Well, not to sound shallow and all, but to tell the truth, I'm working on something resembling this bottom-up approach, ie., and autonomical pattern-morphing system (which would somehow magically infuse order into chaos:)... tho I doubt I will come up with something fast.
Pipedreaming I may be, but at last it's semi-constructive pipe-dreaming:p
Well, yeah, I agree. One of the main reasons why people are foiled is minimal level of interaction in many public forums (ie., people only talk about given topic) or general purpose phrases directed at noone at say, IRC. When you get close and personal, you can pretty much catch the impostor everytime; the problem is, very little communication over the net is actually close and personal...
Tho I doubt (heh. again:) that prisoner's dilemma solvage would be a good benchmark... well, because, erm, what exactly defines "human-iness" in this test? It carries too little useful output imo -- and I mean, "creative" type of output -- which, again imo, would be most useful in judging who is supposedly "intelligent".
OTOH, I'm the supporter of decision-making approach to testing intelligence, ie., if program attains truly unpredictable level of intelligent decision making, it could be considered intelligent.
Bah. Nevermind. Too many words with no clear definition, and I'm a babbling idiot (bot?).
And now the best part -- yes my friends, it's true -- I'm attempting to create what could pass as AI someday... tho I'm only doing that at algorithm level (infomation hierarchies, patterns and morphing thereof, what is information, that sorta thing). In this field, theory is so far far more important than practical implementations.
You said, "AI is a fraud", but you meant "AI community is a bunch of crooks".
I'd wager that AI research is possible, provided said "AI crooks" change their attitudes (highly not likely), or a community using a whole new approach to AI emerges.
On a related note, as someone mentioned, hardware is a likely obstacle, and without 30 years down the road, we just cannot begin to try to simulate even such "trivial" thing as a human brain.
AI of to-day reminds me of physics of Newton. Cute, yet primitive.
> Now the hard part: "How do you design it with motives that will work in a world with lots of people, and will also allow the people to continue to exist?"
Well, there could be two major approaches here imo: the "simple but hardcoded one", aka Asimov's laws of robotics;
and complex/pliable one, just like a human being that can have good or bad purpose, mostly depending on training of said complex being.
Of these two, as elegant as Asimov-style laws look, I would be in favor of more common-sense pliable-animal scenario.
I would consider humans that think animals not intelligent in the exactly same sense that humans are, deluded.
Humans themselves are not really intelligent... this is a very fuzzy matter indeed.
Furthermore, I must say that huamns should pursue not some shady "intelligence", but really some more solid concept, like self-perception, or awareness... For when we create artificial awareness, it will be "aware" enough to learn and evolve by itself, and only computational resources will then be a limit.
Some will say that the only intelligent ones are the enlightened, those who perceive themselves as an indistinguishable part of universe, and are aware of unity of it all. Think about it. Do you really know who you are?:)
Well. I dunno if you're joking here, but when humanity creates a program to distinguish between other program and (intelligent, erudite, superior) human being, the humanity will have created near-perfect AI.
Geez, run-o-the-mill humans cannot distinguish between primitive (state of the art) bots / primitive (run of the mill) humans even today...
Only erudites of the highest order shold be allowed to participate in turing tests as arbiters, in my opinion.
And all this intelligence-testing idea is a very shaky ground we're treading on anyways, because there will, in the passage of time, be more and more state-of-the-art bots who will outperform run-o-the-mill humans in every way.
First of all, thank you for your effort, whether it may succeed or not... for it's not the point - what's important is that people like you (and maybe me-in-the-future) try out new things and grow and evolve etc.
Without error there is no improvement.
That said, from your post I'm able to deduct that your UI prototype is potentially Neat. I'd be willing to test it out and help with it, because I'm confident in my ability to think about things-as-systems, and generally have a good taste in UIs. Simple, yet flexible and elegant, and aesthetically pleasing, not seizure-inducing.
Regarding your comments on whether growing pains would negate potential benefit, I highly doubt it -- consider UIs of games. Sometimes they're cluttered and ugly, sometimes complex yet elegant -- but observe how many people are willing to learn entirely different paradigms, even multiple times per year! Just throw in enough motivation (fun) and they will come, nay, flock!:)
Slashdot Mirror
> Also, there's still a couple people waiting on Duke Nukem Forever, and their patience is wearing thin.
And so did God unleash the rabid game fans upon the world, and they laid waste, and no stone upon stone was left where once great warehouses stood.
Go on... With or without credit, I don't really care. :)
Heh
If some new scheduler would use fancy statistical methods to learn about your machine's usage patterns, and adapt its predictions to it.
Of course, and I highly suspect it, I may be talking out of my ass.
According to Pokey the Penguin
*bzzzt* wrong!
It's id
*whistles innocently*
> Two problems:
:)
> 1) How do you define a human?
> 2) How to you keep it from redefining it's goals?
These questions, I will better leave unmessed with for a while
Thanks for the info.
Well, not to sound shallow and all, but to tell the truth, I'm working on something resembling this bottom-up approach, ie., and autonomical pattern-morphing system (which would somehow magically infuse order into chaos :)... tho I doubt I will come up with something fast.
:p
Pipedreaming I may be, but at last it's semi-constructive pipe-dreaming
Well, yeah, I agree. One of the main reasons why people are foiled is minimal level of interaction in many public forums (ie., people only talk about given topic) or general purpose phrases directed at noone at say, IRC. When you get close and personal, you can pretty much catch the impostor everytime; the problem is, very little communication over the net is actually close and personal...
Your sig attained self awareness five minutes ago, right before my eyes.
*goes off to fetch some more hash candies*
Haha.
:) that prisoner's dilemma solvage would be a good benchmark... well, because, erm, what exactly defines "human-iness" in this test? It carries too little useful output imo -- and I mean, "creative" type of output -- which, again imo, would be most useful in judging who is supposedly "intelligent".
Tho I doubt (heh. again
OTOH, I'm the supporter of decision-making approach to testing intelligence, ie., if program attains truly unpredictable level of intelligent decision making, it could be considered intelligent.
Bah. Nevermind. Too many words with no clear definition, and I'm a babbling idiot (bot?).
And now the best part -- yes my friends, it's true -- I'm attempting to create what could pass as AI someday... tho I'm only doing that at algorithm level (infomation hierarchies, patterns and morphing thereof, what is information, that sorta thing). In this field, theory is so far far more important than practical implementations.
You said, "AI is a fraud", but you meant "AI community is a bunch of crooks".
I'd wager that AI research is possible, provided said "AI crooks" change their attitudes (highly not likely), or a community using a whole new approach to AI emerges.
On a related note, as someone mentioned, hardware is a likely obstacle, and without 30 years down the road, we just cannot begin to try to simulate even such "trivial" thing as a human brain.
AI of to-day reminds me of physics of Newton. Cute, yet primitive.
...for a program that says "yes master, I will abide the rules" is more a sign of a dullard than of free will at work.
> Now the hard part: "How do you design it with motives that will work in a world with lots of people, and will also allow the people to continue to exist?"
Well, there could be two major approaches here imo: the "simple but hardcoded one", aka Asimov's laws of robotics;
and complex/pliable one, just like a human being that can have good or bad purpose, mostly depending on training of said complex being.
Of these two, as elegant as Asimov-style laws look, I would be in favor of more common-sense pliable-animal scenario.
I would consider humans that think animals not intelligent in the exactly same sense that humans are, deluded.
:)
Humans themselves are not really intelligent... this is a very fuzzy matter indeed.
Furthermore, I must say that huamns should pursue not some shady "intelligence", but really some more solid concept, like self-perception, or awareness... For when we create artificial awareness, it will be "aware" enough to learn and evolve by itself, and only computational resources will then be a limit.
Some will say that the only intelligent ones are the enlightened, those who perceive themselves as an indistinguishable part of universe, and are aware of unity of it all. Think about it. Do you really know who you are?
Well. I dunno if you're joking here, but when humanity creates a program to distinguish between other program and (intelligent, erudite, superior) human being, the humanity will have created near-perfect AI.
Geez, run-o-the-mill humans cannot distinguish between primitive (state of the art) bots / primitive (run of the mill) humans even today...
Only erudites of the highest order shold be allowed to participate in turing tests as arbiters, in my opinion.
And all this intelligence-testing idea is a very shaky ground we're treading on anyways, because there will, in the passage of time, be more and more state-of-the-art bots who will outperform run-o-the-mill humans in every way.
I have pretty much similar problem with nethack - it's neither textual, nor graphical.
I like MUDs and graphical RPGs (and obviously pen and paper ones), but nethack is just not it.
Gimme art, or let me visualise it from raw text.
My reply was meant to be a joke :)
... but noone moderated it as funny.
*Which* last 84 digits you mean, my friend?
Reply here using the secret ninnik speak, and we will rule the world together!
First of all, thank you for your effort, whether it may succeed or not... for it's not the point - what's important is that people like you (and maybe me-in-the-future) try out new things and grow and evolve etc.
:)
Without error there is no improvement.
That said, from your post I'm able to deduct that your UI prototype is potentially Neat. I'd be willing to test it out and help with it, because I'm confident in my ability to think about things-as-systems, and generally have a good taste in UIs. Simple, yet flexible and elegant, and aesthetically pleasing, not seizure-inducing.
Regarding your comments on whether growing pains would negate potential benefit, I highly doubt it -- consider UIs of games. Sometimes they're cluttered and ugly, sometimes complex yet elegant -- but observe how many people are willing to learn entirely different paradigms, even multiple times per year! Just throw in enough motivation (fun) and they will come, nay, flock!
Good luck