The computer can only process stuff via one or more cpus. The human mind has no such limitation. The best equivalent would be a computer where every byte is associated with a cpu, for MASSIVE parallelism. And those cpus would have to be able to re-wire their conections over time, based on the data in them and their surrounding cpus.
3 points:
1. Computers have been built that behave like this, usually as part of research systems.
2. Any sufficiently powerful computer architecture can simulate any other. The only real limitations are on storage capacity and processing speed. This means that we don't really need to have computers that do what you describe; we just need computers that are fast enough and have enough storage that the can simulate it at an adequate speed.
3. Intelligence may or may not require a specific architecture to function; this has not yet been demonstrated either way. What you describe may be incidental to the way the human brain has evolved, rather than a requirement of the function it performs.
Quantum phenomena are a fact of life. We can't get away from that. They are also the only satisfactory explanation to date of what is really going on.
Why? Give me a single good reason that consciousness could not be an emergent property of the complexity of the human brain, why it has to be a quantum effect?
On a related topic, I'm not satisfied with the Turing test being the standard. I want to see emotion, I want to see humour, I want to see sarcasm, I want to see love, I want to see empathy.
I don't believe an AI could pass the Turing test without at least the ability to bluff these things -- which means, as far as I'm concerned, understanding them, at least, if not having them.
I'm not satisfied with the Turing test because it makes the implicit assumption that intelligence can only exist by modelling or imitating the way humans think, which strikes me as highly arrogant.
No, actually neanderthals on average had larger brainpans than their successors.
The correlation between brain size and intelligence isn't always direct. Structure is important, too.
Remember, they (neanderthals) invented/developed the 7 things we take for granted: agriculture, animal husbandry, fire, tools, religion, communications, art.
How on Earth can we possibly know that? Particularly if they coexisted with cromagnon man, at which point radio-isotope datings of any cave paintings we might find would not indicate which of them was responsible.
Oh, please, you pick up a small error that Kurzweil has made (which has little influence over his theories, it is a minor point in that respect), and then go on to state as fact:
Hell, no computer can even have an opinion. And that's probably not going to change even with nanotech, because the consciousness seems to "inhabit" the quantum world
There is _no_ evidence to back this up. And I don't consider Penrose's half-assed psuedo-scientific arguments evidence, but the rants of someone too convinced of his own superiority to allow him to consider that the way his own mind does things might not be the only way to achieve similar results.
They're also often willing to consider shorter than their target guidelines.
And you're right that it varies by genre. I understand that 50,000 words is actually fairly average for either romance or young-adult novels. For reference, 50,000 words would probably be slightly under 200 pages when printed in standard paperback format.
I think they originated as the official guidelines of the Science Fiction Writers of America society. They need such guidelines because their membership requirements involve having published certain categories of fiction and having been paid for it at a particular rate, which depends on what type of work it is.
I know of several published novelists who do this every year. When you're into the swing of it, writing at this kind of rate is (apparently) not all that difficult, even if you are aiming for quality.
They tend to make sure they do all the research they require in advance, and have a pretty good idea of what they're going to do with the story, which avoids the major problems. Also note that the point isn't actually to write a novel: it's to write a first draft of the first 50,000 words of a novel. It does not need to be complete within the month, and it certainly does not need to be edited.
"It is perfectly okay to write garbage--as long as you edit brilliantly." C.J. Cherryh
I don't know about you, but I tend to average about 5-600 words per hour when I sit down to write (as opposed to when I sit down to procrastinate while pretending to write). That means that the time commitment to finishing this is approximately 3 hours per day. If you can find that amount of time for it, I think it's an excellent idea. This November, I'll be trying for 25,000 words -- I simply don't have that much time.
Also note: most of the people doing this would probably never write their book without some kind of stunt to motivate them. Sure, most of the results are junk that is never properly edited or submitted to publishers other than PublishAmerica et al, but is this really a problem?
yeah, universities would kind of expect that. Give you homework like take an OS kernel and add multiprocessing to it, give you a weekend.
Hmmm...? Not quite sure what you're getting at here. That's clearly a ridiculous assignment, though.
Do you think languages have got more complex (JAva, C#, C++), with all the libraries you need to learn with them, or the problems have got harder?
You don't need to know all of a language's libraries in order to know the language. I, myself, claim to know Java and C++ (I started learning C# yesterday, I don't claim to know it yet). In terms of the standard libraries, though, I would say I probably know about 70% of C++ and about 30% of Java. Yet this is enough for me to perform day-to-day programming tasks using whichever language I need.
There must be somebody here who can do a reasonable imitation of the style of artwork they've got on that front page. I want to see the conclusion to that storyboard. Preferably involving a long, heavy wooden pole labelled "cluestick".
This just screams "give me valid email addresses so I can sell them to spammers!"
But it says "Yelp! will not share email addresses with third parties." You don't think they'd do that do you? After they said they wouldn't? That's just... immoral.
And _that_ reminds me of one of the weirdest quirks of Rexx: it has DYNAMIC SCOPE!! The variables you have access to from within a procedure depend on where that procedure was called from. How's THAT for wacky
Ahh. I wrote a little interpreter a while back (I called it MOOSE) that did that. It was an interesting experiment, but I never put in the time to make it efficient enough for the application I had in mind.
Because I had a Computer Science degree, I was able to master a simple procedural programming language on my own, with one flimsy, poorly written internal reference document, within a week [...] That, my friends, is the power of a good University degree.
No it isn't. Back in '92 or thereabouts, I was studying away to get the qualifications I would need to get onto a CS course. I learned to program in C, the first 'proper' language I had ever learned. (Until then, it was mostly BASIC, with a little assembly language)
I switched schools a few months later, and was informed that because the tutor at the new school didn't know C, I would need to do my project in Pascal.
So, I learned Pascal. Yes, it took me about a week. Remember this is a long time before I got my degree.
I think what this says is more about the kind of people who get degrees than what you learn on the degree course itself.
Ah. I think it's because we're slashdotting it. I was trying to investigate what was going on, so I looked at it through lynx, which gave the right page... then I tried to grab the source with lynx -source, and got the home page. Trying again got the right page; I think the server's dropping a cached copy of the home page on people whenever it's too busy to serve up individual requests for other pages.
Slashdot Mirror
That's not how they'd do it. They'd buy out somebody else already in the game -- that's how MS enters new markets.
The computer can only process stuff via one or more cpus. The human mind has no such limitation. The best equivalent would be a computer where every byte is associated with a cpu, for MASSIVE parallelism. And those cpus would have to be able to re-wire their conections over time, based on the data in them and their surrounding cpus.
3 points:
1. Computers have been built that behave like this, usually as part of research systems.
2. Any sufficiently powerful computer architecture can simulate any other. The only real limitations are on storage capacity and processing speed. This means that we don't really need to have computers that do what you describe; we just need computers that are fast enough and have enough storage that the can simulate it at an adequate speed.
3. Intelligence may or may not require a specific architecture to function; this has not yet been demonstrated either way. What you describe may be incidental to the way the human brain has evolved, rather than a requirement of the function it performs.
Quantum phenomena are a fact of life. We can't get away from that. They are also the only satisfactory explanation to date of what is really going on.
Why? Give me a single good reason that consciousness could not be an emergent property of the complexity of the human brain, why it has to be a quantum effect?
On a related topic, I'm not satisfied with the Turing test being the standard. I want to see emotion, I want to see humour, I want to see sarcasm, I want to see love, I want to see empathy.
I don't believe an AI could pass the Turing test without at least the ability to bluff these things -- which means, as far as I'm concerned, understanding them, at least, if not having them.
I'm not satisfied with the Turing test because it makes the implicit assumption that intelligence can only exist by modelling or imitating the way humans think, which strikes me as highly arrogant.
No, actually neanderthals on average had larger brainpans than their successors.
The correlation between brain size and intelligence isn't always direct. Structure is important, too.
Remember, they (neanderthals) invented/developed the 7 things we take for granted: agriculture, animal husbandry, fire, tools, religion, communications, art.
How on Earth can we possibly know that? Particularly if they coexisted with cromagnon man, at which point radio-isotope datings of any cave paintings we might find would not indicate which of them was responsible.
Oh, please, you pick up a small error that Kurzweil has made (which has little influence over his theories, it is a minor point in that respect), and then go on to state as fact:
Hell, no computer can even have an opinion. And that's probably not going to change even with nanotech, because the consciousness seems to "inhabit" the quantum world
There is _no_ evidence to back this up. And I don't consider Penrose's half-assed psuedo-scientific arguments evidence, but the rants of someone too convinced of his own superiority to allow him to consider that the way his own mind does things might not be the only way to achieve similar results.
They're also often willing to consider shorter than their target guidelines.
And you're right that it varies by genre. I understand that 50,000 words is actually fairly average for either romance or young-adult novels. For reference, 50,000 words would probably be slightly under 200 pages when printed in standard paperback format.
I think they originated as the official guidelines of the Science Fiction Writers of America society. They need such guidelines because their membership requirements involve having published certain categories of fiction and having been paid for it at a particular rate, which depends on what type of work it is.
I know of several published novelists who do this every year. When you're into the swing of it, writing at this kind of rate is (apparently) not all that difficult, even if you are aiming for quality.
They tend to make sure they do all the research they require in advance, and have a pretty good idea of what they're going to do with the story, which avoids the major problems. Also note that the point isn't actually to write a novel: it's to write a first draft of the first 50,000 words of a novel. It does not need to be complete within the month, and it certainly does not need to be edited.
"It is perfectly okay to write garbage--as long as you edit brilliantly."
C.J. Cherryh
I don't know about you, but I tend to average about 5-600 words per hour when I sit down to write (as opposed to when I sit down to procrastinate while pretending to write). That means that the time commitment to finishing this is approximately 3 hours per day. If you can find that amount of time for it, I think it's an excellent idea. This November, I'll be trying for 25,000 words -- I simply don't have that much time.
Also note: most of the people doing this would probably never write their book without some kind of stunt to motivate them. Sure, most of the results are junk that is never properly edited or submitted to publishers other than PublishAmerica et al, but is this really a problem?
No, I think it's calling people nazis and "dipshits" that is considered a troll.
There's a class of gun that has never been used in a crime? I'm impressed. Have any been sold to the public yet?
"1" is not [...] a way to indicate a probability of something happening.
Yes it is.
I think it was supposed to be a joke.
yeah, universities would kind of expect that. Give you homework like take an OS kernel and add multiprocessing to it, give you a weekend.
Hmmm...? Not quite sure what you're getting at here. That's clearly a ridiculous assignment, though.
Do you think languages have got more complex (JAva, C#, C++), with all the libraries you need to learn with them, or the problems have got harder?
You don't need to know all of a language's libraries in order to know the language. I, myself, claim to know Java and C++ (I started learning C# yesterday, I don't claim to know it yet). In terms of the standard libraries, though, I would say I probably know about 70% of C++ and about 30% of Java. Yet this is enough for me to perform day-to-day programming tasks using whichever language I need.
I'll admit I'm only repeating & summarising other people's commentary on the named case. Have I missed something?
Lupe is a dog
I would have thought, based on the name, that it is a wolf.
Yeuch! I don't think I'm ever going to eat fast-food mayonnaise again.
Findum, Fuckum, and Flee
That sounds like a good name for a firm of lawyers.
There must be somebody here who can do a reasonable imitation of the style of artwork they've got on that front page. I want to see the conclusion to that storyboard. Preferably involving a long, heavy wooden pole labelled "cluestick".
This just screams "give me valid email addresses so I can sell them to spammers!"
But it says "Yelp! will not share email addresses with third parties." You don't think they'd do that do you? After they said they wouldn't? That's just... immoral.
Ford can keep their red paint. It fades like nothing I've ever seen before.
I think this patent covers an older implementation using javascript that the plugin replaced.
But the slashdot thieves will argue that they have some sorta 'right' to steal it.
I suspect they will. It's called the MPAA v Betamax decision; it states you have a legal right to make recordings for the purpose of 'time shifting'.
And _that_ reminds me of one of the weirdest quirks of Rexx: it has DYNAMIC SCOPE!! The variables you have access to from within a procedure depend on where that procedure was called from. How's THAT for wacky
Ahh. I wrote a little interpreter a while back (I called it MOOSE) that did that. It was an interesting experiment, but I never put in the time to make it efficient enough for the application I had in mind.
developing IVR and "database" solutions using REXX and Btrieve, on OS/2! It was awful.
It can't have been worse than the work I did a while back with Microsoft QuickBASIC and BTrieve on DOS with a NetWare server...
Because I had a Computer Science degree, I was able to master a simple procedural programming language on my own, with one flimsy, poorly written internal reference document, within a week
[...]
That, my friends, is the power of a good University degree.
No it isn't. Back in '92 or thereabouts, I was studying away to get the qualifications I would need to get onto a CS course. I learned to program in C, the first 'proper' language I had ever learned. (Until then, it was mostly BASIC, with a little assembly language)
I switched schools a few months later, and was informed that because the tutor at the new school didn't know C, I would need to do my project in Pascal.
So, I learned Pascal. Yes, it took me about a week. Remember this is a long time before I got my degree.
I think what this says is more about the kind of people who get degrees than what you learn on the degree course itself.
Ah. I think it's because we're slashdotting it. I was trying to investigate what was going on, so I looked at it through lynx, which gave the right page... then I tried to grab the source with lynx -source, and got the home page. Trying again got the right page; I think the server's dropping a cached copy of the home page on people whenever it's too busy to serve up individual requests for other pages.