We use NN's for similar purposes at my work. The particular ones that we use are Neuralware's NGO, SNNS (without the JavaGui) and the Nueral Network Toolbox for Matlab.
Since we distribute the code for the final NN, we've found that writing up our own C NN to be ideal. There are many books that tell you how to do this. A feedforward linear NN is very easy to program yourself. The trainer is what get's complicated. The best/cheapest solution that we've found is to use trainers from COTS packages, such as those that I mentioned, and then implement the trained NN with our own code.
As we evolve, we are slowly developing our own trainer with features that are specific to my area of work.
The basic fact, which Neal Stephenson stated so eloquently, is this:
Microsoft is in trouble because their #1 competitor is FREE.
Think about business models. There is one market that cares that OSS is free as in speech (i.e. they can tinker and optimize), and there is another market that cares that the OSS is free as in beer. I like Windows NT and especially windows 2000, but if I need to optimize my OS I don't use it. If I don't want to pay for my OS and still be legal, then I don't us windows.
As far as I'm concerned, the MS OS monopoly is a sinking ship. The fact is that GNU Linux already makes a better server and in my opinion, development platform than any propietary OS. All MS can do is hope that it gets enough marketshare in the PDA market as well as the.net ASP market.
Linux may never catch up to windows in the desktop market, because of the potential decline in the market and the fact that 99% of the population are familiar with windows and don't want the hassle of something new, BUT... Linux will continue to be 1 step behind windows, and should MS ever stop pumping so much effort into the desktop, Linux will overtake windows on the desktop.
Cliff! You ruined the submission by answering it. As an editor of/. you should encourage disscussion, not sequester it with your own knowledge.
In the future, please refrain from making comments that completely answer a submission, rather, enhance the topic by adding thought provoking thoughts, such as...
"What might one do with a Beowulf cluster of such SMP laptops?"
...or inflammetory thoughts...
"Finally, laptops might have enough processing power handle Mozilla!"
...or a comment that brings other market segments into the disscussion...
"I'll bet that OS X would run lightning fast on such a laptop and how many simultaneous Napster downloaded MP3s could you play on such a laptop?"
Just don't mention realistic facts that would make such a system unfeasable! By the way, this comment is mostly just a joke.;-)
I really hope that I am avoiding the usual MS bashing
You are not avoiding it; You are adding to it.
Your second comment about the majority of conformist sheep blindly using MS. You are correct, my friend. You are correct.
The same people just recently became able to comprehend how to use a computer. They have yet to form opinions about whether MS is a good choice or not. Most of them don't even know that MS is a choice or what exactly an operating system is.
My faith in humanity is decreasing everyday.
You should have faith. People will learn. Enough of them will eventually jump off the MS ship to either turn Linux into a viable desktop OS or to force MS to compete and MS will turn Windows into a viable desktop OS. Neither of which has yet happened.
I assure you that I did read the article before commenting. I disagree with what he said regarding the autoclave, that's all.
Most of what I mentioned comes from experience comparing nylon thermoset casting processes with injection molding techniques. The process described in the article sounds like a very similar comparison, except with an autoclave instead of an injection die.
The autoclave uses a vaccuum to remove airpockets in much the same way that injection molding uses high pressure to 'push-out' air-pockets. I based my opinion on that assumption. Of course it may be wrong. Thanks for pointing that out. I'm going to look into it in further detail and see what I can figure out.
The main reason that the Automobile industry will never use this is the same reason that it hasn't gone anywhere with ceramic engines. In the early '80s, the big three auto companies did research on Ceramic engines that could run hotter and were lighter and stronger than steel engine blocks. The reason that the development never went anywhere had nothing to do with ceramics' brittle nature (which is just a design technicallity). The reason was because ceramic engines couldn't be recycled. Towards the middle of the '80s, recycling automobiles became a big deal and it has been ever since.
Carbon fiber paneling won't ever replace aluminum because
Aluminum is Earth's most abundant metal
Aluminum is infinitely more recycleable than carbon fiber composites. The only thing that you can recycle carbon fiber composites into is chopped-carbon fiber composites.
The process from the article is useful for a couple of things. Large carbon fiber shapes can be created because an autoclave is no longer neccessary. Protype carbon fiber components can be created because of no autoclave and cheap molds can be used.
I don't feel that high quality carbon fiber components will ever be created by this process. The vaccuum used in an autoclave is necessary to 'suck' out airpockets and provide a uniformly dense material. The carbon fibers are brittle and the stress concentrations that arrise in microscopic air pockets are enough to fracture the fibers (eventually). This same problem has occured over and over again in other thermoset composite processing techniques (what the article describes). I worked on a thermoset production process for Glass Fiber enhanced Nylon.
These techniques are innovative and they have important applications. However, these applications are limited to low-cost and low-quality components, or components where the traditional processes can't be used (complex geometry components, or, large/small scale components).
I had just installed the current version of Nautilus about 3 weeks ago. Yeah, I hate the find cron job that Mandrake Linux installs on my Iopener. That was one of the first things to go.
Does Ximian Gnome use Medusa? I hate Medusa because I'll be sitting at my PC doing something casual (pr0n, typing, listening to music) and suddenly the harddrive starts spinning like there's no tommorrow! I try to kill the processes, but they keep starting back up. Finally, I uninstalled Nautilus and all of the Medusa garbage.
Medusa is only a file finding utility! Why does it need it's own daemon and to spin my harddrive every half hour? Dios mios!
This is annoying on the same scale that the 'Windows Explorer crashes take down my entire system' is annoying.
Besides Medusa, Nautilus and Ximian Gnome are pretty good.
you are correct that eventually the sequence of characters will come up. However, in 1/2 of the cases, your initial Z will be larger than the size of the string of characters, whether it is in magnitude or precision (or both).
A similar approach may be used with the logistic map (a=4... X(n+1)=4 X(n) * (X(n)-1), 0X(n)1 )
I will avoid splitting the sentences in the future.
What were the/. readers' opinions on this subject? It sounds to me like the person offering the $5000 was scamming everyone because it's theoretically impossible to win. This fellow from Japan knows this, but due to a wording technicality in the contract, he figures he can scam the $5k from the scammer. They both seemed to be good natured chaps about it, but I feel that they still have some deep resentment towards each other.
Even though Hemos said 'IMHO', I wouldn't agree that it was honest. He happens to have a large amount of capital invested in British Telecom.
Anywhere that you can have mass government deregulation, it's a good idea to invest in the telecom company because you know that they will start raping^H^H^H reaping the profits by being a mass monopoly.
Okay, mister optical know-it-all. There's a flaw in your idea (even though I must admit that it is an interesting and cool^H^H^H good one).
One problem is if you send two light sources one-way on a fiber. The problem then would be that you have to stick with glass components. Any plastic material that you emit the light through is birefringent, meaning that it will break the light into the plastic's own X and Y axis. This, by itself, is okay becuase under perfect conditions the light would be reconstructed once it came out of the plastic material. However, plastics exibit photoelasticity, so stress and strains on the plastic would alter the light differently for the two directions, altering the altering the signals differently depending upon the wave axis and the stress in the fibre. A WDM would split the light back into it's original axes, but again, the problem is that the axes were effected individually by the plastic material.
One way to circumvent this is to use all glass fiber and glass components or use multiple optic fibers. One for each signal.
I like your idea for lighting up the keyboard lights using the light from the fiber. Maybe in the future, one of the components in every computer will be a Laser!
Can you tell me how to power my mp3 player off of light? Then I would be impressed.
Whoa, slow down partner. USB and Firewire have something that optical will never have, and that is power. You will never be able to send electricity down an optical cable. The only way to power something would be to send a bright light and use solar panels on the other end--not likely.
Furthermore, SCSI has direct memory access. Unless the new bus has DMA, then SCSI will still have a niche market. IDE? Well, maybe it is time to retire IDE.
Someone tell Tux to get a haircut and go get a job! I'm not going to support any lousy hippy penguins.
I think that Sid said it best in Sid & Nancy, "None of that free-love hippy shit here!" I think that the Linux community needs a new mascot, especially in light of the recent IBM support. We can change the name to Hard-working-nux and the mascot can be a tall white vietnam soldier with short hair. There's a mascot that we can be proud of! You won't see hard-working-nux defacing America with filthy love and peace. hard-working-nux will come back from Vietnam in time to find a nice job and support his 1.5 kids and a wife.
OMG, it's late and this post is acually pretty stupid.
This type of feature isn't new at all. Have you ever used a CAD package? Many of the top packages such as Catia and Unigraphics (to a lesser extent) use a similar type of control. Although I don't have any experience with Ideas, I've heard that it uses gesture based control to a large degree.
The reason that CAD tools have had these features for longer is that they are inherently more complicated. For example, just to rotate and translate your 3-D model, you have 3 axii of rotation and 6 directions of tranlation. How do you do that without some form of Gesture based control?
Maybe the idea of 'gesture based control' as used in web browsers will become the norm of Office/productivity apps. Finally, companies such as Microsoft don't need to dumb down their products anymore, since 99.9% of the office/productivity force use computers. They can concentrate on making them more efficient and not easier to use. I'm still waiting for a webbrowser that can see my eyes and determine which link that I'm looking at.
Slashdot Mirror
We use NN's for similar purposes at my work. The particular ones that we use are Neuralware's NGO, SNNS (without the JavaGui) and the Nueral Network Toolbox for Matlab.
Since we distribute the code for the final NN, we've found that writing up our own C NN to be ideal. There are many books that tell you how to do this. A feedforward linear NN is very easy to program yourself. The trainer is what get's complicated. The best/cheapest solution that we've found is to use trainers from COTS packages, such as those that I mentioned, and then implement the trained NN with our own code.
As we evolve, we are slowly developing our own trainer with features that are specific to my area of work.
Microsoft is in trouble because their #1 competitor is FREE.
Think about business models. There is one market that cares that OSS is free as in speech (i.e. they can tinker and optimize), and there is another market that cares that the OSS is free as in beer. I like Windows NT and especially windows 2000, but if I need to optimize my OS I don't use it. If I don't want to pay for my OS and still be legal, then I don't us windows.
As far as I'm concerned, the MS OS monopoly is a sinking ship. The fact is that GNU Linux already makes a better server and in my opinion, development platform than any propietary OS. All MS can do is hope that it gets enough marketshare in the PDA market as well as the .net ASP market.
Linux may never catch up to windows in the desktop market, because of the potential decline in the market and the fact that 99% of the population are familiar with windows and don't want the hassle of something new, BUT... Linux will continue to be 1 step behind windows, and should MS ever stop pumping so much effort into the desktop, Linux will overtake windows on the desktop.
In the future, please refrain from making comments that completely answer a submission, rather, enhance the topic by adding thought provoking thoughts, such as...
"What might one do with a Beowulf cluster of such SMP laptops?"
...or inflammetory thoughts...
"Finally, laptops might have enough processing power handle Mozilla!"
...or a comment that brings other market segments into the disscussion...
"I'll bet that OS X would run lightning fast on such a laptop and how many simultaneous Napster downloaded MP3s could you play on such a laptop?"
Just don't mention realistic facts that would make such a system unfeasable! By the way, this comment is mostly just a joke. ;-)
Your second comment about the majority of conformist sheep blindly using MS. You are correct, my friend. You are correct.
The same people just recently became able to comprehend how to use a computer. They have yet to form opinions about whether MS is a good choice or not. Most of them don't even know that MS is a choice or what exactly an operating system is.
You should have faith. People will learn. Enough of them will eventually jump off the MS ship to either turn Linux into a viable desktop OS or to force MS to compete and MS will turn Windows into a viable desktop OS. Neither of which has yet happened.Most of what I mentioned comes from experience comparing nylon thermoset casting processes with injection molding techniques. The process described in the article sounds like a very similar comparison, except with an autoclave instead of an injection die.
The autoclave uses a vaccuum to remove airpockets in much the same way that injection molding uses high pressure to 'push-out' air-pockets. I based my opinion on that assumption. Of course it may be wrong. Thanks for pointing that out. I'm going to look into it in further detail and see what I can figure out.
The main reason that the Automobile industry will never use this is the same reason that it hasn't gone anywhere with ceramic engines. In the early '80s, the big three auto companies did research on Ceramic engines that could run hotter and were lighter and stronger than steel engine blocks. The reason that the development never went anywhere had nothing to do with ceramics' brittle nature (which is just a design technicallity). The reason was because ceramic engines couldn't be recycled. Towards the middle of the '80s, recycling automobiles became a big deal and it has been ever since.
Carbon fiber paneling won't ever replace aluminum because
Aluminum is Earth's most abundant metal
Aluminum is infinitely more recycleable than carbon fiber composites. The only thing that you can recycle carbon fiber composites into is chopped-carbon fiber composites.
The process from the article is useful for a couple of things. Large carbon fiber shapes can be created because an autoclave is no longer neccessary. Protype carbon fiber components can be created because of no autoclave and cheap molds can be used.
I don't feel that high quality carbon fiber components will ever be created by this process. The vaccuum used in an autoclave is necessary to 'suck' out airpockets and provide a uniformly dense material. The carbon fibers are brittle and the stress concentrations that arrise in microscopic air pockets are enough to fracture the fibers (eventually). This same problem has occured over and over again in other thermoset composite processing techniques (what the article describes). I worked on a thermoset production process for Glass Fiber enhanced Nylon.
These techniques are innovative and they have important applications. However, these applications are limited to low-cost and low-quality components, or components where the traditional processes can't be used (complex geometry components, or, large/small scale components).
I had just installed the current version of Nautilus about 3 weeks ago. Yeah, I hate the find cron job that Mandrake Linux installs on my Iopener. That was one of the first things to go.
I think that Ruby is kewl. In Japan, Ruby is used more than Python!
Medusa is only a file finding utility! Why does it need it's own daemon and to spin my harddrive every half hour? Dios mios!
This is annoying on the same scale that the 'Windows Explorer crashes take down my entire system' is annoying.
Besides Medusa, Nautilus and Ximian Gnome are pretty good.
On the opposite side of my desk from all of the right handed peripherals!
A similar approach may be used with the logistic map (a=4 ... X(n+1)=4 X(n) * (X(n)-1), 0X(n)1 )
What were the /. readers' opinions on this subject? It sounds to me like the person offering the $5000 was scamming everyone because it's theoretically impossible to win. This fellow from Japan knows this, but due to a wording technicality in the contract, he figures he can scam the $5k from the scammer. They both seemed to be good natured chaps about it, but I feel that they still have some deep resentment towards each other.
Any other opinions?
because you can't compress random data. He can just generate a file of random data. This is just a scam so he can make $100 off of every entry.
Anywhere that you can have mass government deregulation, it's a good idea to invest in the telecom company because you know that they will start raping^H^H^H reaping the profits by being a mass monopoly.
One problem is if you send two light sources one-way on a fiber. The problem then would be that you have to stick with glass components. Any plastic material that you emit the light through is birefringent, meaning that it will break the light into the plastic's own X and Y axis. This, by itself, is okay becuase under perfect conditions the light would be reconstructed once it came out of the plastic material. However, plastics exibit photoelasticity, so stress and strains on the plastic would alter the light differently for the two directions, altering the altering the signals differently depending upon the wave axis and the stress in the fibre. A WDM would split the light back into it's original axes, but again, the problem is that the axes were effected individually by the plastic material.
One way to circumvent this is to use all glass fiber and glass components or use multiple optic fibers. One for each signal.
I like your idea for lighting up the keyboard lights using the light from the fiber. Maybe in the future, one of the components in every computer will be a Laser!
Can you tell me how to power my mp3 player off of light? Then I would be impressed.
Whoa, slow down partner. USB and Firewire have something that optical will never have, and that is power. You will never be able to send electricity down an optical cable. The only way to power something would be to send a bright light and use solar panels on the other end--not likely.
Furthermore, SCSI has direct memory access. Unless the new bus has DMA, then SCSI will still have a niche market. IDE? Well, maybe it is time to retire IDE.
Why can't we just all get along and use the agp bus?
Guppy06 - please spell check.
I think that Sid said it best in Sid & Nancy, "None of that free-love hippy shit here!" I think that the Linux community needs a new mascot, especially in light of the recent IBM support. We can change the name to Hard-working-nux and the mascot can be a tall white vietnam soldier with short hair. There's a mascot that we can be proud of! You won't see hard-working-nux defacing America with filthy love and peace. hard-working-nux will come back from Vietnam in time to find a nice job and support his 1.5 kids and a wife.
OMG, it's late and this post is acually pretty stupid.
Linux is secure by de fault of de haxx0rs that hack de system and find de security holes.
In what way is this sad? Furthermore, why did this post get moderated as Funny???
The reason that CAD tools have had these features for longer is that they are inherently more complicated. For example, just to rotate and translate your 3-D model, you have 3 axii of rotation and 6 directions of tranlation. How do you do that without some form of Gesture based control?
Maybe the idea of 'gesture based control' as used in web browsers will become the norm of Office/productivity apps. Finally, companies such as Microsoft don't need to dumb down their products anymore, since 99.9% of the office/productivity force use computers. They can concentrate on making them more efficient and not easier to use. I'm still waiting for a webbrowser that can see my eyes and determine which link that I'm looking at.
I probably don't need to even point this out, since it is obvious to everyone reading, but what you meant to say was:
I dont see people trading 15Ton locomotives any time soon.
I'm sure it was a typo or something, but please preview your posts before you submit them. I'm glad to have helped.
-grammar nazi
You said:
but what you actually meant was:Please be careful to get that straight in future posts. I'm glad to have helped out.-grammar nazi
I'd like to see the DeBeers' own that!