Well maybe the realized that it's hard
on
The Baby Bootstrap?
·
· Score: 3, Interesting
Bootstrapped learning something useful, even from an information ocean like the internet, is *HARD*.
Doubly so if you have no goals, and your task is just to "learn". It would come back with garbage.
Perhaps the real killer is that even if it did learn something, the information acquired in its unguided search through the internet would be completely alien. You'd then have to launch a second project to figure out what the hell your little guy learned.
And you'd probably figure it out was mostly garbage.
Well if you're right, than there's alot of excellent scientific work that's been done by you and I look forward to learning about it.
But I do have to question your definition of intentionality.
Just because a user emits gesture X, and intended it, doesn't mean it was intended for use in controlling the ipod (or whatever). So in two cases, the same gesture is emitted, but understanding context is required in order to determine if it's suitable.
Or maybe you've found a nice set of gestures which are never intentionally performed for other reasons.... but I would have to question how natural they are.
So somewhere, definitions are awry. If your gestures are truly "natural", than different systems use them intentionally, even in the same context. And figuring out which is which is an "H" hard problem.
Thanks for replying. Sorry about the "insane" crack, I'm on 3 hours of sleep.
If you've effectively solved the problem of determing the user's intent when they make gesture X, as opposed to gesture Y, while allowing people to make natural gestures (as opposed to tortuous twitches).... you should be doing better things with your genius than making new Ipod controllers.
You have apparently cracked problems of cognitive psychology that have been stumping legions of scientists for decades.
If these getures are truly "natural", then they are going to be controlled by different mental processes at different times. Sometimes the part of the brain that wants to skip to the next song, sometimes the part that wants to nod in agreement to the person being talked to (or maybe someone called their name from across the room).
Discriminating what part of the brain intended to make a given gesture is going to require that you effectively read the subject's mind. At the very least, you'd need to follow the ongoing dialogue in the vicinity to determine if the gesture is part of a conversation or not.... So there's a host of "hard" problems that have to be solved to accurately determine intent.
So if you've done all these things, I'm going to retire from my cog neuroscience career, because you're obviously some kind of crazy protege who's surely going to render a century of research by 100,000+ researchers obsolete within the next few years.
I haven't seriously looked for a non academic job ever, is monster.com still how it's done?
Also... you'd like a freak, waving your hand around, nodding your head this way and that as you walk down the street. People would step out of their way to avoid you.
Noone is going to want to walk around doing that.
This sounds like people looking too hard for a "new idea" that can win them bucks. Except that it sucks, and it's not even new (the Media lab's been doing this for many years)
Admittedly, I didn't read the article, but I don't think I have to to know this is a bad idea because I've heard it all before.
How the hell is linking tiny gestural movements to PDA/mobile control going to improve control for people "on the go"? Whether, walking or driving, if I'm actively moving from one place to another, my body needs to be involved in the process. If I have to control head motion while driving or walking to control a PDA, my awareness of my surroundings will be worse.
This whole idea of controlling devices with formerly incidental motions is like a recurring bad dream. Wake up. This idea is bad. It will not work. While we can consciously control these motions with severe training, the default state is that the brain does these things automatically. Any communication medium that forces the user to laboriously reprogram their own brain so that formerly automated behaviours have to reside under exclusive conscious control are impractical.
The most successful user input devices (ie. cars, telephones, pencils, keyboards) have always focussed on elements of interaction that are under direct control in the context of the use of that device (ie I don't control my feet while walking, but I do while sitting in a car, because the walking program isn't engaged)
So? You have limited emissions to a very few sources, instead of having to worry about tens of thousands of catalytic converters and pollution control systems. It is a lot easier to deal with one or very few sources.
Yes it's very beneficial for the environment. The government can send all the fines for breaking EPA guidelines to just a few addresses rather than splitting them up among millions of drivers.
As processor speed, memory and disk space continue to spiral to ever larger values, Microsoft is really going to be put to task in finding ways to make Windows sluggish.
Their task is made more difficult by advances in compiler design which find an eliminate trivial solutions that simply chew up CPU time by computing huging cosine tables and then overwriting them.
New innovation may come from recent advances in polling network devices unnecessarily and hanging various threads until a reply is received. In the case of pulling a device off the network that Windows Explorer had browsed in the last 15 weeks, a given thread can hang for minutes, chewing up processor time in loops that scan network traffic.
The Windows Development team seems optimistic that they can produce the same crippled user interface on new 64 bit architectures that customers have become familiar with, a valuable marketing strategy in teaching consumers to become suspicious of computers with more responsive interfaces.
It's obvious they've become increasingly frustrated by the overwhelming atmosphere of stupidity that has descended over our politics and media lately.
They wanted to lash out at the source of their frustration, but in a way that didn't imperil their status as a reputable (well that's debatable) publication. So they choose the one day of the year when people can go nuts and say what they really mean, and then throw up their hands and say April Fools!
Our society is like a toned-down of Japan in this way, we have a built in release valve for venting our frustration at being bound by certain rules and regulations most of the time. *fwooot*
On the contrary, some jokes get funnier and funnier with retelling.
The problem is that converting one-off jokes into running gags is more art than science, and clearly something these poor guys have absolutely no grasp of.
This RFC was just pitiful, all the more so because of the obvious time investment.
No it's not better. Microsoft still benefits when someone pirates XP, because they get a monkey on their back. And once they get used to XP's features, they'll be reluctant to change, even if that forces them to start paying for it because of new hoops they have to jump through.
It also means they are more likely to use IE and various other MS products.
You are being unrealistic because you assume it takes them the same amount of time to learn as you.
My parents have been working at this hard, for months now. But even the simplest aspects of using a computer have escaped them until recently. These people, god bless them, don't understand that a document, a batch file and a program are the same type of thing on the hard drive. It's just a question of the contents and whether the computer tries to run it or not.
Ditto with so many little things that you and I take for granted.
So what, for you, seems to be a trivial amount of courtesy time, is for them the equivalent of learning to speak a foreign language in terms of time investment. And it's not even as much fun.
My parents have been computer users since 1984. We had an Apple II+ (followed by a IIGS ugh), and then they some variety of PC's at work.
They are very organized and successful people, and they used those early computers with no problem. Word processing, spreadsheets, it was enormously useful and rarely gave us problems. As the years have gone by, the problems they have had with computers have increased exponentially.
This same story is being repeated in every computer owning household in the world.
But if it makes you feel better to stand on your soapbox spewing gibberish about stupid everyone is and if they just follow your 6 simple steps (each of which has 10 implicit substeps that would require hours of training to get through) then everything would be rosy... well keep doing it I guess.
The means to properly operate a computer is as far beyond the reach of the average person as is the ability to tune their car and replace the fan belt. It's not that their stupid,it's that the concepts are completely alien to them. What seems trivial to you is a goddamned nightmare to most adults who grew up without touching a computer.
And there's also a selection bias, many people who are otherwise smart, just aren't good at dealing with computer-type systems. Those people have avoided computers until they couldn't any longer. Again, not that they're stupid (well not all of them), but everyone is good at some things, and bad at others.
Computers have gotten much harder to use successfully as they have gotten more powerful. It is not that the people who put together that 8086 did it "properly", it's that there were far fewer ways to screw it up.
Yes it's true. My own site, which has only recently been offered to consumers at great rates offers VIAGRA and CIALIS tabs for next to nothing. These pills can enhance your performance while you are on one of the FREE VACATIONS that I also offer.
not "could be"...it IS incredibly difficult to build a good walking bot. The best robot labs in the world are still making small, primitive insects that crawl around at a snail's pace. The physics of constructing them are far outside of the grasp of hobbyists.
Eliminating armor is the easier way to do it. What you are proposing raises the bar of design to such a level that you'd see designers spending the bulk of their time and money on the drivetrain and suspension which is not what we are paying to see.
We want simple driving with lots of traction, giving us the potential for lots of damage and kinetic energy, not dune buggies crashing into each other.
And walking bots? You're dreaming. There's a million reasons the military doesn't use them.
You are limited to about 2 square feet of metal sheeting at the biggest weight class, and only 1 mm thick aluminum.
Then tighten up the weight restrictions a bit, making it harder to rig up makeshift armor out of non-sheet shaped metal
This would shift the emphasis towards robots that can deal damage, not take it. You have limited weight and armor, so you use that weight to do as much damage as possible.
This means exposed guts.
Exposed guts means fun times for spikey/smashy robots, and even those flame throwers might hope to melt some wiring.
Think of it like evolution, we shift the environment to make the current optimal solutions impossible, so now everyone has to find a totally new way to go.
Re:Was this really illegel?
on
Book 'Em, Dano
·
· Score: 1
That poor library, they'll be doing a complete reinventory.
Slashdot Mirror
Bootstrapped learning something useful, even from an information ocean like the internet, is *HARD*.
Doubly so if you have no goals, and your task is just to "learn". It would come back with garbage.
Perhaps the real killer is that even if it did learn something, the information acquired in its unguided search through the internet would be completely alien. You'd then have to launch a second project to figure out what the hell your little guy learned.
And you'd probably figure it out was mostly garbage.
Well if you're right, than there's alot of excellent scientific work that's been done by you and I look forward to learning about it.
But I do have to question your definition of intentionality.
Just because a user emits gesture X, and intended it, doesn't mean it was intended for use in controlling the ipod (or whatever). So in two cases, the same gesture is emitted, but understanding context is required in order to determine if it's suitable.
Or maybe you've found a nice set of gestures which are never intentionally performed for other reasons.... but I would have to question how natural they are.
So somewhere, definitions are awry. If your gestures are truly "natural", than different systems use them intentionally, even in the same context. And figuring out which is which is an "H" hard problem.
Thanks for replying. Sorry about the "insane" crack, I'm on 3 hours of sleep.
If you've effectively solved the problem of determing the user's intent when they make gesture X, as opposed to gesture Y, while allowing people to make natural gestures (as opposed to tortuous twitches).... you should be doing better things with your genius than making new Ipod controllers.
You have apparently cracked problems of cognitive psychology that have been stumping legions of scientists for decades.
If these getures are truly "natural", then they are going to be controlled by different mental processes at different times. Sometimes the part of the brain that wants to skip to the next song, sometimes the part that wants to nod in agreement to the person being talked to (or maybe someone called their name from across the room).
Discriminating what part of the brain intended to make a given gesture is going to require that you effectively read the subject's mind. At the very least, you'd need to follow the ongoing dialogue in the vicinity to determine if the gesture is part of a conversation or not.... So there's a host of "hard" problems that have to be solved to accurately determine intent.
So if you've done all these things, I'm going to retire from my cog neuroscience career, because you're obviously some kind of crazy protege who's surely going to render a century of research by 100,000+ researchers obsolete within the next few years.
I haven't seriously looked for a non academic job ever, is monster.com still how it's done?
Also... you'd like a freak, waving your hand around, nodding your head this way and that as you walk down the street. People would step out of their way to avoid you.
Noone is going to want to walk around doing that.
This sounds like people looking too hard for a "new idea" that can win them bucks. Except that it sucks, and it's not even new (the Media lab's been doing this for many years)
Admittedly, I didn't read the article, but I don't think I have to to know this is a bad idea because I've heard it all before.
How the hell is linking tiny gestural movements to PDA/mobile control going to improve control for people "on the go"? Whether, walking or driving, if I'm actively moving from one place to another, my body needs to be involved in the process. If I have to control head motion while driving or walking to control a PDA, my awareness of my surroundings will be worse.
This whole idea of controlling devices with formerly incidental motions is like a recurring bad dream. Wake up. This idea is bad. It will not work. While we can consciously control these motions with severe training, the default state is that the brain does these things automatically. Any communication medium that forces the user to laboriously reprogram their own brain so that formerly automated behaviours have to reside under exclusive conscious control are impractical.
The most successful user input devices (ie. cars, telephones, pencils, keyboards) have always focussed on elements of interaction that are under direct control in the context of the use of that device (ie I don't control my feet while walking, but I do while sitting in a car, because the walking program isn't engaged)
Lets do some back-of-the-napkin rough math
How do you even figure out which side that is?
So? You have limited emissions to a very few sources, instead of having to worry about tens of thousands of catalytic converters and pollution control systems. It is a lot easier to deal with one or very few sources.
Yes it's very beneficial for the environment. The government can send all the fines for breaking EPA guidelines to just a few addresses rather than splitting them up among millions of drivers.
Talk about saving paper!
Up here, at less than 6 cents a kwh, and $4.50 a gallon for gas, running the car on cheap hydro sounds pretty good.
Electricity won't be 6 cents/kwh once everyone starts running their cars with it. Nor would gas stay at $4.50.
*dives into a bunker*
FIRE IN THE HOLE!
As processor speed, memory and disk space continue to spiral to ever larger values, Microsoft is really going to be put to task in finding ways to make Windows sluggish.
Their task is made more difficult by advances in compiler design which find an eliminate trivial solutions that simply chew up CPU time by computing huging cosine tables and then overwriting them.
New innovation may come from recent advances in polling network devices unnecessarily and hanging various threads until a reply is received. In the case of pulling a device off the network that Windows Explorer had browsed in the last 15 weeks, a given thread can hang for minutes, chewing up processor time in loops that scan network traffic.
The Windows Development team seems optimistic that they can produce the same crippled user interface on new 64 bit architectures that customers have become familiar with, a valuable marketing strategy in teaching consumers to become suspicious of computers with more responsive interfaces.
It's obvious they've become increasingly frustrated by the overwhelming atmosphere of stupidity that has descended over our politics and media lately.
They wanted to lash out at the source of their frustration, but in a way that didn't imperil their status as a reputable (well that's debatable) publication. So they choose the one day of the year when people can go nuts and say what they really mean, and then throw up their hands and say April Fools!
Our society is like a toned-down of Japan in this way, we have a built in release valve for venting our frustration at being bound by certain rules and regulations most of the time. *fwooot*
On the contrary, some jokes get funnier and funnier with retelling.
The problem is that converting one-off jokes into running gags is more art than science, and clearly something these poor guys have absolutely no grasp of.
This RFC was just pitiful, all the more so because of the obvious time investment.
Have I finally lost touch with the computing industry or is this just not funny?
Even though this kind of joke has been done to death, I had hoped for a smile to at least cross my lips.
Regardless, he's right.
so so bad
No it's not better. Microsoft still benefits when someone pirates XP, because they get a monkey on their back. And once they get used to XP's features, they'll be reluctant to change, even if that forces them to start paying for it because of new hoops they have to jump through.
It also means they are more likely to use IE and various other MS products.
You are being unrealistic because you assume it takes them the same amount of time to learn as you.
My parents have been working at this hard, for months now. But even the simplest aspects of using a computer have escaped them until recently. These people, god bless them, don't understand that a document, a batch file and a program are the same type of thing on the hard drive. It's just a question of the contents and whether the computer tries to run it or not.
Ditto with so many little things that you and I take for granted.
So what, for you, seems to be a trivial amount of courtesy time, is for them the equivalent of learning to speak a foreign language in terms of time investment. And it's not even as much fun.
I realized the perfect example to shut your gob.
My parents have been computer users since 1984. We had an Apple II+ (followed by a IIGS ugh), and then they some variety of PC's at work.
They are very organized and successful people, and they used those early computers with no problem. Word processing, spreadsheets, it was enormously useful and rarely gave us problems. As the years have gone by, the problems they have had with computers have increased exponentially.
This same story is being repeated in every computer owning household in the world.
But if it makes you feel better to stand on your soapbox spewing gibberish about stupid everyone is and if they just follow your 6 simple steps (each of which has 10 implicit substeps that would require hours of training to get through) then everything would be rosy... well keep doing it I guess.
Would you get over yourself already?
The means to properly operate a computer is as far beyond the reach of the average person as is the ability to tune their car and replace the fan belt. It's not that their stupid,it's that the concepts are completely alien to them. What seems trivial to you is a goddamned nightmare to most adults who grew up without touching a computer.
And there's also a selection bias, many people who are otherwise smart, just aren't good at dealing with computer-type systems. Those people have avoided computers until they couldn't any longer. Again, not that they're stupid (well not all of them), but everyone is good at some things, and bad at others.
Computers have gotten much harder to use successfully as they have gotten more powerful. It is not that the people who put together that 8086 did it "properly", it's that there were far fewer ways to screw it up.
Boy, having a SCUBA hobby has really mellowed you out.
There's a huge difference between the involvement of a passenger in a spaceship and a diver in SCUBA gear.
The appropriate analogy here is flying on an airplane, not scuba diving.
Yes it's true. My own site, which has only recently been offered to consumers at great rates offers VIAGRA and CIALIS tabs for next to nothing. These pills can enhance your performance while you are on one of the FREE VACATIONS that I also offer.
Really annoying actually.
not "could be"...it IS incredibly difficult to build a good walking bot. The best robot labs in the world are still making small, primitive insects that crawl around at a snail's pace. The physics of constructing them are far outside of the grasp of hobbyists.
Eliminating armor is the easier way to do it. What you are proposing raises the bar of design to such a level that you'd see designers spending the bulk of their time and money on the drivetrain and suspension which is not what we are paying to see.
We want simple driving with lots of traction, giving us the potential for lots of damage and kinetic energy, not dune buggies crashing into each other.
And walking bots? You're dreaming. There's a million reasons the military doesn't use them.
Here's an idea:
You are limited to about 2 square feet of metal sheeting at the biggest weight class, and only 1 mm thick aluminum.
Then tighten up the weight restrictions a bit, making it harder to rig up makeshift armor out of non-sheet shaped metal
This would shift the emphasis towards robots that can deal damage, not take it. You have limited weight and armor, so you use that weight to do as much damage as possible.
This means exposed guts.
Exposed guts means fun times for spikey/smashy robots, and even those flame throwers might hope to melt some wiring.
Think of it like evolution, we shift the environment to make the current optimal solutions impossible, so now everyone has to find a totally new way to go.
That poor library, they'll be doing a complete reinventory.
Probably lost some of their best works.