The main question is the final goal. Are you trying for academic or trade success. That is are you trying to teach How to program or to Get a Job.
There is a huge difference. You can get a job programming in language X (For most X anyways) without knowing how to program. However if you know how to program, you can learn language X (For all X) in short order.
It would seem that if you want to get a job, C/C++ woulud be the better lang for you to learn. It is firmly rooted in a low-level environment and more forgiving of lack or order or structure (generally).
If you want to learn how to program though java would seem to be a better starting point. You can more easily reach higher level concepts, OO, Program Architecture/Design, Data Structures, and Algorithms without have to worry about the lowlevel items. Once you have a firm grasp of these concepts it is easier to learn the lower level hardware concepts. Plus once you do, you should already have a sharp view for design elements that can easily be forgotten in C/C++.
To learn to program you should probably go the route of: Boolean Logic - basic understadning of binary and logical concepts. Java - intro to language leading into OO design and program architecture. After that take Algorithms (focus on basic algorithms and algorithmic analysis) Then concurently take advanced algorithms (Graph Algorithms), and Data structures ( De-Queues, queues, stacks, tree/graphs, heaps etc). Then learn C/C++ concurently with Basic Hardware (Flip-Flops, adders, 2's complents and such). Finally do OS Design/Compilers along side of advanced hardware (RISC pipelines, cache etc).
Then you will *know* how to program. Also throw in comparitive languages in there eg lisp, ml, asm, language design etc.
If you really want a job learn to program VB. You'll be brain dead, but so what. It pays.
Check out paynestudios.com The artist there makes dinousar skeltons out of steel. Along with some wire cabling you can make em move, and even snap. I saw them at a trade show last year, and they are very cool. They are for sale too. Less than troody too.
It seems to be a great idea. The only issue is item #3
iltering is opt-out by default. By this, I mean that if a site does not appear in any of your subscribed lists, it gets blocked.
Unfortunately, this is the only real way (short of analyzing content in real-time) to ensure that various types of web spammers,
particularly pornographers, can't fo an end-run around the filtering. Note that I don't think this should be the only mode of
operation; just that the default settings should be "disallow access unless approved by me or someone I have decided to trust".
What happens when junior needs to do a research paper on 18th century mosques with the "default" mode?
The internet/www which is a great research tool suddenly becomes useless without a user with permission to do the list editing. However if you give junior the ability to do editing then you have defeated the point of filtering. Also it would be difficult to find a list that would cover possible research items.
Maybe one of the modes of operation should be that unlisted sites are loged, as a link/thumbnail (thumbnail for the lazy). That way a parent could review the list later and talk to junior if necessary about what they are doing on the net
If anyone is actually interested in doing something like this I sure you could get a lot of support. It's a greay project, and not controlled by a bunch or corporate morons that you don't know.
the problem would seem to be what are you going to key the network off of?
The text on the site? Remove all text from your porn site, make it all gif's.
Change the url all the time and they can't key off of that either. It may work for some things, but it's going to be well abused by lots of people and most likely confused, just like when you give a neural net too much bad data.
It's not going to work on the photo's that's too much data. And of course if you did have a nerual net to tell porn images from non-porn images don't you think that some porn site would have one that would automatically differentiate good porn from bad?
One of the most lossy parts of his photography had
to be getting the filters just right. That is at the time what was the state of color-polarizing lenses? Sligh imperfections could cause color loss
between the lenses. If you project it back though
with the same lenses that you photograph with, you
should be fairly good.<BR><BR>
One of the ways that I was thinking that you could
do something like this was by using a prism. The prism is located directly behind the main lense, which would split the light into colors. This could be mathematicly calculated, and three other lenses could grab ranges of the prism, and redirect them correctly onto three seperate plates. Now using trig, and refraction you could probably caclulate exactly where to place the redirecting lenses, and the photo-plates.<BR><br>
Anyone interested in calculating the size of the camera for this method from the size of the photo plates?<BR><br>
Of course projecting them becomes a little harder, but you could light the plate, send it through a prism, and then redirect each color component through it's own lense. Looking at the projector would easily prove that he didn't use the system.<BR><br>
He would also have had the problem of synchornizing all threee shutters for each plane, but it would be an interesting science project for someone to try.
You are not communicating at FTL.
Ok, you create an entangled pair of particles, or very many pairs in this case. You ship one pair off to alpha centuri. Now when you read one of the particles, you know the value of the other particle. BUT since you can't force the value you get when you read you have not transmitted anything. You simply know the value of the particle that you sent 7 years ago to apha cenrui. To do anything useful with that you would have to send a signal to alpha centauri, which again would take 7 years. SO you have not communicated anything faster that the speed of light.
Johns Hopkins University Applied Physics Lab built it, and I belive the big thing at the time was that it was a million dollars UNDER budget. I'd look for a link, but their site is/.'ed
And unlike some other contractors they apprently don't get metric and english units confused.
Do a search for SBC or single board computer. LinuxDevices has a section on them. Also see Advantech they have a bunch of 5.25 or 3.5 form factor computers with no fans. I got a 300Mhz Geode box (x86 pent class) with ether, sound, video, KB/Mouse, Floppy, UDMA 33(LINUX SUPPORTED) and I am pretty sure that you can get them with TV out support, though I'm not sure if there are linux drivers for em. They also have a few board that support the bigger mobile pentiums, but that would most likely require a fan.
Yes this is a powerPC board, but for ~500$ you can get a 3.5 form factor x86 with dual ether, and real video. I got one for dev at work from Advantech The entire thing run off of a 5v suppy. Get a laptop hard drive (44pin IDE) and you have an entire 5v system that will take a standard power plug from your power suppy.
Emjembedded also has similar stuff.
The article does not mention alot of things that make quantum computing difficult (Programming wise, not implementation wise). At this point in time there are very very few algorithms for quantum computers (That are better than an equiv algorithm for classical computers) because they are difficult to write and prove correct. Let alone develop an operating system for it. The first quantum computing devices are much more likely to supplement a classical computer. I'll kind of be like a special math coprocesser for very specific functions.
It also seemed to leave out the most important quantum computing algorithm, Shor's Algorithm. Shor's Algorithm factors primes in polynomial time. Which basicly comes do to that with it you can crack encryption as quickly as you can encrypt with it. So as you can guess the US gov. dosn't want a quantum computer to replace it's classical ones, except for the ones at the NSA....
Slashdot Mirror
The main question is the final goal. Are you trying for academic or trade success. That is are you trying to teach How to program or to Get a Job.
There is a huge difference. You can get a job programming in language X (For most X anyways) without knowing how to program. However if you know how to program, you can learn language X (For all X) in short order.
It would seem that if you want to get a job, C/C++ woulud be the better lang for you to learn. It is firmly rooted in a low-level environment and more forgiving of lack or order or structure (generally).
If you want to learn how to program though java would seem to be a better starting point. You can more easily reach higher level concepts, OO, Program Architecture/Design, Data Structures, and Algorithms without have to worry about the lowlevel items. Once you have a firm grasp of these concepts it is easier to learn the lower level hardware concepts. Plus once you do, you should already have a sharp view for design elements that can easily be forgotten in C/C++.
To learn to program you should probably go the route of: Boolean Logic - basic understadning of binary and logical concepts. Java - intro to language leading into OO design and program architecture. After that take Algorithms (focus on basic algorithms and algorithmic analysis) Then concurently take advanced algorithms (Graph Algorithms), and Data structures ( De-Queues, queues, stacks, tree/graphs, heaps etc). Then learn C/C++ concurently with Basic Hardware (Flip-Flops, adders, 2's complents and such). Finally do OS Design/Compilers along side of advanced hardware (RISC pipelines, cache etc). Then you will *know* how to program. Also throw in comparitive languages in there eg lisp, ml, asm, language design etc.
If you really want a job learn to program VB. You'll be brain dead, but so what. It pays.
Check out paynestudios.com The artist there makes dinousar skeltons out of steel. Along with some wire cabling you can make em move, and even snap. I saw them at a trade show last year, and they are very cool. They are for sale too. Less than troody too.
It seems to be a great idea. The only issue is item #3
iltering is opt-out by default. By this, I mean that if a site does not appear in any of your subscribed lists, it gets blocked. Unfortunately, this is the only real way (short of analyzing content in real-time) to ensure that various types of web spammers, particularly pornographers, can't fo an end-run around the filtering. Note that I don't think this should be the only mode of operation; just that the default settings should be "disallow access unless approved by me or someone I have decided to trust".
What happens when junior needs to do a research paper on 18th century mosques with the "default" mode?
The internet/www which is a great research tool suddenly becomes useless without a user with permission to do the list editing. However if you give junior the ability to do editing then you have defeated the point of filtering. Also it would be difficult to find a list that would cover possible research items.
Maybe one of the modes of operation should be that unlisted sites are loged, as a link/thumbnail (thumbnail for the lazy). That way a parent could review the list later and talk to junior if necessary about what they are doing on the net
If anyone is actually interested in doing something like this I sure you could get a lot of support. It's a greay project, and not controlled by a bunch or corporate morons that you don't know.
the problem would seem to be what are you going to key the network off of?
The text on the site? Remove all text from your porn site, make it all gif's.
Change the url all the time and they can't key off of that either. It may work for some things, but it's going to be well abused by lots of people and most likely confused, just like when you give a neural net too much bad data.
It's not going to work on the photo's that's too much data. And of course if you did have a nerual net to tell porn images from non-porn images don't you think that some porn site would have one that would automatically differentiate good porn from bad?
One of the most lossy parts of his photography had
to be getting the filters just right. That is at the time what was the state of color-polarizing lenses? Sligh imperfections could cause color loss
between the lenses. If you project it back though
with the same lenses that you photograph with, you
should be fairly good.<BR><BR>
One of the ways that I was thinking that you could
do something like this was by using a prism. The prism is located directly behind the main lense, which would split the light into colors. This could be mathematicly calculated, and three other lenses could grab ranges of the prism, and redirect them correctly onto three seperate plates. Now using trig, and refraction you could probably caclulate exactly where to place the redirecting lenses, and the photo-plates.<BR><br>
Anyone interested in calculating the size of the camera for this method from the size of the photo plates?<BR><br>
Of course projecting them becomes a little harder, but you could light the plate, send it through a prism, and then redirect each color component through it's own lense. Looking at the projector would easily prove that he didn't use the system.<BR><br>
He would also have had the problem of synchornizing all threee shutters for each plane, but it would be an interesting science project for someone to try.
You are not communicating at FTL. Ok, you create an entangled pair of particles, or very many pairs in this case. You ship one pair off to alpha centuri. Now when you read one of the particles, you know the value of the other particle. BUT since you can't force the value you get when you read you have not transmitted anything. You simply know the value of the particle that you sent 7 years ago to apha cenrui. To do anything useful with that you would have to send a signal to alpha centauri, which again would take 7 years. SO you have not communicated anything faster that the speed of light.
Get it?
Johns Hopkins University Applied Physics Lab built it, and I belive the big thing at the time was that it was a million dollars UNDER budget. I'd look for a link, but their site is /.'ed
And unlike some other contractors they apprently don't get metric and english units confused.
Do a search for SBC or single board computer. LinuxDevices has a section on them. Also see Advantech they have a bunch of 5.25 or 3.5 form factor computers with no fans. I got a 300Mhz Geode box (x86 pent class) with ether, sound, video, KB/Mouse, Floppy, UDMA 33(LINUX SUPPORTED) and I am pretty sure that you can get them with TV out support, though I'm not sure if there are linux drivers for em. They also have a few board that support the bigger mobile pentiums, but that would most likely require a fan.
Yes this is a powerPC board, but for ~500$ you can get a 3.5 form factor x86 with dual ether, and real video. I got one for dev at work from Advantech The entire thing run off of a 5v suppy. Get a laptop hard drive (44pin IDE) and you have an entire 5v system that will take a standard power plug from your power suppy. Emjembedded also has similar stuff.
The article does not mention alot of things that make quantum computing difficult (Programming wise, not implementation wise). At this point in time there are very very few algorithms for quantum computers (That are better than an equiv algorithm for classical computers) because they are difficult to write and prove correct. Let alone develop an operating system for it. The first quantum computing devices are much more likely to supplement a classical computer. I'll kind of be like a special math coprocesser for very specific functions. It also seemed to leave out the most important quantum computing algorithm, Shor's Algorithm. Shor's Algorithm factors primes in polynomial time. Which basicly comes do to that with it you can crack encryption as quickly as you can encrypt with it. So as you can guess the US gov. dosn't want a quantum computer to replace it's classical ones, except for the ones at the NSA....
These guys are the ones who built near. more nice pictures and some videos. Near stuff.