For binary it is trivial: Since it is an infinite, noncyclical number, there must be a zero some finite digits after any one. This must repeat ad infinity, so there must be an infinite numbe of zeroes.
Alternative would be that all digits after, say, digit 1,000,000,000 are ones. This could be written as a fraction, and would not be an irrational number.
You can make a similar argument for a decimal representation for pi. If you can come up with a good one, please reply.
Design parameters for a fusion reactor: 1. Shielding: 10m of water or similar as well as magnetic shielding 2. Energy density 10kW/m2 3. Politics: Not in my backyard
Conclusion: Sun 1. Atmosphere and earth magnetic field: perfect 2. perfect almost anywhere 3. 150 million km away: perfect
Correct: You can pick any non-accelerated reference frame as center.
Earth is accelerated around the sun, and the sun is, to a lesser degree, accelerated around the milky way, as well as around the solar systems barycenter. It is anyway usually implied in the phrase "heliocentric reference" that a non-accelerated reference is picked.
Try to write simple laws of physics if your reference point is accelerated in a gravity well.
There has been a long and hostile fight between the Flat Earth Society (FES) and the Geocentrists. The conservative FES claim ancient heritage, and will not accept any newfangled ideas from the Geocentrists.
0. Buy a sacrificial computer. Keyboard and optical mouse can be $10 each. 1. Take a mouse and wire all buttons so they all are "left buttons". 2. Now open a browser to pbskids.org by default. 3. Also set your desktop up to accept a single click to "open" icons on desktop 4. And go from there
Your toddler will first have fun for days on pbskids. Will discover the desktop after that. You can sprinkle it with some icons of interest.
Can MIT come up with something else than "solar powered nano technology autonomous robot swarms". Is this the "Build a really cool solution, and then spend a decade looking for the problem it solves" Would be nice to see some innovation there. Are they not supposed to have the skills and intelligence to think outside the box and go in new directions? - a NON-autonomous robot - a FOSSIL powered something - some new MACRO technology.
There is no problem to get a lens with better resolution than this. A fixed focus would certainly do just fine. At 750nm, and with n.a. ~1, a lens would need a diameter equal to the resolution. The diagonal resolution is sqrt(x^2+y^2) or 16146. So a stepped down lens should have d>12mm.
Secondly, if you look at a distant object, like a cell tower on a distant mountain, your eyes will resolve down to 20 seconds of arc (Sparrow's resolution limit). Your total field of view is 120x60 degrees (of couse not high resolution away from focus, but just to be able to make an image with same resolution as your eyes, and allow the eyes to wander the image), you will need: 120deg/20"=20kpix x 10kpix or 200Mpix.
I can not see how this is a lot, nor can I see how this can be hard for a simple lens to focus properly. I can see how a 10x optical zoom would be problematic, but well understood optics combined with a powerful DSP would fix that. Look at how the fuzzy images on Hubble (before the optics was fixed) was deconvolved with simple math.
I hate to say this as a former Tektronix engineer, but you seldom need a scope, and if you do it is typically application targeted and expensive. So the general purpose scope that all self respecting EE's used to have on the lab desk is a thing of the past. All digital work either has debuggers or with FPGAs, Chipscope Pro or other. No scope needed. And if you really need to see how the eye diagram looks with your 10GB Ethernet, the best scope may be your receiver chip. Hard to find a 40GHz scope anyway.
I actually had a Tek 2440 300MHz Digital Storage for at least a decade, but used it less and less. Became more a educational thing to show kids how AC looked. All serial interfaces are running at muli GHz speed, and RF development is more in the 5.7GHz range and higher (802.11n) Not many scope sampling at 4x or more at those frequencies.
This setup works flawlessly in most settings. Pros: Works like a normal cellphone with dialing options etc. Use your phone lists etc Synchronizes with PC apps Good range
Cons: Have to preconfigure all hotspots. Will not hop between hotspots during same conversation Not intuitive network configuration. Phone jumps to 3G unless you dig down to the setting to force it to wifi only.
A fiber through a neighborhood drops one 10Gb color per highrise. An edgerouter in the basement connects each of the 36 floors with Gb ethernet. Each floor has GigE routers to 30 apartments.
Actual cost of installation in a dense urban deployment: 1km of fiber between each highrise: $60k (This was a shared cost ditch) 10Gb uplink edgerouter $30k 1k ports Gig switches: $8k In building infrastructure(wiring, power etc: $56k
total cost for last mile for dense urban deployment per 100Mb customer: $142
Montly depreciation with 5 year life on equipment and 20 year on infrastructure: $1.65
Provisioning, management, billing, email, DNS servers etc: $3.50 Marketing, DRM, security, law enforcement support, Lobby efforts: $5.50
Adjust these numbers for yourself, and there is no way you can get more than 3x even if you goldplate it all.
Wifi can keep cost down in less dense deployments.
It is remotely comparable. So maybe it will be 100x the cost for the last mile. That makes 100Mb for a few dollars.
The point is that bandwidth is not a limited resource like electricity and water with a natural based scarcity, and an economical model based on that.
The scarcity of bandwidth in the USA is mostly a marketing gimmick to be able to differentiate cost between customers to maximize profit to the detriment of the overall speed of access.
20Mb/s-50Mb/s is common for basic internet in much of the world outside the USA
The single mode, Non-Zero Dispersion Shifted fibers is of course optimized for DWDM. That means that a buyer can put at least 128 colors in the fiber, each with 10Gb/s. With 423 fibers in the bundle, that adds up to 0.5Pb/s.
With 10x oversubscription, this will supply 541 million homes with 100Mb/s broadband each.
That should cover all of the americas with 100 million is USA, 46 million in Brazil, and 12 million in Canada.
Just add harddrives everywhere, and add them together in a virtual folder. Buy 2TB sata drives internally and USB external SATA docks. Mirror them pairwise if you like, but don't do RAID. Just adds more problems than is solves.
Lets say you have 4 computers that will take 6 SATA 1 ESATA and 2 PATA, and with a dual USB sata docking station, you will have 22TB/computer, so you will have 88TB total for all computers. Now add docking for your USB storage off your routers for maybe 90TB. Copy an entire 2TB HDD to a docked HDD, and store offline as many as you like. A bookshelf holds easily 1000TB.
Each computer runs a script to share the drives on startup, so they are all available easily.
If you like, you can add truecrypt to them all as well, and include this in your script.
The scriptS may look like this with Win: MOUNT.CMD "C:\Program Files\TrueCrypt\TrueCrypt"/quit/volume c:\lib\disk1\libM/letter M/cache y/beep "C:\Program Files\TrueCrypt\TrueCrypt"/q/v c:\lib\disk2\libN/lN/b "C:\Program Files\TrueCrypt\TrueCrypt"/q/v c:\lib\disk3\libO/lO/b "C:\Program Files\TrueCrypt\TrueCrypt"/q/v c:\lib\disk5\libP/lP/b.....
SHARE.CMD net share tvM=M:\lib net share tvN=N:\lib net share tvO=O:\lib net share tvP=P:\lib.....
Quick terminology: Spectral color- Pure, single wavelength color, like a laser. Composite color- A combination of many spectral colors of different intensity. To truly reproduce a color, each pixel should be able to not only make one spectral color, but a combination of all of them.
This would be very expensive, and fortunately, our eye have sensors only for Red 580 nm, Green 540nm, and Blue 440 nm (RGB), if we exclude the low light rods. We can therefore get away with RGB screens. There are slight errors. For example, assume each R-G-B pixel emits light matching the eyes R-G-B sensors peak sensitivity. Now, we can reproduce any light stimulation by exiting a linear combination of the three emitters. The eye however is sensitive from 380 nm to 740 nm, and can obviously not create the stimulation for neither 400 nm light, nor 700 nm, as your linear combination of only positive values will not cover these spectral colors (outside the gamut of the display). Take a picture of a prism spectrum or rainbow, and compare the original with what you see on the monitor, and you can see this.
So bottom line, RGB covers almost all colors, but adding emitters allows linear combination to cover more of the possible stimulation, but a high cost for little value. It is primarily the near UV purplish blue below 440 nm and the warm reds near IR that can not be reproduced.
1. Move in with your girlfriend. -1 key 2. Get a front door with keyless entry. -1 key 3. Get keyless entry at work. -1 key (this should be a nobrainer anyway) 4. and so on. 5. Drop the army knife. Get a stainless survival credit card.
You are now down to a car key, and you can keep a spare hidden under the car.
Open WRT has the potential to be fantastic, and used to be a stable release that worked on many platforms, the WRTGL in particular. That focus is lost, and instead it is to make it run on everything with a CPU.
I had used the old "White Russian" version for a long time on the standard WRT54GL (thats where the name OpenWRT comes from), and decided to upgrade, and I could not get it to even work on the reference platform as a basic AP.
I am no networking expert, but I have written a few network stacks, including wireless ones, so I am not completely green. The solution to get it to work came down to set up the whole development environment, and do it myself. Maybe that is the intention, but I just do not have the time to do this. I reverted back to the old "White Russian".
On another box the implementation booted, but lacked so much, that I reverted back to the factory image.
The main plume of the ash blows right across the Shetland islands pretty close to Iceland. The maps with artificial color show it to be black here, any yet, I see nothing. Blue skies, and starry nights. I have see ash from volcanoes where the sun turns red, so I know the scale, and here it is no ash. I have spoken with friends around Europe, and nobody has seen any ash plume. (except Iceland of course)
If anyone have seen any of this ash plume, please respond.
Funny how Google is a better front end to WSJ than WSJ itself. Maybe WSJ is shooting itslef in the foot with this policy. (Same with Mercury News and others)
Slashdot Mirror
For binary it is trivial:
Since it is an infinite, noncyclical number, there must be a zero some finite digits after any one. This must repeat ad infinity, so there must be an infinite numbe of zeroes.
Alternative would be that all digits after, say, digit 1,000,000,000 are ones. This could be written as a fraction, and would not be an irrational number.
You can make a similar argument for a decimal representation for pi. If you can come up with a good one, please reply.
an so are an infinite other digits in that number
Design parameters for a fusion reactor:
1. Shielding: 10m of water or similar as well as magnetic shielding
2. Energy density 10kW/m2
3. Politics: Not in my backyard
Conclusion:
Sun
1. Atmosphere and earth magnetic field: perfect
2. perfect almost anywhere
3. 150 million km away: perfect
Correct: You can pick any non-accelerated reference frame as center.
Earth is accelerated around the sun, and the sun is, to a lesser degree, accelerated around the milky way, as well as around the solar systems barycenter. It is anyway usually implied in the phrase "heliocentric reference" that a non-accelerated reference is picked.
Try to write simple laws of physics if your reference point is accelerated in a gravity well.
There has been a long and hostile fight between the Flat Earth Society (FES) and the Geocentrists. The conservative FES claim ancient heritage, and will not accept any newfangled ideas from the Geocentrists.
0. Buy a sacrificial computer. Keyboard and optical mouse can be $10 each.
1. Take a mouse and wire all buttons so they all are "left buttons".
2. Now open a browser to pbskids.org by default.
3. Also set your desktop up to accept a single click to "open" icons on desktop
4. And go from there
Your toddler will first have fun for days on pbskids. Will discover the desktop after that. You can sprinkle it with some icons of interest.
Can MIT come up with something else than "solar powered nano technology autonomous robot swarms". Is this the "Build a really cool solution, and then spend a decade looking for the problem it solves"
Would be nice to see some innovation there. Are they not supposed to have the skills and intelligence to think outside the box and go in new directions?
- a NON-autonomous robot
- a FOSSIL powered something
- some new MACRO technology.
where is that Æ again?
There is no problem to get a lens with better resolution than this. A fixed focus would certainly do just fine. At 750nm, and with n.a. ~1, a lens would need a diameter equal to the resolution. The diagonal resolution is sqrt(x^2+y^2) or 16146. So a stepped down lens should have d>12mm.
Secondly, if you look at a distant object, like a cell tower on a distant mountain, your eyes will resolve down to 20 seconds of arc (Sparrow's resolution limit). Your total field of view is 120x60 degrees (of couse not high resolution away from focus, but just to be able to make an image with same resolution as your eyes, and allow the eyes to wander the image), you will need: 120deg/20"=20kpix x 10kpix or 200Mpix.
I can not see how this is a lot, nor can I see how this can be hard for a simple lens to focus properly. I can see how a 10x optical zoom would be problematic, but well understood optics combined with a powerful DSP would fix that. Look at how the fuzzy images on Hubble (before the optics was fixed) was deconvolved with simple math.
I hate to say this as a former Tektronix engineer, but you seldom need a scope, and if you do it is typically application targeted and expensive. So the general purpose scope that all self respecting EE's used to have on the lab desk is a thing of the past.
All digital work either has debuggers or with FPGAs, Chipscope Pro or other. No scope needed. And if you really need to see how the eye diagram looks with your 10GB Ethernet, the best scope may be your receiver chip. Hard to find a 40GHz scope anyway.
I actually had a Tek 2440 300MHz Digital Storage for at least a decade, but used it less and less. Became more a educational thing to show kids how AC looked. All serial interfaces are running at muli GHz speed, and RF development is more in the 5.7GHz range and higher (802.11n) Not many scope sampling at 4x or more at those frequencies.
This setup works flawlessly in most settings.
Pros:
Works like a normal cellphone with dialing options etc.
Use your phone lists etc
Synchronizes with PC apps
Good range
Cons:
Have to preconfigure all hotspots.
Will not hop between hotspots during same conversation
Not intuitive network configuration. Phone jumps to 3G unless you dig down to the setting to force it to wifi only.
MIT's webpage on iconoclasm http://fuck-the-skull-of-jesus.mit.edu/ prominently displayed this image for years with little uproar.
That image is truly orders of magnitude more insulting than anything any cartoonists have come up with of mohammed.
Where is the parity here? Are all christians just quitet duds? Freedom of press does not mean you have to accept any insult with no comment.
Comments welcome.
A fiber through a neighborhood drops one 10Gb color per highrise. An edgerouter in the basement connects each of the 36 floors with Gb ethernet. Each floor has GigE routers to 30 apartments.
Actual cost of installation in a dense urban deployment:
1km of fiber between each highrise: $60k (This was a shared cost ditch)
10Gb uplink edgerouter $30k
1k ports Gig switches: $8k
In building infrastructure(wiring, power etc: $56k
total cost for last mile for dense urban deployment per 100Mb customer: $142
Montly depreciation with 5 year life on equipment and 20 year on infrastructure: $1.65
Provisioning, management, billing, email, DNS servers etc: $3.50
Marketing, DRM, security, law enforcement support, Lobby efforts: $5.50
Adjust these numbers for yourself, and there is no way you can get more than 3x even if you goldplate it all.
Wifi can keep cost down in less dense deployments.
It is remotely comparable. So maybe it will be 100x the cost for the last mile. That makes 100Mb for a few dollars.
The point is that bandwidth is not a limited resource like electricity and water with a natural based scarcity, and an economical model based on that.
The scarcity of bandwidth in the USA is mostly a marketing gimmick to be able to differentiate cost between customers to maximize profit to the detriment of the overall speed of access.
20Mb/s-50Mb/s is common for basic internet in much of the world outside the USA
The single mode, Non-Zero Dispersion Shifted fibers is of course optimized for DWDM. That means that a buyer can put at least 128 colors in the fiber, each with 10Gb/s. With 423 fibers in the bundle, that adds up to 0.5Pb/s.
With 10x oversubscription, this will supply 541 million homes with 100Mb/s broadband each.
That should cover all of the americas with 100 million is USA, 46 million in Brazil, and 12 million in Canada.
The cost for each household should be pennies.
Well, if you put all the water after each other, it would reach from here to the moon and back. maybe that helps us grasp such a big number...
Just add harddrives everywhere, and add them together in a virtual folder. Buy 2TB sata drives internally and USB external SATA docks. Mirror them pairwise if you like, but don't do RAID. Just adds more problems than is solves.
Lets say you have 4 computers that will take 6 SATA 1 ESATA and 2 PATA, and with a dual USB sata docking station, you will have 22TB/computer, so you will have 88TB total for all computers. Now add docking for your USB storage off your routers for maybe 90TB. Copy an entire 2TB HDD to a docked HDD, and store offline as many as you like. A bookshelf holds easily 1000TB.
Each computer runs a script to share the drives on startup, so they are all available easily.
If you like, you can add truecrypt to them all as well, and include this in your script.
The scriptS may look like this with Win: /quit /volume c:\lib\disk1\libM /letter M /cache y /beep /q /v c:\lib\disk2\libN /lN /b /q /v c:\lib\disk3\libO /lO /b /q /v c:\lib\disk5\libP /lP /b .....
MOUNT.CMD
"C:\Program Files\TrueCrypt\TrueCrypt"
"C:\Program Files\TrueCrypt\TrueCrypt"
"C:\Program Files\TrueCrypt\TrueCrypt"
"C:\Program Files\TrueCrypt\TrueCrypt"
SHARE.CMD .....
net share tvM=M:\lib
net share tvN=N:\lib
net share tvO=O:\lib
net share tvP=P:\lib
UNSHARE.CMD /delete ....
net share tvM
DISMOUNT.CMD /dismount /force /beep /quit
"C:\Program Files\TrueCrypt\TrueCrypt"
So why the lame photoshopped picture of a cockroach with a backpack? Is TFA unable to produce any documentation at all from images to data?
Quick terminology: Spectral color- Pure, single wavelength color, like a laser. Composite color- A combination of many spectral colors of different intensity.
To truly reproduce a color, each pixel should be able to not only make one spectral color, but a combination of all of them.
This would be very expensive, and fortunately, our eye have sensors only for Red 580 nm, Green 540nm, and Blue 440 nm (RGB), if we exclude the low light rods. We can therefore get away with RGB screens. There are slight errors. For example, assume each R-G-B pixel emits light matching the eyes R-G-B sensors peak sensitivity. Now, we can reproduce any light stimulation by exiting a linear combination of the three emitters. The eye however is sensitive from 380 nm to 740 nm, and can obviously not create the stimulation for neither 400 nm light, nor 700 nm, as your linear combination of only positive values will not cover these spectral colors (outside the gamut of the display). Take a picture of a prism spectrum or rainbow, and compare the original with what you see on the monitor, and you can see this.
So bottom line, RGB covers almost all colors, but adding emitters allows linear combination to cover more of the possible stimulation, but a high cost for little value. It is primarily the near UV purplish blue below 440 nm and the warm reds near IR that can not be reproduced.
1. Move in with your girlfriend. -1 key
2. Get a front door with keyless entry. -1 key
3. Get keyless entry at work. -1 key (this should be a nobrainer anyway)
4. and so on.
5. Drop the army knife. Get a stainless survival credit card.
You are now down to a car key, and you can keep a spare hidden under the car.
Open WRT has the potential to be fantastic, and used to be a stable release that worked on many platforms, the WRTGL in particular. That focus is lost, and instead it is to make it run on everything with a CPU.
I had used the old "White Russian" version for a long time on the standard WRT54GL (thats where the name OpenWRT comes from), and decided to upgrade, and I could not get it to even work on the reference platform as a basic AP.
I am no networking expert, but I have written a few network stacks, including wireless ones, so I am not completely green. The solution to get it to work came down to set up the whole development environment, and do it myself. Maybe that is the intention, but I just do not have the time to do this. I reverted back to the old "White Russian".
On another box the implementation booted, but lacked so much, that I reverted back to the factory image.
Since 1 horse power is 0.7kW, and a cow generate 2kW, 1 cow power of 1 k.p. = 3 h.p.
Either these cows are under extreme pressure, or this is some kind of super-cow.
The main plume of the ash blows right across the Shetland islands pretty close to Iceland. The maps with artificial color show it to be black here, any yet, I see nothing. Blue skies, and starry nights. I have see ash from volcanoes where the sun turns red, so I know the scale, and here it is no ash. I have spoken with friends around Europe, and nobody has seen any ash plume. (except Iceland of course)
If anyone have seen any of this ash plume, please respond.
Discovered that Cisco is using GPL software and not complying with neither disclosing it nor making it available. Good an clear documentation as well.
I was not able to find anyone interested at all.
If you link from /. it is paywalled. As always, if you link from Google, it is all free. Just paste WSJ headline into Google, and go:
http://www.google.com/search?q=H-P+Executives+Face+Bribery+Probes
First hit is the article in full.
Funny how Google is a better front end to WSJ than WSJ itself. Maybe WSJ is shooting itslef in the foot with this policy. (Same with Mercury News and others)