Actually, I'd say that not very many are doing that.
The goal of all the cognitive scientists I've met is to make machines think, just as with A.I. In fact, I've always heard, and was told in my AI class, that A.I. is a branch of cognitive science.
However, there are many approaches to machine thinking that are not considered part of A.I.: neural networks, SVMs, computer vision (signal interpretation), modeling.
So what does A.I. cover then? Well, it's not exactly well defined. If you read A.I. textbooks, you'll find the full of lots of different things. Some would go so far as to even include those things I mentioned that aren't normally considered part of A.I. However, in general, I would say that A.I. is the field that is concerned with 1) Solving the search problem (searching for a solution in a large set of possibilities) 2) Doing it with heuristics.
I'd like to take a moment to note that a famous computer vision paper came out in the 80's that documented a method called Marr-Hildreth, which was for finding edges in images. They created it by using the same technique that eyes use (laplacian of a Gaussian for edge detection - they studied cats to find this out).
A few years later someone improved upon it by throwing out the model completely and NOT doing it the way that people do (Canny).
Cognitive scientists are usually more concerned with getting the machines to do what we want than they are with modeling human thinking techniques.
I don't believe that. I've met too many people that speak multiple languages.
Have you heard about the study that relate how bilingual speakers have completely different EEGs depending on what language they're speaking?
Its almost as if they're thought processes are changed.
I know that I myself think differently depending on what programming language I'm using, and those aren't nearly as different as, say, english is from chinese.
Evolve or die? Have you heard of Chomsky? Language itself is finite! There are only so many things you need to describe all conditions!
Most DSL is ADSL, which stands for Asynchronous Digital Subscriber Line. That means that the connection is not a fixed speed. In other words, it's NOT dedicated bandwidth.
At the main hubs, both cable and dsl have to use some other kind of connection to the internet - T1, T3, etc. to connect to the rest of it, and both of them can have bandwidth problems.
The difference that can make you notice is that cable connections can be split much more easily than DSL connections. This means that companies are much more likely to give service when they've already reached capacity, since it's not much more trouble for them.
As far as capping...you must have access to some cable company that I don't know about. I've looked at the main broadband rating website (I think it's here), and it seems that a lot of companies cap the cable modems themselves, and a few do the subnet routers.
In terms of raw bandwidth, it is dsl, not cable, that is a dead-end. Cable services are slowly switching customers over to digital cable services. When this move is complete, all the wasted bandwidth being absorbed by an inefficient analog broadcasting protocol will be available for trasmission (not like they've even maxed out that part now).
In other words, when it is cost effective to offer speeds that are 100x what they are now, DSL won't be able to cut it without adding more wires, but cable will.
It's complicated because language is complicated. As always, the goals of lawmakers is to make the spirit of the law match the letter of it. Obviously, there have been times when we have failed (the "separation of church and state" concept was brought into law and has caused religious persecution despite the fact that the purpose was to stop religious persecution). Interesting that the bill of rights is rather short to the point and uncomplicated, isn't it?
Making language meet an arbitrary level of precision - the same precision as the spirit of the law - is difficult. That is why it is necessary for the system to be complicated.
I think a better, less complicated approach to law would be to require all lawyers and people who wanted to use the law to learn and speak a limited subset of language that has absolute precision (for example, there would have to not be any words that mean "very" "much" or "too").
The law has gotten so complicated that having another language that everyone had to learn would actually simplify it. George Orwell got it right with newspeak - not that we should have it, but that limiting language limits how you think - and certianly law requires a particular pattern of thinking of it's own, which, if enforced in this manner, would naturally limit the complexity of laws.
The law would certainly be against the DMCA then, since all programmers would readily be able to become lawyers.:)
I'm not sure that's true. I know you can copywrite lyrics.
Plus, keep in mind that he does Polka medleys which include the lyrics to songs EXACTLY as they are. I can see how that might not be protected because it isn't parody at all. It's the same songs, but played faster and with kazoos and an accordian.
If you look at the comics from 30 years ago, you'll see complete stories in one comic - or possibly two at the most. Also, characters where recognizable even if they, say, changed clothes.
Today, both of these things have changed. To get a complete story, you've got to buy 10 or 15 comics. And this isn't only because the stories are longer and more complicated - todays comics seem to have more advertising instead of storyline. This is also prompting what I call "soap opera syndrome" - the comics are becoming WAY to melodramatic to try keep you interested between stories so that you'll buy the next one. Also, a lot of the drawing is done so hastily (note that I'm not saying that the artists are bad), that the only way you can recognize the characters is because they have the same clothing from one frame to the next. Their facial features/size/build keeps changing.
Two things would make me buy comic books again (note that I wasn't alive 30 years ago, but I used to buy old comic books because they where good). 1) Enough with the cliffhangers! They suck! If you need more story, make a longer comic and charge more! 2) Draw well!
No one'severwritten a System Administration guide that you should pay for! Now we can stop blowing our money on books with new information in them and buy ones like this!
You can blame your sysadmin for missing gcc, less, and top. Solaris does give the option of including those tools rather easily. Of course, you won't get a modern version of any of them.
As far as a better shell, yeah, it would be nice if they added the bash-completion package to Solaris since the default package is Bash, but tcsh is available on most Solaris systems and it does that stuff.
As long as you don't want your system to be nifty (what I would call elegant) in any way.
Forget having modern extensions and X11R6 applications. You can't have them.
And color terminals? Add them yourself! And forget about user support! You'll pay them good money if you want support!
What about advancing the gnu tools to the current level? You want recursive grepping? Color "ls"? Tar support for bzip2 and gzip? These are only the common ones that I've noticed are subpar compared to linux - I'm sure there are many others that I don't use. Wait until the next version of Solaris and maybe it'll get added.
The hardware will be great, though - for only ten times what you pay for commodity hardware you get reliability (just ignore the fact that if you buy quality hardware for PCs that cost about twice that amount you'll get the same level of quality).
I've yet to see that Solaris is elegant. It works, but it sure ain't pretty - not even compared to Linux. What they offer is reliability that comes from good hardware.
And don't give me any stuff about not rebooting for 4 years - you can do the same with flavors of Linux designed for that. There's more to it than software stability now.
I guess Linux is just going to be getting better and better!
I hear that Microsoft will be helping out to improve the open source nature of Linux, and Sun will be working hard to make sure it works perfectly on x86 systems!
Surely good days lie ahead of us with this collection of hardworking, selfless, and competent companies backing us up!
"Arguing against the transhuman argument is pretty pointless because it's probably going to happen sooner or later."
Don't you think you're missing something? Like an argument? Let me simplify: let's call the eventual occurrance of transhumans, "the statement." Then what you said, making the substition (and other easier to see word changes) "Claiming that the statement is false is pointless because the statement is true."
Think about this before you think of how advanced our minds will be: the first science was philosophy. The first scientists/mathematicians of the renessaince where philosophers as well (Decarte and Pascal come to mind). We have been working on that problem of advancing the mind of the human race for at least 4000 years, and doing so through every means at our disposal.
We've had an inkling that a computer could exist for about 100, and couldn't really make anything viable until about 40 years ago. And even then, we couldn't do anything on a broad scale until the invention of the transistor.
We soar in one field while we haven't even begun to crawl in the other. I'll let you figure out which is which. Transhumans in 100 years? Maybe 100 millenia.
Quantum...my guess is that they're referring to the quanta of an embedded system - which would be states.
I seem to recall that state machines were the first thing that I learned when I started learning about computer architecture. They are, after all, the basis. What I find disturbing is that programmers that you meet in the workplace have a hard time understanding state machines and they're still in the workplace.
Isn't that a like knowing how for loops work? Or how to do recursion? It's a basic fundamental concept. This is even more true for embedded systems, where many (machine based) optimization techniques involve using a strict Meely or Moore implementation of a state machine (I'm thinking right now about how much better it works if you program using one of these models when writing in VHDL for an FPGA).
Now my ruthless plan to buy the land around Los Angeles, and blow up the fault line so that Los Angeles has a massive earthquake and falls into the ocean can finally succeed!
Then all the rich saps living in Los Angeles will have nowhere to go except another city...MY city - Lutherville. It won't be long before I own their sorry butts! And no one can stop me!
Here you go.I actually like evolution better than I ever liked Outlook.
On that note, let me say specifically, that I want my Office suite to
NOT
include anything for e-mail. First of all, an e-mail client is not a publishing tool. Its used to converse.
Secondly, I don't want something made for publishing with a built in api to require access to the internet for any purpose. That leads to the insecurity, which leads to the dark side - viruses.
Right now, nobody can make money except the US government because it's very difficult.
But if it was electronic, then there would be times when security was breached, and massive amounts could be created. This would depreciate the value of money and throw our economy into a depression.
Even if no one breaks in, it would be a lot easier for the government itself to create money, which could easily create the same problem. Sure, they might not all do much, but if thousands of government agencies are all doing it a little...it'll add up.
Having electronic money is a bit like have electronic war. I think I'd rather have them know what I'm doing.
I have had the experience you mention debugging Windows machines. In fact, my machine used to break for no reason and I wasn't sure why. I know I have a crappy motherboard, bad video capture card and video card and often failing harddrive, and at the time a bad sector in my ram, but I couldn't be sure why things where locking up.
I have since then moved my computer over to Linux. In MY experience as a tech, the problems are reported by the system quite frequently. However, you have to be working under an operating system that's polite enough to tell you about it (via logs, mostly).
If you have intermittent failure of most of the components, the rest of the system should still tell you due to the modularity.
Case in point: I had some RAM with a bad sector somewhere. Occasionally, my computer would use that sector for something critical and my machine would crash. But it always gave the appropriate error message, so I knew why it was crashing.
When something fails in computing, it does fail outright. It might not fail the next time, but a failure is a failure. If the hardware has an error on the microscopic level which doesn't lead to a failure, then it's not a problem.
The main exception to this, I would say, is the CPU. It can spit out bad data to no end, and you won't have any clue why. The motherboard is not much better. Once again, though, those are the two components that I mentioned could be the problem.
I see computers the same way I see programs: other than the processor, pretty much all of it is modular.
As long as this continues to be the case, we won't have serious scaling problems (this is where the programs come in - it is also true for when writing programs). When some complicated component breaks, whatever controls it will tell us. If that breaks, whatever controls IT will tell us.
The list of things that can break without notifying the system can still be kept small - the motherboard itself, and the processor (right now if the memory goes bad you can often get the same problem, though I don't think there's any excuse for that).
Of course, the black box (in terms of seeing what's wrong) - the CPU - will get more difficult to check using test vectors as complexity goes up. The process of making them will get more expensive and time consuming. But that doesn't affect me much. I'll still just be throwing away my CPU if it stops working and buying another one.
those internet appliances didn't do that well. I don't know anyone who owns an e-mail appliance.
When you can buy a commodity computer for $200 or a specialized box for $150, and the computer can do way more (even if you don't need it to), you're going to go with the $200 machine.
At least, that's what people have been doing. It's sort of like putting a stereo into a car even though you don't go there just to listen to the radio. It's a nice touch, and it doesn't add much to the overall price of the car.
More and more, that particular prediction of Ellison's has been proven wrong. Many companies have tried and failed to produce such technologies. Most got burnt by it, but a few got lucky. What do you think java was for, originally?
I use 10% of the applications 90% of the time. One of these 10% just happens to work a lot better with optimizations - that being Mozilla/Galeon/Pheonix (all about the same improvement).
Prelinking has also done wonders for me when using the massive applications. This is a binary feature, but it's one that is implemented quite well in Gentoo thanks to the fact that the developers are optimization nuts.
Optimization makes a big difference on really big, really slow applications, though not much on smaller ones. Not surprisingly, these are the apps that are talked about the most frequently in Gentoo chat channels and the forums with regard to optimization.
Did you see the parent of the parent post? It said that a really good movie was "They Live," and then a bunch of people agreed. That's why I "quoted" They Live.
Of course the quote was from They Live. If you look at my post, I SWITCH to talking about army of darkness because "They Live" really only has that one quote. All the rest of the good quotes are in "Army of Darkness" (where "good" means "in a Duke Nukem' game").
Slashdot Mirror
Actually, I'd say that not very many are doing that.
The goal of all the cognitive scientists I've met is to make machines think, just as with A.I. In fact, I've always heard, and was told in my AI class, that A.I. is a branch of cognitive science.
However, there are many approaches to machine thinking that are not considered part of A.I.:
neural networks, SVMs, computer vision (signal interpretation), modeling.
So what does A.I. cover then? Well, it's not exactly well defined. If you read A.I. textbooks, you'll find the full of lots of different things. Some would go so far as to even include those things I mentioned that aren't normally considered part of A.I. However, in general, I would say that A.I. is the field that is concerned with
1) Solving the search problem (searching for a solution in a large set of possibilities)
2) Doing it with heuristics.
I'd like to take a moment to note that a famous computer vision paper came out in the 80's that documented a method called Marr-Hildreth, which was for finding edges in images. They created it by using the same technique that eyes use (laplacian of a Gaussian for edge detection - they studied cats to find this out).
A few years later someone improved upon it by throwing out the model completely and NOT doing it the way that people do (Canny).
Cognitive scientists are usually more concerned with getting the machines to do what we want than they are with modeling human thinking techniques.
A canvas with stuff on it is called a painting.
An html page that's been filled out is called...an html page.
Since we use the same word for the medium as we do for the finished work, HTML is art if the finished work is art.
Debunked? It has? By whom?
I don't believe that. I've met too many people that speak multiple languages.
Have you heard about the study that relate how bilingual speakers have completely different EEGs depending on what language they're speaking?
Its almost as if they're thought processes are changed.
I know that I myself think differently depending on what programming language I'm using, and those aren't nearly as different as, say, english is from chinese.
Evolve or die? Have you heard of Chomsky? Language itself is finite! There are only so many things you need to describe all conditions!
Almost everything you said there is untrue.
Most DSL is ADSL, which stands for Asynchronous Digital Subscriber Line. That means that the connection is not a fixed speed. In other words, it's NOT dedicated bandwidth.
At the main hubs, both cable and dsl have to use some other kind of connection to the internet - T1, T3, etc. to connect to the rest of it, and both of them can have bandwidth problems.
The difference that can make you notice is that cable connections can be split much more easily than DSL connections. This means that companies are much more likely to give service when they've already reached capacity, since it's not much more trouble for them.
As far as capping...you must have access to some cable company that I don't know about. I've looked at the main broadband rating website (I think it's here), and it seems that a lot of companies cap the cable modems themselves, and a few do the subnet routers.
In terms of raw bandwidth, it is dsl, not cable, that is a dead-end. Cable services are slowly switching customers over to digital cable services. When this move is complete, all the wasted bandwidth being absorbed by an inefficient analog broadcasting protocol will be available for trasmission (not like they've even maxed out that part now).
In other words, when it is cost effective to offer speeds that are 100x what they are now, DSL won't be able to cut it without adding more wires, but cable will.
It's complicated because language is complicated. As always, the goals of lawmakers is to make the spirit of the law match the letter of it. Obviously, there have been times when we have failed (the "separation of church and state" concept was brought into law and has caused religious persecution despite the fact that the purpose was to stop religious persecution). Interesting that the bill of rights is rather short to the point and uncomplicated, isn't it?
:)
Making language meet an arbitrary level of precision - the same precision as the spirit of the law - is difficult. That is why it is necessary for the system to be complicated.
I think a better, less complicated approach to law would be to require all lawyers and people who wanted to use the law to learn and speak a limited subset of language that has absolute precision (for example, there would have to not be any words that mean "very" "much" or "too").
The law has gotten so complicated that having another language that everyone had to learn would actually simplify it. George Orwell got it right with newspeak - not that we should have it, but that limiting language limits how you think - and certianly law requires a particular pattern of thinking of it's own, which, if enforced in this manner, would naturally limit the complexity of laws.
The law would certainly be against the DMCA then, since all programmers would readily be able to become lawyers.
I'm not sure that's true. I know you can copywrite lyrics.
Plus, keep in mind that he does Polka medleys which include the lyrics to songs EXACTLY as they are. I can see how that might not be protected because it isn't parody at all. It's the same songs, but played faster and with kazoos and an accordian.
They have to stop sucking.
If you look at the comics from 30 years ago, you'll see complete stories in one comic - or possibly two at the most. Also, characters where recognizable even if they, say, changed clothes.
Today, both of these things have changed. To get a complete story, you've got to buy 10 or 15 comics. And this isn't only because the stories are longer and more complicated - todays comics seem to have more advertising instead of storyline. This is also prompting what I call "soap opera syndrome" - the comics are becoming WAY to melodramatic to try keep you interested between stories so that you'll buy the next one.
Also, a lot of the drawing is done so hastily (note that I'm not saying that the artists are bad), that the only way you can recognize the characters is because they have the same clothing from one frame to the next. Their facial features/size/build keeps changing.
Two things would make me buy comic books again (note that I wasn't alive 30 years ago, but I used to buy old comic books because they where good).
1) Enough with the cliffhangers! They suck! If you need more story, make a longer comic and charge more!
2) Draw well!
Tracking other things?
No one's ever written a System Administration guide that you should pay for! Now we can stop blowing our money on books with new information in them and buy ones like this!
You can blame your sysadmin for missing gcc, less, and top. Solaris does give the option of including those tools rather easily. Of course, you won't get a modern version of any of them.
As far as a better shell, yeah, it would be nice if they added the bash-completion package to Solaris since the default package is Bash, but tcsh is available on most Solaris systems and it does that stuff.
Probably this one is big. It's not size, it's the amount of bad publicity that the military gets. :)
As long as you don't want your system to be nifty (what I would call elegant) in any way.
Forget having modern extensions and X11R6 applications. You can't have them.
And color terminals? Add them yourself! And forget about user support! You'll pay them good money if you want support!
What about advancing the gnu tools to the current level? You want recursive grepping? Color "ls"? Tar support for bzip2 and gzip? These are only the common ones that I've noticed are subpar compared to linux - I'm sure there are many others that I don't use. Wait until the next version of Solaris and maybe it'll get added.
The hardware will be great, though - for only ten times what you pay for commodity hardware you get reliability (just ignore the fact that if you buy quality hardware for PCs that cost about twice that amount you'll get the same level of quality).
I've yet to see that Solaris is elegant. It works, but it sure ain't pretty - not even compared to Linux. What they offer is reliability that comes from good hardware.
And don't give me any stuff about not rebooting for 4 years - you can do the same with flavors of Linux designed for that. There's more to it than software stability now.
I guess Linux is just going to be getting better and better!
I hear that Microsoft will be helping out to improve the open source nature of Linux, and Sun will be working hard to make sure it works perfectly on x86 systems!
Surely good days lie ahead of us with this collection of hardworking, selfless, and competent companies backing us up!
"Arguing against the transhuman argument is pretty pointless because it's probably going to happen sooner or later."
Don't you think you're missing something? Like an argument? Let me simplify: let's call the eventual occurrance of transhumans, "the statement." Then what you said, making the substition (and other easier to see word changes)
"Claiming that the statement is false is pointless because the statement is true."
Think about this before you think of how advanced our minds will be: the first science was philosophy. The first scientists/mathematicians of the renessaince where philosophers as well (Decarte and Pascal come to mind). We have been working on that problem of advancing the mind of the human race for at least 4000 years, and doing so through every means at our disposal.
We've had an inkling that a computer could exist for about 100, and couldn't really make anything viable until about 40 years ago. And even then, we couldn't do anything on a broad scale until the invention of the transistor.
We soar in one field while we haven't even begun to crawl in the other. I'll let you figure out which is which. Transhumans in 100 years? Maybe 100 millenia.
Text adventures online. 'Nuff said.
Quantum...my guess is that they're referring to the quanta of an embedded system - which would be states.
I seem to recall that state machines were the first thing that I learned when I started learning about computer architecture. They are, after all, the basis. What I find disturbing is that programmers that you meet in the workplace have a hard time understanding state machines and they're still in the workplace.
Isn't that a like knowing how for loops work? Or how to do recursion? It's a basic fundamental concept. This is even more true for embedded systems, where many (machine based) optimization techniques involve using a strict Meely or Moore implementation of a state machine (I'm thinking right now about how much better it works if you program using one of these models when writing in VHDL for an FPGA).
Now my ruthless plan to buy the land around Los Angeles, and blow up the fault line so that Los Angeles has a massive earthquake and falls into the ocean can finally succeed!
Then all the rich saps living in Los Angeles will have nowhere to go except another city...MY city - Lutherville. It won't be long before I own their sorry butts! And no one can stop me!
On that note, let me say specifically, that I want my Office suite to
- NOT
include anything for e-mail. First of all, an e-mail client is not a publishing tool. Its used to converse.Secondly, I don't want something made for publishing with a built in api to require access to the internet for any purpose. That leads to the insecurity, which leads to the dark side - viruses.
Better to keep the two separate.
Right now, nobody can make money except the US government because it's very difficult.
But if it was electronic, then there would be times when security was breached, and massive amounts could be created. This would depreciate the value of money and throw our economy into a depression.
Even if no one breaks in, it would be a lot easier for the government itself to create money, which could easily create the same problem. Sure, they might not all do much, but if thousands of government agencies are all doing it a little...it'll add up.
Having electronic money is a bit like have electronic war. I think I'd rather have them know what I'm doing.
I have had the experience you mention debugging Windows machines. In fact, my machine used to break for no reason and I wasn't sure why. I know I have a crappy motherboard, bad video capture card and video card and often failing harddrive, and at the time a bad sector in my ram, but I couldn't be sure why things where locking up.
I have since then moved my computer over to Linux. In MY experience as a tech, the problems are reported by the system quite frequently. However, you have to be working under an operating system that's polite enough to tell you about it (via logs, mostly).
If you have intermittent failure of most of the components, the rest of the system should still tell you due to the modularity.
Case in point: I had some RAM with a bad sector somewhere. Occasionally, my computer would use that sector for something critical and my machine would crash. But it always gave the appropriate error message, so I knew why it was crashing.
When something fails in computing, it does fail outright. It might not fail the next time, but a failure is a failure. If the hardware has an error on the microscopic level which doesn't lead to a failure, then it's not a problem.
The main exception to this, I would say, is the CPU. It can spit out bad data to no end, and you won't have any clue why. The motherboard is not much better. Once again, though, those are the two components that I mentioned could be the problem.
I see computers the same way I see programs: other than the processor, pretty much all of it is modular.
As long as this continues to be the case, we won't have serious scaling problems (this is where the programs come in - it is also true for when writing programs). When some complicated component breaks, whatever controls it will tell us. If that breaks, whatever controls IT will tell us.
The list of things that can break without notifying the system can still be kept small - the motherboard itself, and the processor (right now if the memory goes bad you can often get the same problem, though I don't think there's any excuse for that).
Of course, the black box (in terms of seeing what's wrong) - the CPU - will get more difficult to check using test vectors as complexity goes up. The process of making them will get more expensive and time consuming. But that doesn't affect me much. I'll still just be throwing away my CPU if it stops working and buying another one.
those internet appliances didn't do that well. I don't know anyone who owns an e-mail appliance.
When you can buy a commodity computer for $200 or a specialized box for $150, and the computer can do way more (even if you don't need it to), you're going to go with the $200 machine.
At least, that's what people have been doing. It's sort of like putting a stereo into a car even though you don't go there just to listen to the radio. It's a nice touch, and it doesn't add much to the overall price of the car.
More and more, that particular prediction of Ellison's has been proven wrong. Many companies have tried and failed to produce such technologies. Most got burnt by it, but a few got lucky. What do you think java was for, originally?
I use 10% of the applications 90% of the time. One of these 10% just happens to work a lot better with optimizations - that being Mozilla/Galeon/Pheonix (all about the same improvement).
Prelinking has also done wonders for me when using the massive applications. This is a binary feature, but it's one that is implemented quite well in Gentoo thanks to the fact that the developers are optimization nuts.
Optimization makes a big difference on really big, really slow applications, though not much on smaller ones. Not surprisingly, these are the apps that are talked about the most frequently in Gentoo chat channels and the forums with regard to optimization.
Did you see the parent of the parent post? It said that a really good movie was "They Live," and then a bunch of people agreed. That's why I "quoted" They Live.
Of course the quote was from They Live. If you look at my post, I SWITCH to talking about army of darkness because "They Live" really only has that one quote. All the rest of the good quotes are in "Army of Darkness" (where "good" means "in a Duke Nukem' game").