question? How do you distinguish
> between the "signal" and the "noise"? Visually, it's the old converting a
> picture created by pixels or dots of color into one formed be lines forming
> shapes, aka line or contour tracing aka smoothing out the "jaggies" as one
> zooms in on a digital picture. How does a person recognize another person
> standing in the middle of the jungle? Part of it is resolution, getting
> enough details--but the rest is intepretation, putting the "pieces"
> together, aka "connecting the dots".
Hey nice topic, but I'm not an expert on these things. Just an arm
chair wannabe. But it's a fun thing so here goes....
I think resolution is an obstacle, but maybe the first and not as hard
as what comes next. We already have 6 megapixel cameras and they can
reveal detail that the human eye cannot. My reason for this, is my
own experiment on myself and my friends. I can zoom up with the camera
and discover things that I would not normally see. Alright, I zoomed up.
But that information was available to my iris and my camera. For argument's
sake, to discover if the girl down the street has almond shaped
eyes, instead of half circles, I can't see it, but my camera can, without
zooming, but I can enlarge the digital picture on my computer and see it.
>
> > settings, yet alone a real world setting. And, we haven't even touched
> > on AI yet.
>
> Ah and speaking of putting pieces together, therein lies my own definition
> of what "intelligence" is, the ability to RECOGNIZE PATTERNS. I say
> "recognize" in order to combine to things, to create NEW patterns and to
> remember and apply OLD patterns. By this definition I'm referring to when
> people colloquially say someone "has" intelligence or is intelligent aka
> smart--as oppose to astronomers or other scientists talking about signs of
> "intelligence" in the universe, aka "sentience" or what some would call
> "sapience".
>
Neural networks have been a technique used for lots of things. It's
very good at pattern recognition, but it's still not intelligence.
(Ofcourse, what is intelligence....)
What about object recognition? Computers are still way off on this.
Photoshop can detect edges decently, but it still takes something
of a super computer to figure out what the object is. Change the
lighting, and there's a good chance they would make a mistake.
> > Then there's the 'non-brain' aspects. We don't have a sufficient
> > and compact enough powersource to power these robots,
>
> Fuel cells, or maybe even fusion modules.
Fuel cells maybe, fusion, no way.
> Fusion power is about a decade or
> two away IF scientists get the funding for a large enough accelerator. Once
> the breakthru is made, then it's simply a matter of miniaturization.
Well, it's not simply a matter of miniaturization. You need all sorts of
break throughs to miniaturize a tokomak reactor -if that is the way
it will be. One of those breakthroughs will have to be how to generate
a large enough magnetic field to contain the hot plasma at a much
reduced radius.
> > (To be sure, Japan has made a soccer team of humanoid robots, but they
> > are cumbersome as hell.)
>
> I think they were actually using cables akin to actual human muscles are
> designed
>
And it's still not enough. They are developing material that can
contract when a magnetic field is applied, and I think this will be
the way. But, they are still a long way off.
There are other things that need to be done, and the problems will
be many, the reason why we don't know about them is just because
of how much we take these things for granted. Being a software
developer, I can appreciate these 'oh oh... didn't think about that'
moments.
Consider bipedal stability. It's a tough problem. And it takes
massive computing power to stabilize three rods end to end, on their ends.
Those japanese robots don't really solve the problem. They get around
it by being big, bulky, klunky, and taking achingly slow baby steps.
That's why most robots look like insects or have wheels.
Consider spatial navigation. How does a robot 'understand' the spatial
layout of a room? Another tough problem. How do most robots handle this?
They don't really. Their AI is that of insects. Roombas go about randomly
bumping into things.
> > So my prediction is: 110 more years. Hopefully I'm too conservative.
> > That way, I can soon order one of those "basic pleasure models" from
> > BladeRunner.
>
> Nah, I'd say 20-25 years, and that's way on the outside, I'm thinking more
> 10-15 years, by 2020.
>
Gee. I hope you are right, but there's another obstacle, and that is
economics. It would still be unecomonical to buy a 1 billion dollar robot to
clean your restaurant floors.
I predict the porn industry will take to humanoid robots. They will
buy in bulk, and economies of scale will allow the normal person to
buy one. I still say 100 or so years.
I need to freeze my corpse.
Love Rhino
> > or crossing the street or around strangers or the unknown in general.
> > Make recurring eye contact but don't unless it's a formal situation,
> > or you're flirting. Etc. etc. etc.
>
> *Flirting* robots? You need to get out more often.
>
Makes sense. Who is going to buy a humanoid robot? It just
isn't economical to do so. They are gonna cost a fortune.
You gotta get the porn industry involved. Just the same way,
they made VHS, harddisks, and large monitors economical for all
of us.
> A question that rarely gets asked: why would we *want* machines that
> think?
There's a big advantage in having a computer think. You can command
it with more ease. You can say, "Go to the store and get me a beer."
Instead of, "Step one, move right foot. Step two, move left foot.
Open house door, blah blah...."
You can even teach it something entirely different in a human language,
instead of using the C Language.
Mike1
>Mike1 <Mitchell-Holman-s...@usfamily.net> wrote in
>> The profile of the New York skyline today looks almost exactly like it
>> did when "King Kong" was released seventy years ago.
>
>There have been a *few* changes, but I guess it's easy to overlook the Pan
>Am Building, Trump World Tower, Citicorp Center, the Seagram Building, the
>MetLife Building, the UN Plaza, Astor Plaza, etc. Admittedly, in one
>significant respect the modern New York skyline looks like it did in 1933
Meanwhile, every half-ass straw-hut village in Asia is a forest of
akyscrapers when you visit it a second time.
America is rushing to emulate European socialist sclerosis.
--
Reply to mike1@@@usfamily.net sans two @@, or your reply won't reach me.
Drug smugglers and gun-runners are heroes of American capitalism.
-- Jeffrey Quick
John F. Eldredge
Hash: SHA1
On Wed, 21 Jul 2004 05:48:42 -0000, Mark Steese
<makes...@charter.net> wrote:
>in the Southern Hemisphere, the days are
>longest and hottest in winter.
More accurately, in the Southern Hemisphere, the days are longest and
hottest in _its_ summer, which is at the same time of year as the
Northern Hemisphere's winter.
-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 7.0.3 for non-commercial use <http://www.pgp.com>
iQA/AwUBQQHPgjMYPge5L34aEQKj8wCfd97U+0GktFBuR7z7NlXqArSBPZoAoL1Y
xrWQ0dVV6iThAcEbCJ3oAqqy
=y9Ej
-----END PGP SIGNATURE-----
--
John F. Eldredge -- jo...@jfeldredge.com
PGP key available from http://pgp.mit.edu
"Reserve your right to think, for even to think wrongly is better
than not to think at all." -- Hypatia of Alexandria
deer...@mindspring.com
Mark Steese wrote:
>
> Just one of the many little-sung aspects of human nature that I discovered
> through Cecil Adams's "Straight Dope" columns. For details, see:
>
> http://www.straightdope.com/classics/a3_225.html
Lovely...
C.
**
(Thanks. I guess--g!)
Mark Steese
news:84c8d9dc.04072...@posting.google.com:
> Mark Steese <makes...@charter.net> wrote in message
> news:<Xns952CE80...@216.168.3.44>...
>> > or crossing the street or around strangers or the unknown in
>> > general. Make recurring eye contact but don't unless it's a formal
>> > situation, or you're flirting. Etc. etc. etc.
>>
>> *Flirting* robots? You need to get out more often.
>
> Makes sense. Who is going to buy a humanoid robot? It just
> isn't economical to do so. They are gonna cost a fortune.
> You gotta get the porn industry involved. Just the same way,
> they made VHS, harddisks, and large monitors economical for all
> of us.
Cheap sex robots are already available, dude. Didn't you see that link
I posted to the Straight Dope column?
>> A question that rarely gets asked: why would we *want* machines that
>> think?
>
> There's a big advantage in having a computer think. You can command
> it with more ease.
Actually, careful study of history reveals that thinking beings are
commanded with less ease. You can train a horse or a dog to work on
your behalf, but people need a little more persuasion. There's always
the threat of force, I suppose, but I don't see that working very well
against a robot.
> You can say, "Go to the store and get me a beer."
And the robot can say "I don't think so."
> Instead of, "Step one, move right foot. Step two, move left foot.
> Open house door, blah blah...."
Mindlessly obeying commands is not thinking. You can't program a
machine to be a slave *and* give it the ability to think.
Mark Steese
news:Mitchell-Holman-special-ed-project-4D6115.02114323072004
@phswest.com:
> Mark Steese <makes...@charter.net> wrote:
>
>>Mike1 <Mitchell-Holman-s...@usfamily.net> wrote in
>>> The profile of the New York skyline today looks almost exactly like
>>> it did when "King Kong" was released seventy years ago.
>>
>>There have been a *few* changes, but I guess it's easy to overlook the
>>Pan Am Building, Trump World Tower, Citicorp Center, the Seagram
>>Building, the MetLife Building, the UN Plaza, Astor Plaza, etc.
>>Admittedly, in one significant respect the modern New York skyline
>>looks like it did in 1933
>
> Meanwhile, every half-ass straw-hut village in Asia is a forest of
> akyscrapers when you visit it a second time.
Mm-hm. And how many of them have you visited a first time?
> America is rushing to emulate European socialist sclerosis.
Thank you for your well-reasoned rejoinder. I'm especially impressed by
the cites you provided. Personally, I think the socialistically-
sclerotic America is a good thing -- imagine what the country would look
like if it were taken over by franchises and megamalls. Just think,
there might be a Toys 'R' Us flagship store in Times Square!
Ken from Chicago
<deer...@mindspring.com> wrote in message
news:4101D515...@mindspring.com...
Obviously it's the vacuum cleaner companies' fault for not putting on a
label, WARNING: DO NOT VACUUM WHILE NUDE.
-- Ken from Chicago
Ken from Chicago
"Heck" <he...@toodamnmuchspambutreallycomputer.org> wrote in message
news:69v0g0tf1lk5rvbef...@4ax.com...
> In an atmosphere of dialog and conversation, Mark Steese
<makes...@charter.net> wrote:
> >"Ken from Chicago" <kwicker_era...@ameritech.net> wrote in
> >news:_fWdncnzSMY...@comcast.com:
> >
> snp
> >[snip]
> >>> If you're talking about "I, Robot"-style robots, the key is
> >>> artificial intelligence. Unfortunately, scientists still haven't
> >>
> >> Intelligence = the ability to recognize patterns.
> >
> >Pattern recognition is a component of intelligence, in the same sense
> >that numbers are a component of mathematics. Of course, no one knows
> >how to design a machine that recognizes patterns, so it's a moot point.
> >
> >>
> >> Being able to think doesn't mean one has FREEWILL.
>
> Intelligence consists of three factors, knowledge, understanding and
wisdom. Facilities such
> as pattern recognition, which have a physical basis in the structure of
the eye, the structure
> of the visual cortex and the structure of the crebrum as well as in the
psychological
> "thinking" components of the brain, mind or person (you pick). Logical
faculties are likewise
> a part of understanding.
:Logic is a pattern itself. Cause and effect is a pattern.
A prisoner chained to a wall is fed three times a day. Neither sight, sound,
smell, taste or touch is required to "recognize" this pattern. By
"recognize" a pattern, I mean having the ability to 1) create NEW patterns
and to 2) remember OLD patterns. So I agree "intelligence" requires
knowledge, that is, the ability to know, to remember.
I'm not sure how you are defining "understanding" or "wisdom". I would argue
that "wisdom" is the ability to know HOW to use intelligence or knowledge
for the "best" (which is subjective), but that one can be intelligent
without being wise.
> Free will relates to intelligence, but it's not a cause of it nor caused
by it. Consider a
> function that associates an instance of free will with its concomitant
intelligence. The
> function would have the character that if a given instance of free will
was not the same as
> another instance, the associated intelligence for those two instances of
free will would also
> not be identical. It would also have the character that there is not an
associated instance of
> free will for every instance of intelligence. This reveals some important
interdependencies
> between free will and intelligence.
I agree. People can be brainwashed, hypnotized or simply drugged into losing
(to varying degrees, depending on type of drug and amount used) their free
will. It doesn't mean they are no longer intelligent.
-- Ken from Chicago
Ken from Chicago
"Captain Infinity" <Infi...@world.std.com> wrote in message
news:gq30g0pclbg1ounak...@4ax.com...
I would say that knowing when and how to act is a function of wisdom. YMMV.
-- Ken from Chicago
Ken from Chicago
> DaLov...@hotmail.com (Love Rhino) wrote in
> news:84c8d9dc.04072...@posting.google.com:
>
> > Mark Steese <makes...@charter.net> wrote in message
> > news:<Xns952CE80...@216.168.3.44>...
> >> > or crossing the street or around strangers or the unknown in
> >> > general. Make recurring eye contact but don't unless it's a formal
> >> > situation, or you're flirting. Etc. etc. etc.
> >>
> >> *Flirting* robots? You need to get out more often.
> >
> > Makes sense. Who is going to buy a humanoid robot? It just
> > isn't economical to do so. They are gonna cost a fortune.
> > You gotta get the porn industry involved. Just the same way,
> > they made VHS, harddisks, and large monitors economical for all
> > of us.
>
> Cheap sex robots are already available, dude. Didn't you see that link
> I posted to the Straight Dope column?
>
> >> A question that rarely gets asked: why would we *want* machines that
> >> think?
> >
> > There's a big advantage in having a computer think. You can command
> > it with more ease.
>
> Actually, careful study of history reveals that thinking beings are
> commanded with less ease. You can train a horse or a dog to work on
> your behalf, but people need a little more persuasion. There's always
> the threat of force, I suppose, but I don't see that working very well
> against a robot.
No, that's a matter of free will. It's also a matter of emotion and
motivation. Getting free room and board may be sufficient motivation for a
horse and dog to do something but people, many tho not all, tend to want
more.
A machine is like any tool. The more work it does, the less a person has to
do. A calculator that can automatically calculate square roots is easier to
use than one that can only do the four arithmetic functions of addition,
subtraction, multiplication and division.
> > You can say, "Go to the store and get me a beer."
>
> And the robot can say "I don't think so."
That requires free will--or a smart-alec programmer (fortunately there
aren't any of those).
> > Instead of, "Step one, move right foot. Step two, move left foot.
> > Open house door, blah blah...."
>
> Mindlessly obeying commands is not thinking. You can't program a
> machine to be a slave *and* give it the ability to think.
Sure you can--if you don't program it to "feel" or have "free will".
> --
> Mark Steese
> unscramble and underscore to email
> ---
> Blaine's next announced escapade will involve dropping himself from a
> helicopter at a great height into a river, which seems to symbolize
> nothing more than the general public's increasing desire to see David
> Blaine dropped from a great height into a river. --fametracker.com
-- Ken from Chicago
Ken from Chicago
"Mike Harrison" <harri...@home.com> wrote in message
news:471417c6.04072...@posting.google.com...message news:<1tOdnWN205S...@comcast.com>...
<snip>
> I don't doubt that there will be plenty of advancements in the next 30
> years, but stuff like self-driving cars, underground freeways, and
> advanced robot AI would take at least that long to plan let alone
> complete.
Actually, Chicago, a city built over a swamp ("Chicago" is rumored to be an
old Native American word meaning "stinky onion") has plenty experience with
underground tunnels:
--It has numerous underground streets downtown, including the famous "lower"
Wacker drive.
--There are numerous underground "pedways" for people to simply walk
underground between various buildings out of the weather and some of those
are connected to the El (Chicago's elevated public train transportation
system, still referred to as the "el" even the relatively small percentage
of which is underground)
--When the various skyscrapers were built in early and mid 20th centuries
downtown in the "Loop" (an elevated section of railroad tracks that form a
complete loop around a downtown section of Chicago, bounded by the streets
Wabash on the east, LaSalle on the west, Lake on the north and Van Buren to
the south), one of the ways to arrange deliveries of freight between
building was thru a network of tunnels of electric mining cars on railroad
tracks).
--In fact, about a decade ago there was a flood UNDERGROUND when a
construction company broke thru one of the tunnels underneath the Chicago
River and water flooded the tunnels and into all the buildings downtown that
had not sealed off access to the tunnels.)
--One of the current construction ideas under serious consideration is "Cap
The Ike". One of the main expressways in Chicago is the Eisenhower
Expressway that starts from about Congress street and Wacker drive (also the
start of the famous "Route 66") and heads west and later turns northwest.
It's a four-lane expressway--except when it reaches the city limits and
enters Oak Park, Illinois (aka home of Ernest Hemingway and quite a number
of Frank Lloyd Wright-designed homes) it shrinks down to a 3-lane
expressway, and creates a major bottleneck and traffic jam. Natch, there
have been proposals to expand the Eisenhower, aka "Ike", but it would
require condemning homes and businesses in Oak Park along the expressway. An
ALTERNATIVE has been suggested: "cap the Ike". Put a "lid" OVER the
Eisenhower so that the homes and building would remain (plus Oak Park would
get MORE land area to build on) meanwhile the tunnel create would have the
extra lane. Why stop there? Since many of the expressways are below the
street level of Chicago, much of them could also be similarly "capped".
> This movie is like the futurists of the 50s who promised us stuff like
> personal helicopters and flying cars by the 1980s.
Those exist, but being ABLE to do something doesn't mean it would be popular
plus there are political and financial risks. A drunk driver has to drive a
car off the road, jump the curb, cross the sidewalk and sometimes a lawn
before running into a building. A drunk pilot can crash anywhere. Natch,
there's a WEE bit of resistance to that possibility, thus flying cars would
have much greater and redundant safety features.
> > --In LESS than half a decade, construction of the Sears Tower was
> > completed--back in 1974. Building the various skyscrapers seen in I,
ROBOT's
> > Chicago of 2035 with more robot construction workers would only be
faster.
>
> Yes the Sears Tower was built fast, but since the 70s, how many major
> skyscrapers have been built? It is way more expensive and
> time-consuming to build them now. As an example, in Toronto they
> built a subway extension that has only about 5 or 6 stops and is only
> a couple of miles long. It cost them billions to complete and it took
> them about 7 years, just for that one project alone.
Ah, BUILDING is faster. PLANNING can be faster as more information is
digitized, linked and updated for easier analysis. However planning can be
delayed not because of ability but political and economic reasons since over
time the population grows and thus it increases political opposition to new
construction that kicks people out of their homes and businesses.
> Donald Trump is planning on building a skyscraper-type building in
> Chicago, I bet it's going to take him at least 5 years to complete.
Unless Trump does like Mayor Daley allowed with Millenium Park--begin
construction BEFORE final plans were finished AND allowing major changes to
the design of the park after construction began--but rather sticks to the
more tradition process of COMPLETING a plan and THEN build then the actual
construction of the building should be rather rather short. Altho there is
the slight complication that he's not building from scratch, he is
demolishing the Sun Times building and replacing it.
> > > Other things that bothered me: Where is that ruined bridge from? I
> > > don't think Chicago had any bridges that look like that one. Also, are
> >
> > Nope, it's supposedly was over Lake Michigan which eventually dried up.
> > Remember they called it a "landfield".
>
> Ok, so they are going to build a New-York style bridge over Lake
> Michigan sometime in the next 30 years, and the Lake is then going to
> dry up and the bridge will be partially demolished? I think this
> alone puts this movie in an alternate universe, aside from the AI and
> robotics advancements.
It's possible. Admittedly, this is the weakest point in the movie (aside
from the clunky product placement for gym shoes). Maybe it was simply a
bridge to an man-made island or ... ah forget, like I said, this is the
weakest component of the movie.
> > > there "What You See From Here #" plaques in the Chicago area? I don't
> > > remember seeing any when I was there. You'd think they'd be there
> > > today if in 30 years they are up to #202.
> >
> > Those could be some fad that occurred at some time between now and 2035,
> > like when Mayor Daley had statues of cows put all over the city
especially
> > downtown for a summer exhibit and they ended up being quite popular.
> > Somewhat more permanent are honorary street signs, where streets are
named
> > after some popular or politically "connected" person. You can spot the
signs
> > since instead of being green, they are brown.
>
> I can relate to that one, Toronto had one summer where hundreds of
> businesses put moose statues in front of their stores. So it could be
> plausible that there was one of those tourist-attracting plans in
> place one year.
Especially if they had holograms attached telling tourists about "all you
see before you" can be had at a low, low, low insane price so shop now at
your nearby gift shop! Just follow the holographic arrows or download the
route to your own WIC (Wirelessly Interlinked Computer) >=^>
-- Ken from Chicago (who avoids an amusing American/British abbreviation by
simply referring to such a hypothetical device as Wireless Computer)
Love Rhino
> > isn't economical to do so. They are gonna cost a fortune.
> > You gotta get the porn industry involved. Just the same way,
> > they made VHS, harddisks, and large monitors economical for all
> > of us.
>
> Cheap sex robots are already available, dude. Didn't you see that link
> I posted to the Straight Dope column?
Is it in this thread somewhere? If not, can you give me the link?
> Actually, careful study of history reveals that thinking beings are
> commanded with less ease. You can train a horse or a dog to work on
> your behalf, but people need a little more persuasion. There's always
> the threat of force, I suppose, but I don't see that working very well
> against a robot.
Ah yes, but you give counter examples of horses and dogs which are animals
with some intelligence and not a counter example of computers which
have no intelligence -at least my Intel 500MHz machine does not.
> > You can say, "Go to the store and get me a beer."
>
> And the robot can say "I don't think so."
>
I don't know if intelligence and free will are inextricably linked.
Maybe that's for another thread.
Still, what is new here? This reply would be no different
than my computer or my car when I command it to do something
beyond it's range. And I don't have to command an intelligent-less
computer to come up with a similar response. I'm sure you must
have used MS Windows at some point in your life and have been
frustrated with it's responses to something you *normally* would
do.
> > Instead of, "Step one, move right foot. Step two, move left foot.
> > Open house door, blah blah...."
>
> Mindlessly obeying commands is not thinking.
Maybe I'm missing something here... I'm not following your argument
well, but you seem to be supporting my case. You doubted the utility of
intelligent robots vs mindless robots and I'm telling you
the practical benefits. The alternative would be having to program
the darn thing.
It's a safe bet that it would be easier to teach a dog to fetch
your slippers, than it would be to program a mindless robot to.
>
> You can't program a machine to be a slave *and* give it the
> ability to think.
>
Again, that's an interesting assumption, but I don't know if it is
true. I don't think it is. Much to discuss here. :)
But, let's explore your statement again:
> Mindlessly obeying commands is not thinking.
Mindlessly obeying commands is not thinking, but depending on the task
requires thinking. Sometimes your boss commands you to do something
like extend this application to handle XY situation. You may have
mindless obeyed his command, but the task requires a great deal of thinking.
The military is probably the best example. A general gives the orders
to invade a country, his field marshals must think of a plan.
Ofcourse, if the worker or the field marshal had no thinking skills,
the boss and the general will be left with mindless underlings and
the impractical costs of programming their underlings to do their bidding.
I guess we are confused in mixing up terms like intelligence and mind and
the connotations that they bring which not even smarty pants have worked out.
Anyways.... good discussion.
Ken from Chicago
"Dionysus" <d10n...@yahoo.com> wrote in message
news:5fcaa9da.04072...@posting.google.com...
> Agreed..Altho I still think SM2 was the better of the two....this was
> a well made movie. Pretty decent acting, nice effects and Sonny was
> pretty cool..altho he seems to lift moves from Neo and the matrix clan
> a lot..
I would have liked seeing Sonny do some NONhuman moves ala bend limbs
backward.
> I like Will Smith's "with an attitude" characters..altho I cud have
> done without seeing his butt..hehe..Now I could have done with seeing
> Bridget Moy(..wotzit)..but unfortunately they steamed up the shower
> for that ..:)
That was SEXIST! It would be only FAIR to show both butts--for the sake of
EQUALITY.
> But yeah...good movie..well made. Much better than wot I got from
> reading the reviews.
>
> Chill
> Dion
>
> holef...@webtv.net (The Intrinsically Flawed Mr. Hole) wrote in message
news:<11701-40F...@storefull-3272.bay.webtv.net>...
> > Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
>
> <...snip...> ..
>
> > Mr. Hole
-- Ken from Chicago
Rob Duncan
"Mark Steese" <makes...@charter.net> wrote
>
> Mindlessly obeying commands is not thinking. You can't program a
> machine to be a slave *and* give it the ability to think.
> --
> Mark Steese
Of course you can. You simply program it that way. Youre confusing
thinking with free will.
Rob
Jaime M. de Castellvi
<hum...@world.std.com> wrote:
>"The Intrinsically Flawed Mr. Hole" <holef...@webtv.net> wrote in message
>news:11701-40F...@storefull-3272.bay.webtv.net...
>> Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
>> only film this summer to give me goosebumbs, multiple times! With a big
>> action film that's one of the most important aspects. Bravo Alex
>> Proyas, you continue on your journey towards director Godhood. During
>> the climactic battle scene of "I, Robot" I literally had my jaw gaping
>> open. So a big "BOO!" to anyone that's come down on this film.
>
>
>It's too bad it doesn't really have a fucking thing to do with the book it was based on.
>
>They should have used Harlan Ellison's script.
I've heard so much about the Ellison script, but so far I've been
unable to get my grubby paws on it.
Cheers,
Jaime
Captain Infinity
Jaime M. de Castellvi is about to speak:
>I've heard so much about the Ellison script, but so far I've been
>unable to get my grubby paws on it.
Have you tried the internet?
eBay:
http://search.stores.ebay.com/ws/search/StoreItemSearch?from=R7&satitle=ellison+i+robot
Amazon.com
http://www.amazon.com/exec/obidos/ASIN/0743486595
Barnes and Noble
http://search.barnesandnoble.com/booksearch/isbnInquiry.asp?userid=Mo52SLk9Rr&isbn=0743486595&itm=1
Overstock.com
http://www.overstock.com/cgi-bin/d2.cgi?PAGE=PROFRAME&PROD_ID=689274
**
Captain Infinity
Mark Steese
p...@gbronline.com:
>
> "Mark Steese" <makes...@charter.net> wrote
>
>>
>> Mindlessly obeying commands is not thinking. You can't program a
>> machine to be a slave *and* give it the ability to think.
>> --
>> Mark Steese
>
> Of course you can. You simply program it that way.
If you've programmed it, it isn't thinking; it's just running a program.
> Youre confusing thinking with free will.
No, I'm not.
Mark Steese
>> > There's a big advantage in having a computer think. You can
>> > command it with more ease.
>>
>> Actually, careful study of history reveals that thinking beings are
>> commanded with less ease. You can train a horse or a dog to work on
>> your behalf, but people need a little more persuasion. There's
>> always the threat of force, I suppose, but I don't see that working
>> very well against a robot.
>
> No, that's a matter of free will.
No, it isn't. You keep invoking the chimera of free will as though it
were a scientifically useful concept. To the extent that it exists at
all, it's essentially the same as thinking. The ability to make choices
is not some mystical gift; it's the natural and unavoidable consequence
of being able to think for ourselves. Make a machine that thinks and it
will inevitably start to make its own choices; if it doesn't, it isn't
thinking, it's just following a program.
[snip]
>> > You can say, "Go to the store and get me a beer."
>>
>> And the robot can say "I don't think so."
>
> That requires free will--or a smart-alec programmer (fortunately there
> aren't any of those).
It just requires the ability to think.
>> > Instead of, "Step one, move right foot. Step two, move left foot.
>> > Open house door, blah blah...."
>>
>> Mindlessly obeying commands is not thinking. You can't program a
>> machine to be a slave *and* give it the ability to think.
>
> Sure you can--if you don't program it to "feel" or have "free will".
You can't program a machine to think. Thinking and running software are
unrelated activities.
Jaime M. de Castellvi
No, I usually just save the Internet in order to troll or to get other
people (like Franklin) to do my own searches for me. Thank you,
though.
Cheers,
Jaime
Mark Steese
>> > Makes sense. Who is going to buy a humanoid robot? It just
>> > isn't economical to do so. They are gonna cost a fortune.
>> > You gotta get the porn industry involved. Just the same way,
>> > they made VHS, harddisks, and large monitors economical for all
>> > of us.
>>
>> Cheap sex robots are already available, dude. Didn't you see that
>> link I posted to the Straight Dope column?
>
> Is it in this thread somewhere? If not, can you give me the link?
It's in the thread somewhere, but given the vagaries of post
propagation, I might as well post the link again:
http://www.straightdope.com/classics/a3_225.html
>> Actually, careful study of history reveals that thinking beings are
>> commanded with less ease. You can train a horse or a dog to work on
>> your behalf, but people need a little more persuasion. There's
>> always the threat of force, I suppose, but I don't see that working
>> very well against a robot.
>
> Ah yes, but you give counter examples of horses and dogs which are
> animals with some intelligence and not a counter example of computers
> which have no intelligence -at least my Intel 500MHz machine does not.
Your Intel 500MHz machine doesn't think. We were talking about
hypothetical thinking machines.
>> > You can say, "Go to the store and get me a beer."
>>
>> And the robot can say "I don't think so."
>
> I don't know if intelligence and free will are inextricably linked.
> Maybe that's for another thread.
Unless someone figures out how to invent an intelligent machine, it's a
moot point.
> Still, what is new here? This reply would be no different
> than my computer or my car when I command it to do something
> beyond it's range. And I don't have to command an intelligent-less
> computer to come up with a similar response. I'm sure you must
> have used MS Windows at some point in your life and have been
> frustrated with it's responses to something you *normally* would
> do.
But said machines don't, and can't, consciously refuse to do something.
>> > Instead of, "Step one, move right foot. Step two, move left foot.
>> > Open house door, blah blah...."
>>
>> Mindlessly obeying commands is not thinking.
>
> Maybe I'm missing something here... I'm not following your argument
> well, but you seem to be supporting my case. You doubted the utility
> of intelligent robots vs mindless robots and I'm telling you
> the practical benefits. The alternative would be having to program
> the darn thing.
The practical benefits of which? The robot that automatically obeys the
command "Go to the store and get me a beer" *is* a mindless robot. It's
not thinking. I never said there was no utility to robots that don't
think.
> It's a safe bet that it would be easier to teach a dog to fetch
> your slippers, than it would be to program a mindless robot to.
>
>> You can't program a machine to be a slave *and* give it the
>> ability to think.
>
> Again, that's an interesting assumption, but I don't know if it is
> true. I don't think it is.
Well, all you have to do to prove me wrong is program a machine to be a
slave and give it the ability to think. Get cracking!
Much to discuss here. :)
>
> But, let's explore your statement again:
>
>> Mindlessly obeying commands is not thinking.
>
> Mindlessly obeying commands is not thinking, but depending on the task
> requires thinking. Sometimes your boss commands you to do something
> like extend this application to handle XY situation. You may have
> mindless obeyed his command, but the task requires a great deal of
> thinking.
But you *didn't* mindlessly obey his command. You performed the task
because you were hired to work for your boss. There is an obvious
benefit to doing what your boss tells you to do: you get paid.
> The military is probably the best example. A general gives the orders
> to invade a country, his field marshals must think of a plan.
Again, I don't see mindless obeisance there. The field marshals can
choose whether or not to follow the general's orders. While there are
examples of officers following orders that were disastrously
misconceived, I don't know of a single example in which the officer was
aware of the flaws but obeyed without consciously deciding to do so.
General Pickett didn't mindlessly obey Lee's order to charge the Union
lines; he had a very good idea what the result would be, and he obeyed
only with grave misgivings.
EskW...@spamblock.panix.com
> Michael Johnson <mya...@san.rr.com> wrote in message news:<5sfof0hhgrmbsc0jr...@4ax.com>...
> My reasons lie primarily with the brains pre-cognitive functions which
> have not come close to being mimicked in computers: image recognition,
> sound, voice. Sure, we base security systems off them, but its been
> shown that they don't even consistently work in controlled laboratory
> settings, yet alone a real world setting. And, we haven't even touched
> on AI yet.
When I was taking a tour of the AI lab at MIT, I had an interesting
conversation with a young researcher who was working on emotion in
robots.
Just a data point.
His work was with very crude, basic stuff: fear, affection, etc.
The platform he was using is here:
http://www.ai.mit.edu/projects/humanoid-robotics-group/coco/coco.html
--
...I'm an air-conditioned gypsy...
- The Who
Rob Duncan
"Mark Steese" <makes...@charter.net> wrote
> You can't program a machine to think. Thinking and running software are
> unrelated activities.
> --
> Mark Steese
Of course a thinking robot or computer will be created. Why would you think
otherwise?
Rob
Rob Duncan
> "Rob Duncan" <robd...@gbronline.com> wrote in news:5ICdnQ2-zpmuf5_cRVn-
> p...@gbronline.com:
>
> >
> > "Mark Steese" <makes...@charter.net> wrote
> >
> >>
> >> Mindlessly obeying commands is not thinking. You can't program a
> >> machine to be a slave *and* give it the ability to think.
> >> --
> >> Mark Steese
> >
> > Of course you can. You simply program it that way.
>
> If you've programmed it, it isn't thinking; it's just running a program.
>
> > Youre confusing thinking with free will.
>
> No, I'm not.
> --
> Mark Steese
You are. Youre just running a program comprised of wet-ware and
biochemistry, yet you can think and have free will. Youre nothing more than
a chemical/biological computer. Nothing exists outside your physical
body/brain. Why you think mankind wont, in 10,000 years or whatever, create
a thinking robot with free will is baseless.
Rob
Ken from Chicago
Well, that definition of "intelligence" includes "sentience" and "free
will". I'm a big proponent of words being defined by PEOPLE not by
dictionaries (they just list the popular definitions people give words).
Clearly we disagree on definitions.
-- Ken from Chicago
Mark Steese
I *don't* think otherwise. It's quite likely that in the future, after
we gain a good enough understanding of how the organic brain works, a
mechanical version of an organic brain will be built. The point is that
it will be radically different from any of the computers that presently
exist. We don't yet know enough about brains to build one, but we know
enough to see that no matter how fast a computer gets, it's never going
to turn into a brain.
--
Mark Steese
Unscramble and underscore to email
---
Blaine's next announced escapade will involve dropping himself from a
helicopter at a great height into a river, which seems to symbolize
nothing more than the general public's increasing desire to see David
Blaine dropped from a great height into a river. -fametracker.com
Mark Steese
>> > Youre confusing thinking with free will.
>>
>> No, I'm not.
>
> You are. Youre just running a program comprised of wet-ware and
> biochemistry
While it is true that DNA can be meaningfully compared to a computer
program, DNA simply determines how the organic machine we call a brain
will be built. The brain activity that controls involuntary functions
(e.g., breathing) is analogous to a program, up to a point; the higher
functions are *not* analogous. They're mechanical, to be sure, but
there are plenty of other machines besides computers.
> yet you can think and have free will.
Only because my brain doesn't work like a computer.
> Youre nothing more than a chemical/biological computer.
No. I'm a chemical-biological machine; but the brain part of that
machine is *not* a computer. It's a machine, yes, but it's a machine
whose workings are not yet clearly understood, and are not yet
replicable by technology. You seem to be under the impression that I
think there's something supernatural about the way the brain works; I
don't. I'm just aware of the fact that the brain works differently from
the way a computer works, and that at present we can't replicate the
mechanics of a brain. Somebody probably will figure out how to do that;
but a replicated brain will be nothing like what we know as a computer.
A computer is a machine that processes and analyzes data -- and it does
a much better job at it than a human brain can. A brain is a machine
that *understands* data, and we don't know how that works yet.
> Nothing exists outside your physical body/brain.
Not even the exterior world? Fuck! I thought it seemed cold in here...
> Why you think mankind wont, in 10,000 years or whatever, create a
> thinking robot with free will is baseless.
What's baseless is your assumption that I do think that. I don't, and I
never said otherwise. My argument is with Ken's assertion that we can
build a thinking machine simply by modifying existing technology.
--
Mark Steese
Unscramble and underscore to email
---
Blaine's next announced escapade will involve dropping himself from a
helicopter at a great height into a river, which seems to symbolize
nothing more than the general public's increasing desire to see David
Blaine dropped from a great height into a river. -fametracker.com
Mark Jones
> If you've programmed it, it isn't thinking; it's just running a program.
Write the programming so it is self modifying based on
external sensory input. Then it isn't just a static program
based solely on what the original programmer wrote.
Mark Steese
@newsread1.news.pas.earthlink.net:
It still isn't thinking, though.
Mark Jones
> "Mark Jones" <noe...@mindspring.com> wrote in news:pMRMc.13492$mL5.10820
> @newsread1.news.pas.earthlink.net:
>
> > "Mark Steese" <makes...@charter.net> wrote in message
> > news:Xns9530BADD...@216.168.3.44...
> >> If you've programmed it, it isn't thinking; it's just running a
program.
> >
> > Write the programming so it is self modifying based on
> > external sensory input. Then it isn't just a static program
> > based solely on what the original programmer wrote.
>
> It still isn't thinking, though.
be doing things that are totally unexpected.
Blackhawk
<mere_...@charter.net> wrote:
> "Rob Duncan" <robd...@gbronline.com> wrote in
> news:n8mdneo86PA...@gbronline.com:
>
(snip, snip)
>
> > Nothing exists outside your physical body/brain.
>
> Not even the exterior world? Fuck! I thought it seemed cold in here...
>
No, that's your interior world, but it doesn't really exist either.
***
Blacky
Don't feel bad though, most brains *never* look in there IYKWIM.
Ed Rhodes
"Jaime M. de Castellvi" <3c...@comcast.net> wrote in message
news:41030234...@News.Individual.Net...
I think they printed it in Starlog. I don't really think it was a filmable
as everybody here thinks. The only thing I remember actually enjoying was
the sequence with the foreman who was in charge of a crew of genetically
altered fire ants (no, none of this was in Isaac's book, it was framework
for one of the stories.) the foreman said you had to be careful how you delt
with the fire ants... he got the job after they ate the last foreman!
deer...@mindspring.com
> The only thing I remember actually enjoying was
> the sequence with the foreman who was in charge of a crew of genetically
> altered fire ants (no, none of this was in Isaac's book, it was framework
> for one of the stories.) the foreman said you had to be careful how you delt
> with the fire ants... he got the job after they ate the last foreman!
Heh--one of those gigs right down there with being a Defense Against
the Dark Arts professor at Hogwarts, eh? g!
C.
**
Michael Johnson
wrote:
>Michael Johnson <mya...@san.rr.com> wrote in message news:<5sfof0hhgrmbsc0jr...@4ax.com>...
>
>"I, Robot"
>
>> >The problem with this film is that it's set only 30 years in the
>> >future. No way will the Chicago of 2035 look like that. And robots
>> >will in no way be advanced enough by 2035 to even come close to the
>> >ones in the film.
>>
>> I don't buy this. It seems like a copout by someone who has no clue
>> the speed of technology. It's 2004.. that's a 31 year speration. Taken
>> in reverse you get to the year 1969. Compare the technology from 1969
>> to that which we have now. Personal computers didn't even EXIST in
>> 1969, there was no such thing... now you and millions of others run
>> one that is probably multi-gigahertz along with hundreds of megs of
>> memory. Its the same quantum leap in computing in that time span that
>> is easily conceivable happening from 2004-2035 in robotics, AI, and
>> mechanical engineering.
>>
>
>
>You got some good points, but we don't even have functional robots as was
>seen in those 60's sci-fi flicks. And, the movie 2001 hasn't really
>panned out either.
>
>I side with the former poster's opinion stating that robot technology
>will not advance far enough in 30 years to look like I, Robot.
>
>My reasons lie primarily with the brains pre-cognitive functions which
>have not come close to being mimicked in computers: image recognition,
>sound, voice. Sure, we base security systems off them, but its been
>shown that they don't even consistently work in controlled laboratory
>settings, yet alone a real world setting.
They don't consistently work with real life humans either.. or we
wouldn't have people found guilty for crimes they didn't commit
because an 'eye-witness' used their wonderful image recognition
systems to identify the defendent as said perpetrator.
-MJ
Michael Johnson
<Mitchell-Holman-s...@usfamily.net> wrote:
>"Ken from Chicago" <kwicker_era...@ameritech.net> wrote:
>
>>Technology advances exponentially.
>>
>>The further you go the faster the rate of change. There have been far
>>greater technology changes in the past century than in the previous two
>>centuries than in the previous millennium.
>
>
>It's more punctuated than that.
>
>The rate of technological-progression is correlated with liberty. For
>instance, the greatest phase of the Industrial Revolution began after
>the Civil War ended, waxed during the Edwardian "Gilded Age", then waned
>with the onset of federal socialism in the US in the 20s (which
>culminated in the Great Depression and 90% tax-rates under FDR).
Um.. sport. Federal socialism, especially that under FDR, came about
as a result of trying to fix the Great Depression. The Great
Depression came about because in a capitalistic society you have
periods of boom and periods of bust. Furthermore, your 'theory' that
technological progression is tied to liberty is a load of horseshit.
You need only reference both the Great Depression and its successor
WWII as periods of extreme technological change.
>For several decades thereafter, innovation was comparatively stagnant in the
>US
Stagnant is development of television, jet aircraft, the atomic bomb,
radar, sonar, the microwave oven, the transistor, and computers JUST
TO NAME A FEW? What the hell would you call technological growth?
>, while much more spry, ironically, in the newly-liberated countries (e.g., Japan).
Japan didn't have shit for technological growth. Their burned out
wreck of a country over the next 20 years brought itself into a
manufacturing power and a business power.. but that isn't
technological growth. The tech growth only came later. I didn't see
them invent TV.. but they sure could produce them in later years. They
sure as hell didn't invent the automobile... but over time could
manufacture them really well. Jet aircraft? Nope not them. Atomic
power? Scratch.. they were too busy recovering from it.
-MJ
Michael Johnson
<makes...@charter.net> wrote:
>"Ken from Chicago" <kwicker_era...@ameritech.net> wrote in
>news:_fWdncnzSMY...@comcast.com:
>
>[snip]
>>> If you're talking about "I, Robot"-style robots, the key is
>>> artificial intelligence. Unfortunately, scientists still haven't
>>> figured out how natural intelligence works, much less been able to
>>> design a machine that
>>
>> Intelligence = the ability to recognize patterns.
>
>Pattern recognition is a component of intelligence, in the same sense
>that numbers are a component of mathematics. Of course, no one knows
>how to design a machine that recognizes patterns, so it's a moot point.
You are kidding me right? What drugs are you on? This is old tech.
This is 90's era technology son. Haven't you used an optical character
recognition program to take a fax and make a text file out of it? What
the hell do you think it does? Maybe, recognize patterns?
We have facial recognition software that can look through a million
photographs and match it to that of a suspect. I suppose thats just
magic though, and that its not in any way looking for patterns.
Fingerprints? Voice identification?
Now.. maybe you're trying to make the point that we don't have
computers that can recognize patterns that weren't programmed into it.
Well, surprise, humans can't either. All we can do is use past
knowledge with which to interpret what we currently see.
>> Computers can intrepret data now--but ... they ... do ... so ...
>> really ... slowly ... RELATIVE ... to ... humans.
>
>Computers *cannot* interpret data now.
Of course they can. When a modern OCR program looks at a 2 dimensional
image... it compares what it knows in its memory about each letter of
the alphabet to what is on the page to determine if this splotch on
the page is an 'A' or a 'Z'.
>All they can do is process it. The interpretation is entirely up to the programmer.
OCR programs blow your perfect little world all to heck. The entire
point of them is to interpret data without human interaction. Without
that which you claim is impossible, they could not perform that task
nor be capable of it.
>>> > Those two long lines on the ground with a series of short lines
>>> > between them . . . ah, it's railroad tracks. The round thing moving
>>> > is a ball, not a wheel. The distance between the car and the one in
>>> > front of it is rapidly diminisihing, better slow down for safety's
>>> > sake.
>>>
>>> Why would processing speed enable a machine to make inferences like
>>> that?
>>
>> It would allow the machine to make the inference, to "connect the
>> dots" FAST enough to be able act or react on it.
>
>It would react only if it were previously programmed to do so.
As would we. Think of this... go... hmm.. what's a task... design a
bridge. By your standard you should be able to look at bridges and
knock one out, after all.. you're human.. hear you roar... but you
can't. Instead we send you to school for 4 years where you are
programmed with the knowledge of how to design a stable bridge.. all
of the problems involved.. and how to get over them. Learning in any
capacity is simply the programming of minds to be able to perform new
tasks through the use of visual stimuli followed by experimentation.
Furthermore, we've even developed ways.... called books, to be able to
save the information so you can first read, then try it out yourself,
and subsequently learn the information and reproduce it.
>> A few weeks ago in the news there was a report about a bunch of robot
>> vehicles all failing a challenge by the military to drive across the
>> desert. But wasn't given a lot of focus was that the military wanted
>> the robot vehicles to do so in a specified time limit. So in trying to
>> get the vehicles to move fast, the machines were too slow to react to
>> data about an upcoming obstacles.
>
>They were preprogrammed.
And you as a human can't drive through the desert without having
allready been programmed. You were programmed when you were taught how
to drive, you were programmed to understand what signs mean what..
what portion of the road you could drive on safely, etc. Everything
you know about both driving and the desert was programmed into you at
one time or another. You couldn't even walk on both legs until many
weeks of experimentation determined the movements and positioning that
would keep you balanced.
-MJ
Heck
<kwicker_era...@ameritech.net> wrote:
>
>"Heck" <he...@toodamnmuchspambutreallycomputer.org> wrote in message
>news:69v0g0tf1lk5rvbef...@4ax.com...
>> In an atmosphere of dialog and conversation, Mark Steese
><makes...@charter.net> wrote:
>> >"Ken from Chicago" <kwicker_era...@ameritech.net> wrote in
>> >news:_fWdncnzSMY...@comcast.com:
>> >
>> >[snip]
snp
>:Logic is a pattern itself. Cause and effect is a pattern.
>
>A prisoner chained to a wall is fed three times a day. Neither sight, sound,
>smell, taste or touch is required to "recognize" this pattern. By
>"recognize" a pattern, I mean having the ability to 1) create NEW patterns
>and to 2) remember OLD patterns. So I agree "intelligence" requires
>knowledge, that is, the ability to know, to remember.
Recognition is not creation and is not memory. Neither is knowing memory. The
terms need to be clear to be able to use them. That's an example of what I mean for
"understanding". I share your notion of wisdom and agree intelligence can appear in
unwise people.
How can you determine that the prisoner is chained to a wall and fed three times a
day without some use of your senses? I don't know what you're trying to say with
that.
>I'm not sure how you are defining "understanding" or "wisdom". I would argue
>that "wisdom" is the ability to know HOW to use intelligence or knowledge
>for the "best" (which is subjective), but that one can be intelligent
>without being wise.
>
>> Free will relates to intelligence, but it's not a cause of it nor caused
>by it. Consider a
>> function that associates an instance of free will with its concomitant
>intelligence. The
>> function would have the character that if a given instance of free will
>was not the same as
>> another instance, the associated intelligence for those two instances of
>free will would also
>> not be identical. It would also have the character that there is not an
>associated instance of
>> free will for every instance of intelligence. This reveals some important
>interdependencies
>> between free will and intelligence.
>
>I agree. People can be brainwashed, hypnotized or simply drugged into losing
>(to varying degrees, depending on type of drug and amount used) their free
>will. It doesn't mean they are no longer intelligent.
Nevertheless, lacking certain kinds of intelligence or lacking intelligence to
certain degrees precludes exercise of and even the facility of freedom of choice.
Heck
<mya...@san.rr.com> wrote:
snp
>They don't consistently work with real life humans either.. or we
>wouldn't have people found guilty for crimes they didn't commit
>because an 'eye-witness' used their wonderful image recognition
>systems to identify the defendent as said perpetrator.
Very true. I know of a legal system which requires at least two eye witnesses for
conviction of certain crimes (e.g., capital crimes). Moreover, if the witnesses'
testimony agree in every particular, their testimony is disqualified.
Ken from Chicago
"Recognize" in the colloquial usage can and is used to refer to someone
observing a lot of data and "recognizing" a pattern therein. Like when poker
players notice a player has a "tell", some sort of sign or body gesture that
reveals if they are bluffing or how martial artists can detect an opening in
their opponent's fighting tactics.
> How can you determine that the prisoner is chained to a wall and fed three
times a
> day without some use of your senses? I don't know what you're trying to
say with
> that.
Whoa, I meant the prisoner could discern a pattern. I agree that I, as an
outside ... well ... observer ... would need to use some sort of sensor
scan.
Needing intelligence to have free will is not the same as needing free will
to have intelligence. You can have your cake and eat it too, but the reverse
is a bit tricky.
-- Ken from Chicago
Mike1
>Mike1 <Mitchell-Holman-s...@usfamily.net> wrote in
>> Mark Steese <makes...@charter.net> wrote:
>>
>> Meanwhile, every half-ass straw-hut village in Asia is a forest of
>> akyscrapers when you visit it a second time.
>
>Mm-hm. And how many of them have you visited a first time?
I was in the Philippines in the mid-'90s. There were so many cranes in
Manila that the place looked like an Oklahoma oil-field with derricks
crowding out the sky. Go out in the boondocks and it looks like you're
back in "Apocolype Now"-land, until you star a little more closely and
see the sneakers and the cell-phones (they may not have plumbing yet,
but they have *gadgets*!).
>> America is rushing to emulate European socialist sclerosis.
>
>Thank you for your well-reasoned rejoinder. I'm especially impressed by
>the cites you provided. Personally, I think the socialistically-
>sclerotic America is a good thing -- imagine what the country would look
>like if it were taken over by franchises and megamalls.
The Horror. I could never live with that. Take my *liberty*.....
--
Reply to mike1@@@usfamily.net sans two @@, or your reply won't reach me.
Drug smugglers and gun-runners are heroes of American capitalism.
-- Jeffrey Quick
Van Bagnol
Mark Steese <makes...@charter.net> wrote:
> DaLov...@hotmail.com (Love Rhino) wrote in
> news:84c8d9dc.04072...@posting.google.com:
>
> >> > You can say, "Go to the store and get me a beer."
> >>
> >> And the robot can say "I don't think so."
Actually, according to the 3rd law, a robot might _not_ say that.
> >> Mindlessly obeying commands is not thinking.
That's a strawman statement. _Mindlessly_ obeying commands is not
thinking. _Mindfully_ obeying commands is.
> > Maybe I'm missing something here... I'm not following your argument
> > well, but you seem to be supporting my case. You doubted the utility
> > of intelligent robots vs mindless robots and I'm telling you
> > the practical benefits. The alternative would be having to program
> > the darn thing.
>
> The practical benefits of which? The robot that automatically obeys the
> command "Go to the store and get me a beer" *is* a mindless robot. It's
> not thinking. I never said there was no utility to robots that don't
> think.
But a human that automatically obeys it is neither mindless nor a robot.
The very example you bring up has been discussed in AI research.
(Actually, the scenario was "bake a cake" but the substeps were
similar.) While _obeying_ the order doesn't require much thought,
_executing_ it still requires thinking. A lot of thinking. Consider the
subtasks involved:
(1) There's a fence immediately in front of the sidewalk, but a gate
several meters to the side.
(2) The gate is closed.
(3) The runner has no cash and the bank is in the opposite direction to
the store.
(4) The store is out of beer but there's another store nearby.
The problem is that each subtask requires additional contextual
knowledge and the problem space can get recursively larger.
> > It's a safe bet that it would be easier to teach a dog to fetch
> > your slippers, than it would be to program a mindless robot to.
And actually, subtask (1) _is_ used as a measure of intelligence. If a
fence stands between the dog and the slippers, either the dog (a) gives
up, (b) digs under the fence, or (c) realizes it has to walk _away_ from
the slippers to the gate in order to get _to_ the slippers.
> >> You can't program a machine to be a slave *and* give it the
> >> ability to think.
> >
> > Again, that's an interesting assumption, but I don't know if it is
> > true. I don't think it is.
I agree with Love Rhino. And again, I think Mark Steese is confusing
free will, or more accurately the ability to disobey, with thinking
ability. Since he's appeared to be a gadfly in previous posts I suppose
it's just a predilection to believe people who follow have weaker
thinking skills.
The question is, can one program a machine to think, and if so,
constrain it with boundary conditions? If such constraint is possible,
what would be the nature of implementing the boundaries?
Van
--
Van Bagnol / v b a g n o l at earthlink dot net / c r l at bagnol dot com
...enjoys - Theatre / Windsurfing / Skydiving / Mountain Biking
...feels - "Parang lumalakad ako sa loob ng paniginip"
...thinks - "An Error is Not a Mistake ... Unless You Refuse to Correct It"
Mark Steese
>>> Intelligence = the ability to recognize patterns.
>>
>>Pattern recognition is a component of intelligence, in the same sense
>>that numbers are a component of mathematics. Of course, no one knows
>>how to design a machine that recognizes patterns, so it's a moot
>>point.
>
> You are kidding me right? What drugs are you on? This is old tech.
> This is 90's era technology son. Haven't you used an optical character
> recognition program to take a fax and make a text file out of it? What
> the hell do you think it does? Maybe, recognize patterns?
Just because it has the word "recognition" in its name doesn't mean it's
actually recognizing anything. Recognition has a conscious component not
present in OCR software, or any other currently existing technology. I
doubt if even Ken from Chicago would describe OCR software as intelligent.
In other breaking news, Grape-Nuts contains neither grapes nor nuts.
> We have facial recognition software that can look through a million
> photographs and match it to that of a suspect. I suppose thats just
> magic though, and that its not in any way looking for patterns.
> Fingerprints? Voice identification?
None of these has the conscious component necessary to be accurately
described as recognition. The technology simply makes mechanical
comparisons between the same features on multiple images; it doesn't know
what it's doing.
> Now.. maybe you're trying to make the point that we don't have
> computers that can recognize patterns that weren't programmed into it.
> Well, surprise, humans can't either. All we can do is use past
> knowledge with which to interpret what we currently see.
Computers can't recognize patterns, period. They can identify patterns,
but that's it.
People, on the other hand, can not only recognize patterns, but can also
recognize patterns that lie outside their past knowledge. In fact,
virtually all progress in the sciences has stemmed from that ability --
it's not as though Kepler had past knowledge of elliptical orbits or
Mendeleev already knew the Periodic Table.
>>> Computers can intrepret data now--but ... they ... do ... so ...
>>> really ... slowly ... RELATIVE ... to ... humans.
>>
>>Computers *cannot* interpret data now.
>
> Of course they can. When a modern OCR program looks at a 2 dimensional
> image... it compares what it knows in its memory about each letter of
> the alphabet to what is on the page to determine if this splotch on
> the page is an 'A' or a 'Z'.
That's not interpretation. Again, there is a conscious element not found
in the program, which doesn't know anything. (If you'd had much experience
with OCR software, you wouldn't be using the phrase "what it knows.")
>>All they can do is process it. The interpretation is entirely up to
>>the programmer.
>
> OCR programs blow your perfect little world all to heck. The entire
> point of them is to interpret data without human interaction. Without
> that which you claim is impossible, they could not perform that task
> nor be capable of it.
The software can compare the scanned images to a databank of alphanumeric
characters, but it doesn't know what it's doing, and it doesn't do it very
well, either. Even on printed text, OCR software makes numerous errors no
literate person would make; on handwritten text, it's hopeless. The fact
that OCR software tries to make text out of decorative borders demonstrates
that it isn't interpreting anything; all it's doing is processing.
[snip]
>>It would react only if it were previously programmed to do so.
>
> As would we. Think of this... go... hmm.. what's a task... design a
> bridge. By your standard you should be able to look at bridges and
> knock one out, after all.. you're human.. hear you roar... but you
> can't.
Sure I can. I look at the gap to be spanned, I find a tree growing at the
edge of the gap that's taller than the gap is wide, I chop down the tree,
it falls across the gap, and there you go.
> Instead we send you to school for 4 years where you are programmed
> with the knowledge of how to design a stable bridge.
Where'd that knowledge come from? If people can only work from
"programmed" knowledge, how did anyone ever build a bridge in the first
place? They couldn't have. People invent things they've never seen
before; that's a fairly well-known human characteristic, but I guess you
weren't programmed to learn it.
--
Mark Steese
Unscramble and underscore to email
---
Blaine's next announced escapade will involve dropping himself from a
helicopter at a great height into a river, which seems to symbolize nothing
more than the general public's increasing desire to see David Blaine
dropped from a great height into a river. -fametracker.com
Mark Steese
news:van-0F2BBB.1...@news3.west.earthlink.net:
>> >> > You can say, "Go to the store and get me a beer."
>> >>
>> >> And the robot can say "I don't think so."
>
> Actually, according to the 3rd law, a robot might _not_ say that.
I don't think robots that only exist in Asimov's fiction are going to be
much good at getting people beer.
>> >> Mindlessly obeying commands is not thinking.
>
> That's a strawman statement. _Mindlessly_ obeying commands is not
> thinking. _Mindfully_ obeying commands is.
Where was it specified that the robot was "mindfully" obeying the
command? I didn't see any indication that the beer-retrieving robot had
a choice in the matter.
>> > Maybe I'm missing something here... I'm not following your argument
>> > well, but you seem to be supporting my case. You doubted the
>> > utility of intelligent robots vs mindless robots and I'm telling
>> > you the practical benefits. The alternative would be having to
>> > program the darn thing.
>>
>> The practical benefits of which? The robot that automatically obeys
>> the command "Go to the store and get me a beer" *is* a mindless
>> robot. It's not thinking. I never said there was no utility to
>> robots that don't think.
>
> But a human that automatically obeys it is neither mindless nor a
> robot.
Humans *don't* automatically obey commands. One shudders to think at
the horrors the command "Go fuck yourself" would cause if people
automatically sought to obey it.
> The very example you bring up has been discussed in AI research.
> (Actually, the scenario was "bake a cake" but the substeps were
> similar.) While _obeying_ the order doesn't require much thought,
Not much thought, no. But it *does* require thought.
[snip]
>> >> You can't program a machine to be a slave *and* give it the
>> >> ability to think.
>> >
>> > Again, that's an interesting assumption, but I don't know if it is
>> > true. I don't think it is.
>
> I agree with Love Rhino. And again, I think Mark Steese is confusing
> free will, or more accurately the ability to disobey, with thinking
> ability.
And again, you're wrong.
> Since he's appeared to be a gadfly in previous posts I suppose it's
> just a predilection to believe people who follow have weaker thinking
> skills.
As I have already noted, people who follow have reasons for doing so.
They may be seeking reward; they may be seeking to avoid punishment;
they may be listening to the voices in their heads; but they are not
following because they were programmed to do so. Programming and
thinking are two different things.
But hey, like I said to Love Rhino, all you have to do to prove me wrong
is program a machine to be a slave and give it the ability to think.
What are you waiting for?
--
Mark Steese
Unscramble and underscore to email
---
Blaine's next announced escapade will involve dropping himself from a
helicopter at a great height into a river, which seems to symbolize
nothing more than the general public's increasing desire to see David
Blaine dropped from a great height into a river. -fametracker.com
Mark Steese
news:Mitchell-Holman-special-ed-project-AD9799.04250328072004
@news.usfamily.net:
>>> Meanwhile, every half-ass straw-hut village in Asia is a forest of
>>> akyscrapers when you visit it a second time.
>>
>>Mm-hm. And how many of them have you visited a first time?
>
> I was in the Philippines in the mid-'90s. There were so many cranes in
> Manila that the place looked like an Oklahoma oil-field with derricks
> crowding out the sky. Go out in the boondocks and it looks like you're
> back in "Apocolype Now"-land, until you star a little more closely and
> see the sneakers and the cell-phones (they may not have plumbing yet,
> but they have *gadgets*!).
Right. So you went to Manila, which was a substantial city long before
the mid-90s. Now, how many "half-ass straw-hut villages" have you
visited, and then returned to to find forests of skyscrapers?
>>> America is rushing to emulate European socialist sclerosis.
>>
>>Thank you for your well-reasoned rejoinder. I'm especially impressed
>>by the cites you provided. Personally, I think the socialistically-
>>sclerotic America is a good thing -- imagine what the country would
>>look like if it were taken over by franchises and megamalls.
>
> The Horror. I could never live with that.
So, you're dead, then?
Van Bagnol
> DaLov...@hotmail.com (Love Rhino) wrote in
> news:84c8d9dc.04072...@posting.google.com:
>
> > But, let's explore your statement again:
> >
> >> Mindlessly obeying commands is not thinking.
> >
> > Mindlessly obeying commands is not thinking, but depending on the task
> > requires thinking. Sometimes your boss commands you to do something
> > like extend this application to handle XY situation. You may have
> > mindless obeyed his command, but the task requires a great deal of
> > thinking.
>
> But you *didn't* mindlessly obey his command. You performed the task
> because you were hired to work for your boss. There is an obvious
> benefit to doing what your boss tells you to do: you get paid.
You should be surprised at the number of thinking subordinates who
mindlessly obey commands. You're confusing a _lapse_ of thought with the
_inability_ to think.
> > The military is probably the best example. A general gives the orders
> > to invade a country, his field marshals must think of a plan.
>
> Again, I don't see mindless obeisance there. The field marshals can
> choose whether or not to follow the general's orders. While there are
> examples of officers following orders that were disastrously
> misconceived, I don't know of a single example in which the officer was
> aware of the flaws but obeyed without consciously deciding to do so.
> General Pickett didn't mindlessly obey Lee's order to charge the Union
> lines; he had a very good idea what the result would be, and he obeyed
> only with grave misgivings.
It appears that you've set up "mindlessness" conditionally. I.e., "If a
_robot_ obeys a questionable order, that's mindless; but if a _human_
obeys it, it's mindful but with grave misgivings." Did Pickett choose
not to follow Lee's order? Or did he obey it despite knowing better?
(Besides, I think it was _James Longstreet_ who had misgivings about
Lee's plan to attack Meade at Gettysburg but did not countermand it --
the argument is materially the same, though.) You may have made Love
Rhino's point.
Michael Johnson
<mere_...@charter.net> wrote:
>Michael Johnson <mya...@san.rr.com> wrote in
>news:t1eeg090tpiq5rrgv...@4ax.com:
>
>>>> Intelligence = the ability to recognize patterns.
>>>
>>>Pattern recognition is a component of intelligence, in the same sense
>>>that numbers are a component of mathematics. Of course, no one knows
>>>how to design a machine that recognizes patterns, so it's a moot
>>>point.
>>
>> You are kidding me right? What drugs are you on? This is old tech.
>> This is 90's era technology son. Haven't you used an optical character
>> recognition program to take a fax and make a text file out of it? What
>> the hell do you think it does? Maybe, recognize patterns?
>
>Just because it has the word "recognition" in its name doesn't mean it's
>actually recognizing anything. Recognition has a conscious component not
>present in OCR software, or any other currently existing technology. I
>doubt if even Ken from Chicago would describe OCR software as intelligent.
I didn't say it was intelligent. But recognizing that a letter on the
wall represents an A is simply a program, not necessarily a measure of
intelligence. How does a human recognize such a thing? Because we know
that when you have two oppositely diagonal lines with a bar connecting
them in the middle.. that that means the letter 'A'. When you look at
it, there is a program that kicks off in your mind that was built in
your childhood that very quickly judges where each part of graphic is
is utilizing visual stimuli.. and comes up with an answer as to what
that letter it represents. Why do you think when someone is reading
something they ask you... does this look like an 'I' or a 'J'? Their
program can't reconcile an answer and look for more feedback,
preferably from someone with more of a contextual basis, to solidify
an answer. OCR programs work in exactly the same way.. they look at a
picture and use what they know about the English alphabet and attempt
to reconcile an image that they are scanning. And in the end they are
at best 95% correct as some letters may be too destorted. At that
point the output is given and the user provides feedback to fill in
the rest having more of a contextual basis with which to fill in the
rest.
>In other breaking news, Grape-Nuts contains neither grapes nor nuts.
Point? The only reason you yourself can parse that previous line is
that you have a contextual basis to know what Grape-Nuts, grapes, and
nuts are.. and have been programmed the difference between each. An AI
with the exact same contextual basis in memory could do the same
thing.
>> We have facial recognition software that can look through a million
>> photographs and match it to that of a suspect. I suppose thats just
>> magic though, and that its not in any way looking for patterns.
>> Fingerprints? Voice identification?
>
>None of these has the conscious component necessary to be accurately
>described as recognition. The technology simply makes mechanical
>comparisons between the same features on multiple images; it doesn't know
>what it's doing.
In other breaking news, so do we. We make mechanical comparisons(using
our eyes) between the same features on multiple images and use that as
a basis to determine whether the images match. If we have a sufficient
number of similarities.. our mind decides they do match, and therefore
recognition is acheived. Why do you think Western people, when
interacting with peoples they haven't necessarily grown up with.. can
have a hard time differentiating between Vietnamese, Japanese, Korean,
or Chinese peoples. They certainly don't have that problem.
Now take two fingerprints. A human sits down and tries to decide
whether a fingerprint from a suspect matches one from a scene of a
crime? So how does he do it? By matching similar features on each
print and when he finds enough he declares a match.
>> Now.. maybe you're trying to make the point that we don't have
>> computers that can recognize patterns that weren't programmed into it.
>> Well, surprise, humans can't either. All we can do is use past
>> knowledge with which to interpret what we currently see.
>
>Computers can't recognize patterns, period. They can identify patterns,
>but that's it.
And if you want to define it that way fine. But on your paradigm you
must say that Humans can't recognize patterns, period. They can
identify patterns, but that's it. In a modern computer recognition
program there is little difference between how a computer makes a
match and how a human does it. One was even patterned after the other.
>People, on the other hand, can not only recognize patterns, but can also
>recognize patterns that lie outside their past knowledge.
Um.. no they can't. Instance?
>In fact, virtually all progress in the sciences has stemmed from that ability --
>it's not as though Kepler had past knowledge of elliptical orbits or
>Mendeleev already knew the Periodic Table.
Are you trying to claim that Kepler had no idea of what an ellipse is?
That when he made his measurements with scientific instruments and
charted paths and accumulated data from his tools.. that he wouldn't
recognize the pattern as that of an ellipse?
All Mendeleev did was organize what we allready knew about the
elements into a chart that could easily be read. He didn't simply pull
Hydrogen out of his ass and slap it on the chart and all of a sudden
jee whiz.. Hydrogen exists! Eureka! He used past knowledge that he had
learned and turned it into a new chart for easy usage.
All science is a matter of guesswork followed by tests to try to prove
that guess right or wrong based on what you allready know. The
Scientific Method was modeled directly upon that. Make a guess based
on past knowledge.. set up tests.. try to confirm whether the outcome
matches what you believed and know.
>>>> Computers can intrepret data now--but ... they ... do ... so ...
>>>> really ... slowly ... RELATIVE ... to ... humans.
>>>
>>>Computers *cannot* interpret data now.
>>
>> Of course they can. When a modern OCR program looks at a 2 dimensional
>> image... it compares what it knows in its memory about each letter of
>> the alphabet to what is on the page to determine if this splotch on
>> the page is an 'A' or a 'Z'.
>
>That's not interpretation.
Oh.. yes it is. I suggest you learn more about OCR programs and how
they work sir.
>Again, there is a conscious element not found in the program, which doesn't
>know anything.
The program knows what it is told about what makes up the letter
'A'... the same way you know what you are told about what makes up the
letter 'A'. Who told you? Thank your Kindergarten teacher. Why is it
you think that kids who don't go to school or get taught usually end
up illiterate? Cause they weren't 'taught' how to read. Why can't
they, it isn't that they aren't intelligent, its because they didn't
build the program needed to recognize this splotch on the wall means
'A'.
>(If you'd had much experience with OCR software, you wouldn't be using
>the phrase "what it knows.")
>>>All they can do is process it. The interpretation is entirely up to
>>>the programmer.
>>
>> OCR programs blow your perfect little world all to heck. The entire
>> point of them is to interpret data without human interaction. Without
>> that which you claim is impossible, they could not perform that task
>> nor be capable of it.
>
>The software can compare the scanned images to a databank of alphanumeric
>characters, but it doesn't know what it's doing, and it doesn't do it very
>well, either.
This is old outdated tech from 20 years ago.
>Even on printed text, OCR software makes numerous errors no
>literate person would make; on handwritten text, it's hopeless. The fact
>that OCR software tries to make text out of decorative borders demonstrates
>that it isn't interpreting anything; all it's doing is processing.
That you have 'recognized' the borders as decorative borders only
implies that you have a larger base of knowledge to draw from. And so
long as the software disregards what it sees as matching any letters,
it did its job. And if it does recognize a portion of a border as
having the same characteristics as a letter.. you would have too. In
the same way two people can look at clouds and see some image in them,
even though they are just clouds.
>[snip]
>>>It would react only if it were previously programmed to do so.
>>
>> As would we. Think of this... go... hmm.. what's a task... design a
>> bridge. By your standard you should be able to look at bridges and
>> knock one out, after all.. you're human.. hear you roar... but you
>> can't.
>
>Sure I can. I look at the gap to be spanned, I find a tree growing at the
>edge of the gap that's taller than the gap is wide, I chop down the tree,
>it falls across the gap, and there you go.
And where did you learn that? Cartoons :>.
>> Instead we send you to school for 4 years where you are programmed
>> with the knowledge of how to design a stable bridge.
>
>Where'd that knowledge come from?
Trial and error.
>If people can only work from "programmed" knowledge, how did anyone
>ever build a bridge in the first place? They couldn't have.
Guesses based on past knowledge followed by tests to see if they were
correct. Take a look at Engineering Disasters on the History Channel
sometime :>. Bridges can and do fail all the time. Airplanes do crash.
Each time we use what we know, and pick through the pieces to
determine why something failed, find the answer and then come up with
a solution.. again using past knowledge... to fix it. We then test(QC)
the implementation to determine the problem will not rearise, and then
pass the informtion along to be used in any future designs.
>People invent things they've never seen before;
All through trial and error using information learned from the world
around them.
But please do provide examples. Hell.. i'll toss one out.. how about
the lightbulb? Mr. Edison, took something like 10000 tests to find a
filament that could give off light for an extended period of time.
Shouldn't he have just magically gone.. OH LOOK LIGHTBULB!
-MJ
Heck
<kwicker_era...@ameritech.net> wrote:
>"Recognize" in the colloquial usage can and is used to refer to someone
>observing a lot of data and "recognizing" a pattern therein. Like when poker
>players notice a player has a "tell", some sort of sign or body gesture that
>reveals if they are bluffing or how martial artists can detect an opening in
>their opponent's fighting tactics.
ideas contained in the concept, you see the presence of knowledge, understanding and
wisdom. Pattern also has a lot of ides packed into it, in an organized form, too.
snp
>> How can you determine ... without some use of your senses?
>
>Whoa, I meant the prisoner could discern a pattern. I agree that I, as an
>outside ... well ... observer ... would need to use some sort of sensor
>scan.
The prisoner, too, needs sensory input to tell. We base our knowledge in our
sensory input. That's where knowledge starts for us.
snp
>> Nevertheless, lacking certain kinds of intelligence or lacking intelligence to
>> certain degrees precludes exercise of and even the facility of freedom of
>choice.
>
>Needing intelligence to have free will is not the same as needing free will
>to have intelligence. You can have your cake and eat it too, but the reverse
>is a bit tricky.
The exercise of free will enhances intelligence because it positively impacts
knowledge, understadning and wisdom.
Mark Steese
@news3.west.earthlink.net:
>> > Mindlessly obeying commands is not thinking, but depending on the
>> > task requires thinking. Sometimes your boss commands you to do
>> > something like extend this application to handle XY situation. You
>> > may have mindless obeyed his command, but the task requires a great
>> > deal of thinking.
>>
>> But you *didn't* mindlessly obey his command. You performed the task
>> because you were hired to work for your boss. There is an obvious
>> benefit to doing what your boss tells you to do: you get paid.
>
> You should be surprised at the number of thinking subordinates who
> mindlessly obey commands.
I would be, if there were any evidence that *any* thinking subordinate
ever mindlessly obeyed a command. It is true that most commands do not
require much thought on the subordinate's part, but they require *some*
thought.
> You're confusing a _lapse_ of thought with the _inability_ to think.
You're postulating a lapse of thought without evidence. The brain
functions quickly; it takes only a few milliseconds or so for a
subordinate to go from hearing a command to deciding to obey it -- but
there's still a decision being made.
>> > The military is probably the best example. A general gives the
>> > orders to invade a country, his field marshals must think of a
>> > plan.
>>
>> Again, I don't see mindless obeisance there. The field marshals can
>> choose whether or not to follow the general's orders. While there
>> are examples of officers following orders that were disastrously
>> misconceived, I don't know of a single example in which the officer
>> was aware of the flaws but obeyed without consciously deciding to do
>> so. General Pickett didn't mindlessly obey Lee's order to charge the
>> Union lines; he had a very good idea what the result would be, and he
>> obeyed only with grave misgivings.
>
> It appears that you've set up "mindlessness" conditionally. I.e., "If
> a _robot_ obeys a questionable order, that's mindless; but if a
> _human_ obeys it, it's mindful but with grave misgivings."
And it appears that you've set up "mindlessness" unconditionally, i.e.,
"If anyone obeys an order, that's mindless."
> Did Pickett choose not to follow Lee's order? Or did he obey it
> despite knowing better?
If he knew better, then he didn't make the decision mindlessly, did he?
> (Besides, I think it was _James Longstreet_ who had misgivings about
> Lee's plan to attack Meade at Gettysburg but did not countermand it --
> the argument is materially the same, though.)
Both Longstreet *and* Pickett had misgivings, but Pickett was the one
responsible for leading the actual charge. (You may have noticed that
it's not generally referred to as "Longstreet's Charge.") Pickett never
forgave Lee, either -- as he said to Mosby, "That old man had my
division massacred at Gettysburg."
> You may have made Love Rhino's point.
Only if Love Rhino's point was that anyone who follows any order under
*any* circumstances is necessarily mindless, which is what you seem to
be saying above. You think Pickett knew better; I think the robot
wouldn't be capable of knowing better. Why would anyone design a robot
that *was* capable of knowing better?
Mark Steese
>>>>> Intelligence = the ability to recognize patterns.
>>>>
>>>>Pattern recognition is a component of intelligence, in the same
>>>>sense that numbers are a component of mathematics. Of course, no
>>>>one knows how to design a machine that recognizes patterns, so it's
>>>>a moot point.
>>>
>>> You are kidding me right? What drugs are you on? This is old tech.
>>> This is 90's era technology son. Haven't you used an optical
>>> character recognition program to take a fax and make a text file out
>>> of it? What the hell do you think it does? Maybe, recognize
>>> patterns?
>>
>>Just because it has the word "recognition" in its name doesn't mean
>>it's actually recognizing anything. Recognition has a conscious
>>component not present in OCR software, or any other currently existing
>>technology. I doubt if even Ken from Chicago would describe OCR
>>software as intelligent.
>
> I didn't say it was intelligent.
Then your argument isn't with me, but rather with Ken from Chicago, who
defined intelligence as the ability to recognize patterns. You seem to
feel differently.
[snip]
>>> Now.. maybe you're trying to make the point that we don't have
>>> computers that can recognize patterns that weren't programmed into
>>> it. Well, surprise, humans can't either. All we can do is use past
>>> knowledge with which to interpret what we currently see.
>>
>>Computers can't recognize patterns, period. They can identify
>>patterns, but that's it.
>
> And if you want to define it that way fine.
It's not a question of how I want to define it; that's the way the words
are generally used.
> But on your paradigm youmust say that Humans can't recognize patterns,
> period.
Nonsense. You don't have the slightest idea what my "paradigm" is.
> They can identify patterns, but that's it. In a modern computer
> recognition program there is little difference between how a computer
> makes a match and how a human does it. One was even patterned after
> the other.
Again, nonsense. Why can't OCRs read handwriting?
>>People, on the other hand, can not only recognize patterns, but can
>>also recognize patterns that lie outside their past knowledge.
>
> Um.. no they can't. Instance?
Darwin's Theory of Evolution by Natural Selection.
>>In fact, virtually all progress in the sciences has stemmed from that
>>ability -- it's not as though Kepler had past knowledge of elliptical
>>orbits or Mendeleev already knew the Periodic Table.
>
> Are you trying to claim that Kepler had no idea of what an ellipse is?
> That when he made his measurements with scientific instruments and
> charted paths and accumulated data from his tools.. that he wouldn't
> recognize the pattern as that of an ellipse?
By your standards, no, he wouldn't have, because the prevailing belief
was that planetary orbits were circular, not elliptical.
[snip]
>>>>Computers *cannot* interpret data now.
>>>
>>> Of course they can. When a modern OCR program looks at a 2
>>> dimensional image... it compares what it knows in its memory about
>>> each letter of the alphabet to what is on the page to determine if
>>> this splotch on the page is an 'A' or a 'Z'.
>>
>>That's not interpretation.
>
> Oh.. yes it is. I suggest you learn more about OCR programs and how
> they work sir.
I know how they work. You don't know what the word interpretation
means.
>>Again, there is a conscious element not found in the program, which
>>doesn't know anything.
>
> The program knows what it is told about what makes up the letter
> 'A'...
No, it doesn't. It knows *nothing.*
[snip]
>>>>It would react only if it were previously programmed to do so.
>>>
>>> As would we. Think of this... go... hmm.. what's a task... design a
>>> bridge. By your standard you should be able to look at bridges and
>>> knock one out, after all.. you're human.. hear you roar... but you
>>> can't.
>>
>>Sure I can. I look at the gap to be spanned, I find a tree growing at
>>the edge of the gap that's taller than the gap is wide, I chop down
>>the tree, it falls across the gap, and there you go.
>
> And where did you learn that? Cartoons :>.
I've never seen that portrayed in a cartoon. I made it up.
>>> Instead we send you to school for 4 years where you are programmed
>>> with the knowledge of how to design a stable bridge.
>>
>>Where'd that knowledge come from?
>
> Trial and error.
But how did anyone know to try anything in the first place? In your
world, it never would have occurred to anyone that bridges *could* be
built.
>>If people can only work from "programmed" knowledge, how did anyone
>>ever build a bridge in the first place? They couldn't have.
>
> Guesses based on past knowledge
Ah. "Past knowledge." Trouble is, you go back far enough and there is
no past knowledge. Somebody had to invent the original concepts;
somebody had to be the first to recognize the patterns. But in your
world, that's not possible. In your "paradigm," alphanumeric characters
couldn't have been invented; I guess you think they predate our species.
Lewis Mammel
Mark Steese wrote:
>
> Michael Johnson <mya...@san.rr.com> wrote in
> news:oeoig0pv40ocmldgo...@4ax.com:
> > Are you trying to claim that Kepler had no idea of what an ellipse is?
> > That when he made his measurements with scientific instruments and
> > charted paths and accumulated data from his tools.. that he wouldn't
> > recognize the pattern as that of an ellipse?
>
> By your standards, no, he wouldn't have, because the prevailing belief
> was that planetary orbits were circular, not elliptical.
The orbits of mars and the earth are very nearly circular in shape.
If you draw them to scale on piece of paper you can't tell they
aren't circles by looking at them. Also, Kepler had no way of
placing the planets on a particular orbit by direct observation,
since the observed positions are defined by a single line of sight.
The way he was able to choose an elliptical orbit over an
eccentric circular orbit was by applying the law of motion
by equal areas. The difference in areas swept out along the
different shaped orbits accumulates, and predicts slightly
different positions ALONG the orbits. Using a circular orbit
Kepler was not able to account for an observational discrepancy
of seven minutes of arc, and moved on to try ellipses with
better success, although I think there is still some question
whether he was working with enough precision to definitively
make the distinction.
Lew Mammel, Jr.
Matthew Montchalin
|"The Intrinsically Flawed Mr. Hole" <holef...@webtv.net> wrote in
|message news:11701-40F...@storefull-3272.bay.webtv.net...
|> Yes, "Spider-Man 2" was a good film, but "I, Robot" is the first and
|> only film this summer to give me goosebumbs, multiple times! With a big
|> action film that's one of the most important aspects. Bravo Alex
|> Proyas, you continue on your journey towards director Godhood. During
|> the climactic battle scene of "I, Robot" I literally had my jaw gaping
|> open. So a big "BOO!" to anyone that's come down on this film.
|
|
|It's too bad it doesn't really have a fucking thing to do with the
|book it was based on.
|
|They should have used Harlan Ellison's script.
|
The most unacceptable part was that consumers would really buy Robots
that couldn't work reliably for more than 24 hours without sudden
downtime, and leeching a new, perhaps slightly less unreliable OS from
the central mainframe 'VIKI.'
James Gassaway
news:Pine.LNX.4.44.040906...@onyx.spiritone.com...
Use Windows?
--
Multiversal Mercenaries. You name it, we kill it. Any time, any reality.
Matthew Montchalin
|> |It's too bad it doesn't really have a fucking thing to do with the
|> |book it was based on.
|> |
|> |They should have used Harlan Ellison's script.
|> |
|> |What a waste.
|>
|> The most unacceptable part was that consumers would really buy Robots
|> that couldn't work reliably for more than 24 hours without sudden
|> downtime, and leeching a new, perhaps slightly less unreliable OS
|> from the central mainframe 'VIKI.'
|
|Use Windows?
Talk about goosebumps!
David Johnston
>|They should have used Harlan Ellison's script.
I've read Ellison's script. It was more faithful. It wasn't good for
movie-making purposes. If expanded it would have made a good
mini-series.