This is just a stupid idea. what happens when you merge onto a fast moving freeway at 55 MPH? You get REAR ENDED. If everyone is going 70, you had BETTER be going 70 +- 10% too.
Besides, if they're just recording the "max speed" I've seen that jump around if you go through an area of poor reception. The receiver thinks you've jumped from location A to location B really quickly, result, extremely short lived spike in speed of 20% or more. Sometimes when walking around, I look down and see that my max walking speed was actually 15 MPH... Not likely.
Unless they're using this to do something like fine customers whose sustained speed was over 90 MPH for 2 minutes or more, this is just a stupid way to get short term money, and lose long term customers.
Re:That makes no sense.
on
Flywheel UPS
·
· Score: 1
That's likely about the same amount as your desktop computer's fan. Of course your computer's fan isn't buried in the ground, but 55db isn't loud by any means. Most people's comfortable easy listening level on their radio is say around 65db @ 1 Meter.
Yes. And the base 2 log of this number is CONSIDERABLY more than even a rough generous estimate of the number of atoms in the universe. So one might in fact expect chez scheme to have trouble computing it..... because it takes far more bits than you could ever store to represent that number...
No, there are some problems that not even a Beowulf will help you with...
CMUCL is the Carnegie Mellon University Common Lisp compiler which you can find out more about at http://www.cons.org/cmucl
(unfortunately I seem to have found a bug with respect to read-sequence... so I can't run the "io-byte" test on CMUCL, but CLISP also works for that test, CLISP is a common lisp system available at clisp.sourceforge.net)
Here's what I've got so far... it does many of the basic tests....
btw: rot13 is a method of converting a string of text by rotating all characters by 13 positions... try a google search, or use "tr"
;(setf extensions:*bytes-consed-between-gcs* 30000000) ; use this with CMUCL to prevent it from doing GC too often on a sparc without the incremental GC
(defun ack (x y) (declare (type fixnum x) (type fixnum y)) (cond ((= x 0) (+ 1 y)) ((= y 0) (ack (- x 1) 1)) (t (ack (- x 1) (ack x (- y 1))))))
(defun io-str (file) "although read-sequence is probably better for speed, the more natural way to do this in lisp is with a read-line, assuming the file is line oriented." (with-open-file (f file) (do ((str (read-line f nil) (read-line f nil))) ((not str) nil) (write-line str))))
(defun io-byte (file) "here's the probably significantly faster io routine" (with-open-file (f file ) (if (not f) (format t "f isn't a good file~%")) (do* ((arr (make-string (* 1024 32)) arr) (pos (read-sequence arr f ) (read-sequence arr f))) (( (length s) n) (setf len (length s))) ; (format t "~S~%" (switcheroo (expander s)) ) )))))
I'm writing the equivalent routines in Common Lisp as we speak...
So far CMUCL has been quite impressive on our sparc here at work, but I have to get home and check out my linux box which is quite similar to the one used in the tests.
I'd like to point out that the only usefulness of this test is to show what speed one-off programs will run at in various languages.
This really totally ignores the features of each language, (for example they couldn't do a C based hash-table test even... DUH), and it is heavily biased toward a c/perl/awk style of programming.... pre-allocating arrays and filling them, multiple hash tables... The descriptions even give you an algorithm, rather than a problem to solve... so they pre-suppose a way to solve a given problem.
For example, most common lisp compilers, and all scheme compilers will optimize away the function calls in the ackermann example. Furthermore, it is much more natural in lisp to compute some of these results by the use of concatenate and map(-into)... Some compilers (esp functional languages) will prove you could never use the results of your little loop that has no side effects and returns no value, and simply optimize it away entirely !!!:-)
I'll give some results here tomorrow for a few of the tests... the easy ones that don't require me to download large datasets over a slow link... in CMUCL on a dual pII 233 w/ 128 MB of ram running linux.
A dining cryptographer network allows you to communicate data without anyone being able to determine where the data is coming from.
It does require sharing a lot of bandwidth.
However it seems like some Gnutella like clone with a dining cryptographer based protocol would be a very useful tool for free speech. Then you only need servers operating in LAND based countries.... Get enough of them and since no-one knows where the data is actually coming from, no-one can attack it...
It would help to have laws that made it hard for the local governments to attack a website participating in such a network based on the fact that there is absolutely no evidence that any given site is hosting the data.
While you might lose some servers in the less free countries, if you have randomized replication, and multiple countries involved, you'd have a very robust information service.
I suspect that there's got to be some database from which the words were taken (ie. more than a pen and spiral notebook). The database would have to be widely available to make it worthwhile, otherwise the database has to be distributed with the key and the ciphertext.
Furthermore, the sentences have correct (or close to correct) grammatical structure. So there's a constraint on how you form the sentences... More than just first or second order random selection of words would suggest...
let's assume that the image itself is not involved, because you couldn't create a jpeg image without a computer.
As someone has already pointed out though, while it's always harder to decrypt something where the algorithm is unknown, a real cypher challenge would release the algorithm as well, and see whether the attacker can either decode the message without the key, or guess/derive the key from the message and any other known quantities (ie. plaintext fragments etc)...
Re:How do we discover when the problem is solved?
on
Solving Chess?
·
· Score: 1
There's already an RFC for a new Infinite Monkey Protocol...
How about a 4th option, get some good text books, maybe a little caffeine if you can't stay up, and read them, do the problems yourself, and study on your own time, especially problems you think are related to your job at first, because that will be the most motivating..
I can recommend Structure and Interpretation of Computer Programs, or Scheme and the Art of Programming as good basic computing science texts.
Once you've seen various kinds of abstraction (ie. in the above books) then you might want to look for books on computational complexity, and automata theory (this is a major part of the foundation of theoretical computer science). They also are very relevent areas for anyone who works with text programming (ie. automata were largely developed as a theoretical foundation for computer languages, and HTML and soforth are just kinds of languages)
You'll also want an algorithms book. Cormen Leiserson and Rivest is one good basic book. Of course there's always "the Bible" (ie. Knuth's The Art of Computer Programming vols. 1-3 and 4 if/when he's done with it.)
Yes, it's easier to learn with someone to help teach you, but it's also easier to learn something from a book, than the hard way (ie. doing it wrong over and over again at work;-) )
If you want more suggestions...
NOTE: why doesn't slashdot have a "private message" feature so people who want to set up out of band communications can??
Just because you don't know any apps written in Lisp doesn't mean it's a bad language.
Also, just because one example of a Lisp program (Yahoo store) doesn't work a well as you'd like doesn't mean Lisp is a bad language. Think of all the shitty C/C++ or Perl code that sucks even harder.... By your metric you've just eliminated those languages from any kind of competition.
Autocad is (or was) largely written in Lisp, and Nichimen graphics http://www.nichimen.com sells one of the more popular 3D graphics engines for artists of console games, all written in LISP that's right, high end 3D graphics in Lisp...
Face it, most people who think Lisp is just for ivory towers, or is too slow, or blablabla don't know anything about the language...
For Scheme see : Structure and Interpretation of Computer Programs, and Scheme and the Art of Programming...
For Common Lisp (the industrial grade standardized lisp) see Paul Grahams books "On Lisp" and "ANSI Common Lisp"
Download one of the free Lisps: clisp.sourceforge.net or www.cons.org/cmucl...
This benchmark measures performance with multiple dimensions (ie. performance with respect to the size of the problem, the number of processors available, and the size of the cache) It's numerical intensive so it gives a good idea of what scientific programming performance would be like, not say graphics performance...
Scheme (a dialect of Lisp) is a great language for Lego Mindstorm. The syntax is simple, it could be byte-code compiled easily. There are tons of implementations to start with. It's often used to teach introductory programming, and some of the best CS texts are written towards Scheme.
"Structure and Interpretation of Computer Programs" and "Scheme and the Art of Programming" for example.
The biggest problem would be the garbage collector, but that problem is not a huge one as incremental real time garbage collectors exist, and the interrupt handlers would all be written in something lower level anyway.
Anyone want to take a stab at making "Mindthtorm" (lego LISP)??
Even if they aren't real options (which you don't need any cash to exercise), if it's, for example, an employee purchase plan which just allows you to buy stock itself at a discount you can raise the money by selling the stock short...
Sell 100 shares of stock at $20 on the open market. Take $1500 of the $2000 in cash you have from the short-sale, give to company in exchange for 100 shares at the discount price of, say $15. Now you owe someone 100 shares in one account, but in some other account you have these shares you owe, plus you have $500 in cash.
Make $5/share without putting out ANY serious cash.
The only problem with this is if the stock shoots through the roof and you have a fixed "waiting period" before you can actually take delivery of the shares you purchased at the discount price... Because the broker will require you to keep a "margin" (basically some percentage of the amount that the stock you owe is worth)
If your company were large, you could buy open-market call options which would cover your risk, but eat into your profits.
But don't pass up an opportunity to make GUARANTEED MONEY because you don't have the cash. If it's really guaranteed the way this sort of thing is, there's no risk to someone who lends you money and they get a cut. Banks will do this for you, so at the very least, you could borrow money from a bank to cover the margin.
VPN refers to SLIP connections because the way it works is approximately this:
(note: see the freshmeat.net entry for vpnd which will get you to the homepage if you want more info)
vpnd gets itself a pty (pseudo terminal). Then it creates a SLIP interface on that terminal (ie. it is able to write data to the terminal and have it enter the linux routing mechanism as an IP packet)
Now when it receives packets on the slip interface (ie. a packet headed for the virtual network) it encrypts it, and sends the data inside an IP packet down a regular interface (ie. eth0 or your ppp0 connection) when the remote daemon receives this packet on its eth0 interface, it extracts the data, decrypts it and sends it out the SLIP connection on the opposite end where it is routed by the remote kernel to whatever program is listening on a remote SLIP interface socket.
the same thing happens when you go the other way..
so the SLIP part is actually VIRTUAL and does not require a modem or a serial line (though I believe that it can be used over a modem/serial line.
So how is any of this abuse?? You're specifically ASSIGNED those rights to do all of that.
That's what the authors WANT you to do with it.
The authors only get money from the sale of the hard copy books that their publisher sells, and they KNOW that, and don't expect royalties from other people.
since when do we advocate "free software, so long as you don't actually USE any of that pesky freedom"
Ionizing radiation causes the loss or gain of electrons, creating a particle (like one of the carbons or oxygens in your body) which has an unpaired electron (called a free radical) this causes rapid chemical reactions which can damage cells. These types of radiation are KNOWN TO BE BAD FOR YOU.
Ionizing radiation examples are: X rays, Gamma Rays, Alpha particle, Beta particles, and possibly Ultra Violet light.
Non-ionizing radiation does not cause unpaired electrons. Examples of non-ionizing radiation are: Visible light, infra-red light, microwaves, and other radio waves.
Various of these types of radiation are known to be bad for you (don't stick your head in the microwave) but the mechanisms, dosage effects and chronic vs. acute exposure effects are not well known. They often cause problems because of molecular resonance, which causes heating of molecules (ie. cooking food). But there may be other effects which are not well studied.
Particle vs. Electromag: Beta particles are high energy electrons, they are ionizing radiation, can be blocked by a few millimeters of glass/plastic/metal and are not an issue from computer monitors (get a geiger counter and check it if you want).
Alpha particles are helium nuclei without any electrons. They can be found streaming around inside most smoke detectors. They are VERY bad for you, but they travel only a few centimeters through air, and can be blocked by paper or plastic or clothing. They are a big problem if you get them inside of you (ie. swallow an alpha source).
neutrons are a problem for those who are exposed, but most people aren't. so don't worry about them.
Everything else is electromagnetic (transferred by massless photons) and most of the "radiation" from computer monitors that people worry about is the MAGNETIC fields of very low frequency (ie 60Hz). There's nothing much you can do to shield yourself from these, though monitors with built in circuits can try to cancel out the fields.
This study does not deal with Ultra Low Frequency magnetic radiation, but rather with Gamma Rays, which are an extremely high frequency ionizing electro-magnetic radiation above the frequency of X-Rays. You'd expect any anti-oxidant chemical to potentially help vs. ionizing radiation and perhaps caffeine has some anti-oxidant properties?? I dunno
Disclaimer: I am not a health Physicist, just a guy who has a lot of info shoved into his head.
Certification is equivalent, in my mind, to saying "let us do the thinking for you".
The problem with this "community" is that it's trying to act like a company, and trying to provide some sort of "united front" in some sort of attempt to "take over the world"
bullshit.
write software. Use the software. Enjoy the fact that you have quality software, and the freedom to read, modify, change, and redistribute it.
but stop pretending to speak for everyone else who also enjoys those freedoms.
We're intelligent people who can decide for and amongst ourselves whether we want to participate in software development under certain various licensing terms.
We simply do not need some certification from on high. This is "cathedralic" to use ESR's own analogies.
It's not a doughnut, think more like a trumpet facing you. Stuff is on the inside wall of big (infinite) trumpet that you can zoom into or out of..
I kinda like the inverse better, the fishey view. In that model you're on the surface of a hemisphere (or some such thing which recedes away from you as you move outward), and you can push stuff outside to the edges where it gets compressed, but stuff in the middle is mostly undistorted. It's a better model for high res displays like a desktop, where you'd like to see things in the center in full nice undistorted quality, but be able to push things out to the edges and have them be visible but unimportant.
Slashdot Mirror
This is just a stupid idea. what happens when you merge onto a fast moving freeway at 55 MPH? You get REAR ENDED. If everyone is going 70, you had BETTER be going 70 +- 10% too.
Besides, if they're just recording the "max speed" I've seen that jump around if you go through an area of poor reception. The receiver thinks you've jumped from location A to location B really quickly, result, extremely short lived spike in speed of 20% or more. Sometimes when walking around, I look down and see that my max walking speed was actually 15 MPH... Not likely.
Unless they're using this to do something like fine customers whose sustained speed was over 90 MPH for 2 minutes or more, this is just a stupid way to get short term money, and lose long term customers.
That's likely about the same amount as your desktop computer's fan. Of course your computer's fan isn't buried in the ground, but 55db isn't loud by any means. Most people's comfortable easy listening level on their radio is say around 65db @ 1 Meter.
Yes. And the base 2 log of this number is CONSIDERABLY more than even a rough generous estimate of the number of atoms in the universe. So one might in fact expect chez scheme to have trouble computing it..... because it takes far more bits than you could ever store to represent that number...
h tml
No, there are some problems that not even a Beowulf will help you with...
for more info on the ackermann function go to:
http://mathworld.wolfram.com/AckermannFunction.
or google it up...
CMUCL is the Carnegie Mellon University Common Lisp compiler which you can find out more about at http://www.cons.org/cmucl
:element-type 'integer))
:element-type 'simple-string
:initial-element "abcdefghij")) :element-type 'simple-string
:initial-element "")))
(unfortunately I seem to have found a bug with respect to read-sequence... so I can't run the "io-byte" test on CMUCL, but CLISP also works for that test, CLISP is a common lisp system available at clisp.sourceforge.net)
Here's what I've got so far... it does many of the basic tests....
btw: rot13 is a method of converting a string of text by rotating all characters by 13 positions... try a google search, or use "tr"
;(setf extensions:*bytes-consed-between-gcs* 30000000) ; use this with CMUCL to prevent it from doing GC too often on a sparc without the incremental GC
(defun ack (x y)
(declare (type fixnum x) (type fixnum y))
(cond ((= x 0) (+ 1 y))
((= y 0) (ack (- x 1) 1))
(t (ack (- x 1) (ack x (- y 1))))))
(defun array-int (n)
(declare (type integer n))
(let ((a1 (make-array n
(a2 nil))
(dotimes (x n nil)
(setf (aref a1 x) x))
(setf a2 (copy-seq a1))
nil))
(defun array-str (n)
(let* ((a (make-array n
(b (make-array n
(map-into b #'(lambda (x) (concatenate 'simple-string x " ")) a)
nil))
;(defun assoc-build (file)
; (make-hash-table (
;)))
(defun inc-loop (n)
(declare (type (unsigned-byte 32) n))
(do ((i 0 (1+ i)))
((= i n) nil)
(declare (type (unsigned-byte 32) i))))
(defun io-str (file)
"although read-sequence is probably better for speed, the more
natural way to do this in lisp is with a read-line, assuming the file
is line oriented."
(with-open-file
(f file)
(do ((str (read-line f nil) (read-line f nil)))
((not str) nil)
(write-line str))))
(defun io-byte (file)
"here's the probably significantly faster io routine"
(with-open-file
(f file )
(if (not f) (format t "f isn't a good file~%"))
(do* ((arr (make-string (* 1024 32)) arr)
(pos (read-sequence arr f ) (read-sequence arr f)))
(( (length s) n) (setf len (length s)))
; (format t "~S~%" (switcheroo (expander s)) )
)))))
(defun sum-test (filename)
(with-open-file
(input filename)
(do* ((x (read input nil ) (read input nil))
(sum x (if x (+ sum x) sum)))
((not x) sum)
(declare (type (or float nil) x)
(type float sum)))))
I'm writing the equivalent routines in Common Lisp as we speak...
:-)
So far CMUCL has been quite impressive on our sparc here at work, but I have to get home and check out my linux box which is quite similar to the one used in the tests.
I'd like to point out that the only usefulness of this test is to show what speed one-off programs will run at in various languages.
This really totally ignores the features of each language, (for example they couldn't do a C based hash-table test even... DUH), and it is heavily biased toward a c/perl/awk style of programming.... pre-allocating arrays and filling them, multiple hash tables... The descriptions even give you an algorithm, rather than a problem to solve... so they pre-suppose a way to solve a given problem.
For example, most common lisp compilers, and all scheme compilers will optimize away the function calls in the ackermann example. Furthermore, it is much more natural in lisp to compute some of these results by the use of concatenate and map(-into)... Some compilers (esp functional languages) will prove you could never use the results of your little loop that has no side effects and returns no value, and simply optimize it away entirely !!!
I'll give some results here tomorrow for a few of the tests... the easy ones that don't require me to download large datasets over a slow link... in CMUCL on a dual pII 233 w/ 128 MB of ram running linux.
Last time I looked, BeFS didn't journal user data.
That was a while back, anyone know what the current state is?
A dining cryptographer network allows you to communicate data without anyone being able to determine where the data is coming from.
It does require sharing a lot of bandwidth.
However it seems like some Gnutella like clone with a dining cryptographer based protocol would be a very useful tool for free speech. Then you only need servers operating in LAND based countries.... Get enough of them and since no-one knows where the data is actually coming from, no-one can attack it...
It would help to have laws that made it hard for the local governments to attack a website participating in such a network based on the fact that there is absolutely no evidence that any given site is hosting the data.
While you might lose some servers in the less free countries, if you have randomized replication, and multiple countries involved, you'd have a very robust information service.
Me: i'm thinking of going scuba diving in monterey this weekend
AliceBot: What is this "thinking"?
The defense rests your honor.
I suspect that there's got to be some database from which the words were taken (ie. more than a pen and spiral notebook). The database would have to be widely available to make it worthwhile, otherwise the database has to be distributed with the key and the ciphertext.
Furthermore, the sentences have correct (or close to correct) grammatical structure. So there's a constraint on how you form the sentences... More than just first or second order random selection of words would suggest...
let's assume that the image itself is not involved, because you couldn't create a jpeg image without a computer.
As someone has already pointed out though, while it's always harder to decrypt something where the algorithm is unknown, a real cypher challenge would release the algorithm as well, and see whether the attacker can either decode the message without the key, or guess/derive the key from the message and any other known quantities (ie. plaintext fragments etc)...
There's already an RFC for a new Infinite Monkey Protocol...
ftp://ftp.isi.edu/in-notes/rfc2795.txt
How about a 4th option, get some good text books, maybe a little caffeine if you can't stay up, and read them, do the problems yourself, and study on your own time, especially problems you think are related to your job at first, because that will be the most motivating..
;-) )
...
I can recommend Structure and Interpretation of Computer Programs, or Scheme and the Art of Programming as good basic computing science texts.
Once you've seen various kinds of abstraction (ie. in the above books) then you might want to look for books on computational complexity, and automata theory (this is a major part of the foundation of theoretical computer science). They also are very relevent areas for anyone who works with text programming (ie. automata were largely developed as a theoretical foundation for computer languages, and HTML and soforth are just kinds of languages)
You'll also want an algorithms book. Cormen Leiserson and Rivest is one good basic book. Of course there's always "the Bible" (ie. Knuth's The Art of Computer Programming vols. 1-3 and 4 if/when he's done with it.)
Yes, it's easier to learn with someone to help teach you, but it's also easier to learn something from a book, than the hard way (ie. doing it wrong over and over again at work
If you want more suggestions
NOTE: why doesn't slashdot have a "private message" feature so people who want to set up out of band communications can??
Just because you don't know any apps written in Lisp doesn't mean it's a bad language.
Also, just because one example of a Lisp program (Yahoo store) doesn't work a well as you'd like doesn't mean Lisp is a bad language. Think of all the shitty C/C++ or Perl code that sucks even harder.... By your metric you've just eliminated those languages from any kind of competition.
Autocad is (or was) largely written in Lisp, and Nichimen graphics http://www.nichimen.com sells one of the more popular 3D graphics engines for artists of console games, all written in LISP that's right, high end 3D graphics in Lisp...
Face it, most people who think Lisp is just for ivory towers, or is too slow, or blablabla don't know anything about the language...
For Scheme see : Structure and Interpretation of Computer Programs, and Scheme and the Art of Programming...
For Common Lisp (the industrial grade standardized lisp) see Paul Grahams books "On Lisp" and "ANSI Common Lisp"
Download one of the free Lisps: clisp.sourceforge.net or www.cons.org/cmucl...
Read "comp.lang.lisp" archives...
http://www.scl.ameslab.gov/HINT/
This benchmark measures performance with multiple dimensions (ie. performance with respect to the size of the problem, the number of processors available, and the size of the cache) It's numerical intensive so it gives a good idea of what scientific programming performance would be like, not say graphics performance...
Scheme (a dialect of Lisp) is a great language for Lego Mindstorm. The syntax is simple, it could be byte-code compiled easily. There are tons of implementations to start with. It's often used to teach introductory programming, and some of the best CS texts are written towards Scheme.
"Structure and Interpretation of Computer Programs" and "Scheme and the Art of Programming" for example.
The biggest problem would be the garbage collector, but that problem is not a huge one as incremental real time garbage collectors exist, and the interrupt handlers would all be written in something lower level anyway.
Anyone want to take a stab at making "Mindthtorm" (lego LISP)??
Even if they aren't real options (which you don't need any cash to exercise), if it's, for example, an employee purchase plan which just allows you to buy stock itself at a discount you can raise the money by selling the stock short...
Sell 100 shares of stock at $20 on the open market. Take $1500 of the $2000 in cash you have from the short-sale, give to company in exchange for 100 shares at the discount price of, say $15.
Now you owe someone 100 shares in one account, but in some other account you have these shares you owe, plus you have $500 in cash.
Make $5/share without putting out ANY serious cash.
The only problem with this is if the stock shoots through the roof and you have a fixed "waiting period" before you can actually take delivery of the shares you purchased at the discount price... Because the broker will require you to keep a "margin" (basically some percentage of the amount that the stock you owe is worth)
If your company were large, you could buy open-market call options which would cover your risk, but eat into your profits.
But don't pass up an opportunity to make GUARANTEED MONEY because you don't have the cash. If it's really guaranteed the way this sort of thing is, there's no risk to someone who lends you money and they get a cut. Banks will do this for you, so at the very least, you could borrow money from a bank to cover the margin.
VPN refers to SLIP connections because the way it works is approximately this:
(note: see the freshmeat.net entry for vpnd which will get you to the homepage if you want more info)
vpnd gets itself a pty (pseudo terminal). Then it creates a SLIP interface on that terminal (ie. it is able to write data to the terminal and have it enter the linux routing mechanism as an IP packet)
Now when it receives packets on the slip interface (ie. a packet headed for the virtual network) it encrypts it, and sends the data inside an IP packet down a regular interface (ie. eth0 or your ppp0 connection) when the remote daemon receives this packet on its eth0 interface, it extracts the data, decrypts it and sends it out the SLIP connection on the opposite end where it is routed by the remote kernel to whatever program is listening on a remote SLIP interface socket.
the same thing happens when you go the other way..
so the SLIP part is actually VIRTUAL and does not require a modem or a serial line (though I believe that it can be used over a modem/serial line.
So how is any of this abuse?? You're specifically ASSIGNED those rights to do all of that.
That's what the authors WANT you to do with it.
The authors only get money from the sale of the hard copy books that their publisher sells, and they KNOW that, and don't expect royalties from other people.
since when do we advocate "free software, so long as you don't actually USE any of that pesky freedom"
There seems to be a bit of confusion with regards to the types of radiation that exist, and where they come from
types of radiation:
1) Ionizing
2) Non-ionizing
3) Particle
4) Electromagnetic
Ionizing radiation causes the loss or gain of electrons, creating a particle (like one of the carbons or oxygens in your body) which has an unpaired electron (called a free radical) this causes rapid chemical reactions which can damage cells. These types of radiation are KNOWN TO BE BAD FOR YOU.
Ionizing radiation examples are: X rays, Gamma Rays, Alpha particle, Beta particles, and possibly Ultra Violet light.
Non-ionizing radiation does not cause unpaired electrons. Examples of non-ionizing radiation are:
Visible light, infra-red light, microwaves, and other radio waves.
Various of these types of radiation are known to be bad for you (don't stick your head in the microwave) but the mechanisms, dosage effects and chronic vs. acute exposure effects are not well known. They often cause problems because of molecular resonance, which causes heating of molecules (ie. cooking food). But there may be other effects which are not well studied.
Particle vs. Electromag: Beta particles are high energy electrons, they are ionizing radiation, can be blocked by a few millimeters of glass/plastic/metal and are not an issue from computer monitors (get a geiger counter and check it if you want).
Alpha particles are helium nuclei without any electrons. They can be found streaming around inside most smoke detectors. They are VERY bad for you, but they travel only a few centimeters through air, and can be blocked by paper or plastic or clothing. They are a big problem if you get them inside of you (ie. swallow an alpha source).
neutrons are a problem for those who are exposed, but most people aren't. so don't worry about them.
Everything else is electromagnetic (transferred by massless photons) and most of the "radiation" from computer monitors that people worry about is the MAGNETIC fields of very low frequency (ie 60Hz). There's nothing much you can do to shield yourself from these, though monitors with built in circuits can try to cancel out the fields.
This study does not deal with Ultra Low Frequency magnetic radiation, but rather with Gamma Rays, which are an extremely high frequency ionizing electro-magnetic radiation above the frequency of X-Rays. You'd expect any anti-oxidant chemical to potentially help vs. ionizing radiation and perhaps caffeine has some anti-oxidant properties?? I dunno
Disclaimer: I am not a health Physicist, just a guy who has a lot of info shoved into his head.
Certification is equivalent, in my mind, to saying "let us do the thinking for you".
The problem with this "community" is that it's trying to act like a company, and trying to provide some sort of "united front" in some sort of attempt to "take over the world"
bullshit.
write software. Use the software. Enjoy the fact that you have quality software, and the freedom to read, modify, change, and redistribute it.
but stop pretending to speak for everyone else who also enjoys those freedoms.
We're intelligent people who can decide for and amongst ourselves whether we want to participate in software development under certain various licensing terms.
We simply do not need some certification from on high. This is "cathedralic" to use ESR's own analogies.
It's not a doughnut, think more like a trumpet facing you. Stuff is on the inside wall of big (infinite) trumpet that you can zoom into or out of..
I kinda like the inverse better, the fishey view. In that model you're on the surface of a hemisphere (or some such thing which recedes away from you as you move outward), and you can push stuff outside to the edges where it gets compressed, but stuff in the middle is mostly undistorted. It's a better model for high res displays like a desktop, where you'd like to see things in the center in full nice undistorted quality, but be able to push things out to the edges and have them be visible but unimportant.