This experiment can be done with waves in water and a dam with small openings, with similar result. I think "science" is a slightly misinterpreted by the author of the article.
Particularly amusing is his misunderstanding that this will only work with _red_ light... It works with any light, but due to varied refraction of the different wavelengths of blended/white light, lasers should illustrate the interference much clearer (since they emit only a single wavelength).
The mindboggling thing about the double-slit experinment is, that light of a certain wavelength can only be broken into pieces with a finite amount of energy, you cannot split a photon. And yet a single photon (an indivisable amount of electro-magcanetic energy at a certain wavelength) interfere with itself, generating an interference pattern when passing though a "double slit".
There are several external influences that decides if a program is "optimized enough". These influences are often local to spcific users, and thus cannot be tested easily by benchmarking.
Now, I'm all for the optimizations based on profiling and don't optimize utill the need is there. I'm just trying to say that it's pretty complicated trying to figure out when the need is there in a more general setting.
Repetition: (probably the most common problem) It's ok if a task takes long if it's not done too often. If you have to repeat the task many times, performance of that task becomes as many times more critical as you repeat it.
Multitasking/Parralelism: Other tasks would like to get to the processor too. If your program is just fast enough to do it's job, it's too slow to do it while notepad is running along side it. Now, I run colinux alongside on my windows desktop, so performance of "other" programs does matter. Also, execessive memory usage slows down all the other programs on my machine down too (hi there mozilla:)
Power: My laptop can scale the cpu-frequency and use that to save power, so i can sit out in the sun programming longer. If a program spends too many cycles on a job, i'm gonna have to go inside sooner. and this programs BEGS for optimizations from my point of view.
Many more of these side-effects of optimization occur, so don't think you can easily (or exactly) evalute when certain code is efficient enough for others, only for yourself.
This also shows a fundamental advantage in the OpenSource model. I can run a profiler myself and try to perform some optimization, instead of selecting not to use the program (or use another).
he is now suing Google, Yahoo (which used Google as its search engine at the time), AOL (for using Google to enhance its search results) and Time Warner (because they're the same company as AOL) for libel
Yeah, right, that information will be:
'alarming, false, misleading and injurious'
Most likely it's gonna be true with such a zealeous litigation based on search results.
I've setup SVN to replace CVS at 2 companies now, and they're both very happy.
Here's what I say:
* Just leave the old CVS data where it is, you can always use your old methods to get hold of it...
* The most precise record of how the past (which was recorded in CVS) was is in...... CVS
* When starting a new project --- use subversion, and move the components required for that project from CVS to subversion --- minimum work, maximum benefit
I just wish my country (Denmark) had a patent-law that could help business' like the american patent-law.
It's clear to me that this patent helps "stimulate the inventive genius" and makes it possible for companies to do do research that would never be done without the patent-law, thus helping the community at large.
It's not _directly_ anti-american, but "America" gives a shit about global evironment (check international negotiations) and thus enviromentalism becomes opposed to "America", which "America" usually interpret as Anti-American.
("America" meaning the official actions of the USA)
The problem with biometrics is, that they are symmetric in nature, this allows anyone you ever authenticated against to effectively take over your id.
Of course, that's harder with biometrics than with passwords, but why take the chance?
I'd much rather use challenge based authentication, and the use my own, bio-metric and password protected, device to store the information (usually private keys) for answering the challenges.
As i understand, you DO need a web-broser to interface with the box, so why not just connect the PC directly to the stereo?
If you are worried that will feed the sound when you're watching pr0n into the stereo add one more soundcard.
I got a P3-700 sitting under my video, hooked to the stereo and TV. It does PVR, mp3-play, routing, dhcp server, NAT, print-deamon, wireless access point,...
The whole thing set me back around $190, one year ago.
I severely doubt that any positive effects on memory and reaction times is _caused_ by wireless phones.
Actually (at least in europe), there are demographic differences between cell-phoners and the rest.
One difference being that younger people are over-overrepresented in the set of cell-phoners...accidentaly, young persons are documented to have better memory and reaction time than average people:)
A: "You stole my watch" B: "No, i bought that" A: "No, you stole mine!" B: "Okay, show me somthing that identifies your watch" A: "Can't do that... it's a secret watch!" B: "uuuhhmmm, if you really think it's your watch, I already GOT the secret!"
This places everyone who ever accepts a signature from anther person in a position to man-in-the-middle attack that person, pretty clever.... DUH!
What is needed in computer-systems is zero-knowledge proofs, or at least a challenge/response method influenced by both parties.
The real use for bio-metrics is to verify relations between identity and private-key possesion in a trusted 3rd-patry setting, not for authentication itself.
As a defensive step, I have e-mailed SCO, stating that I am cercerned about their claims that I am infringing on their property, asking for an indication of exactly how I'm infringing on their property, asking (politely!) that they show me the offending code.
Of course, I am not in a position to license Linux from SCO without that information. So I'm just waiting for them to mail me back:)
(BTW: I live in a country where the LAW is actually reasonable, I don't think SCO could convince any Danish judge with their outrageous claims:)
googling for "rechageable faq" yeilds lots of hist, for example: http://www.buchmann.ca/faq.asp. But here's some advice anyway:
* Basically, you want rechargeable batteries whenever chances are your thingy with the batteries will not get nicked, since that defeats the investment in rechageables.
* Choose batteries with the highest mAh rating, which says how much energy you can draw from the battery.
* Choose NiMH for camera's and other power-suckers.
* Buy a charger with seperate charging cicuits for each battery.
* NiCD batteries have some problems with memory, but good chargers with "decharge" can usually fix that. (actually, NiCD batteries are hard to come by these days)
patented non-linear data mapping technology, called VME (Virtual Matrix Encryption)
From the patent:
A data security method and apparatus that provides an exceptional degree of security at low computational cost. The data security arrangement differs from known data security measures in several fundamental aspects. Most notably, the content of the message is not sent with the encrypted data. Rather, the encrypted data consists of pointers to locations within a virtual matrix, a large (arbitrarily large), continuously-changing array of values. The encryption technique is therefore referred to as Virtual Matrix Encryption. Furthermore, the data security arrangement uses a very large key of one million bits or more which creates a level of security much higher than any other existing method. The key is not transferred but is instead created from a file of any size that is available on both a computer used to send a secure message and a computer used to receive a secure message. The term Virtual Key Cryptographic as used herein to refer to techniques in which a key is recreated at a remote location from an electronic file without any transmission of the key itself. The file may be a system file, a file downloaded from the Internet, etc. A smaller, transaction-specific key, e.g., a 2,048 bit key, is sent end-to-end and is used in conjunction with the very large key to avoid a security hazard in instances where the same file is used repeatedly to create the very large key.
So, it would _seem_ a bit like:
1. build matrix:
A B C
D E F
G H I
2. to cipher up the letter F which is at row 2, col 3 send (2,3).
3. mutate matrix, goto 2
So the real "crypto" lies in the mutation of the matrix... how that is done is not described... maybe it's just x-or'ed onto itself or whatnot.
The way the key is found has nothing to do with the value of the crypto, so don't even begin to critisize how easy it must be for an attacker to guess which file is being used as key.
The trust model of X509 Cerifitates is fundamentally flawed, in that it does not mimic the trust model applied in "the real world", but an authoritarian one.
In the real world, you trust someone if enough "peers" that you trust trust that someone, and probably a bit less:)
So if Microsoft ultimately wins a major share of the market, then using a Microsoft DRM client may be necessary in order to buy music.
The solution to that problem is already being effectively implemented in the Computer Programs area: generate a vast industri, that doesn't pay taxes or royalties, providing people with unprotected material.
The difference between encoding and encryption is, that encryption is parametrised with a key.
ROT-13,... is an encoding
ROT, is an encryption, since it's parametrised.
<fun>Some encodings are not surjective, which makes them kind of in a grey area, since the interpretation of those elements which are not the image of any input is basically a key... This nicely explains why different text-encodings are so god-damn cryptic to get working in applications:)</fun>
I would venture the hypothesis, that any strong encryption has a mathematical model. Otherwise reasoning about the encryption would be rather hard.
The important thing, is to not provide Free (as in beer) training to one OS vendor, radically unbalancing the competition in the OS market.
The danish goverment spend millions of dollars each year on "teaching the people to use IT", which basically boils down to giving users a training course on all M$-OS and Office products.
I suggest having a mix of OS'es, so that the students have different experiences and learn from comparing those.
I myself is a student at DAIMI where machines with SunOS, HPUX (well not that many anymore) IRIX, GNU/Linux and Windows (Using vmware), and yes it's a pain with the differences between computers but:
1. You can just select to use the same OS every time 2. You learn a lot by seeing different solutions to the same problem
Here in Denmark, nobody, except the police, carry guns that are now for sport. Police guns are weak, and VERY RARELY (once every few years) kill.
1-10 people are killed each year by guns, fom a population of 5.5million.
Draw your own conclusions....
If Americans have problems protecting themselves from "villans" with guns, maybe they should start thinking a bit about how those villains got their guns.
Slashdot Mirror
Anyone who can guess what fluctuations in background radiation, or gieger count-intervals i used to generate my OTP is worthy of the predicate "GOD"
This experiment can be done with waves in water and a dam with small openings, with similar result. I think "science" is a slightly misinterpreted by the author of the article.
Particularly amusing is his misunderstanding that this will only work with _red_ light... It works with any light, but due to varied refraction of the different wavelengths of blended/white light, lasers should illustrate the interference much clearer (since they emit only a single wavelength).
The mindboggling thing about the double-slit experinment is, that light of a certain wavelength can only be broken into pieces with a finite amount of energy, you cannot split a photon. And yet a single photon (an indivisable amount of electro-magcanetic energy at a certain wavelength) interfere with itself, generating an interference pattern when passing though a "double slit".
There are several external influences that decides if a program is "optimized enough". These influences are often local to spcific users, and thus cannot be tested easily by benchmarking.
:)
Now, I'm all for the optimizations based on profiling and don't optimize utill the need is there. I'm just trying to say that it's pretty complicated trying to figure out when the need is there in a more general setting.
Repetition: (probably the most common problem) It's ok if a task takes long if it's not done too often. If you have to repeat the task many times, performance of that task becomes as many times more critical as you repeat it.
Multitasking/Parralelism: Other tasks would like to get to the processor too. If your program is just fast enough to do it's job, it's too slow to do it while notepad is running along side it. Now, I run colinux alongside on my windows desktop, so performance of "other" programs does matter. Also, execessive memory usage slows down all the other programs on my machine down too (hi there mozilla
Power: My laptop can scale the cpu-frequency and use that to save power, so i can sit out in the sun programming longer. If a program spends too many cycles on a job, i'm gonna have to go inside sooner. and this programs BEGS for optimizations from my point of view.
Many more of these side-effects of optimization occur, so don't think you can easily (or exactly) evalute when certain code is efficient enough for others, only for yourself.
This also shows a fundamental advantage in the OpenSource model. I can run a profiler myself and try to perform some optimization, instead of selecting not to use the program (or use another).
Yeah, right, that information will be:
Most likely it's gonna be true with such a zealeous litigation based on search results.
I've setup SVN to replace CVS at 2 companies now, and they're both very happy.
...... CVS
Here's what I say:
* Just leave the old CVS data where it is, you can always use your old methods to get hold of it...
* The most precise record of how the past (which was recorded in CVS) was is in
* When starting a new project --- use subversion, and move the components required for that project from CVS to subversion --- minimum work, maximum benefit
I just wish my country (Denmark) had a patent-law that could help business' like the american patent-law.
It's clear to me that this patent helps "stimulate the inventive genius" and makes it possible for companies to do do research that would never be done without the patent-law, thus helping the community at large.
It's not _directly_ anti-american, but "America" gives a shit about global evironment (check international negotiations) and thus enviromentalism becomes opposed to "America", which "America" usually interpret as Anti-American.
("America" meaning the official actions of the USA)
predicting evolution is a loosong game
BTW: if so many species really do go extinct, it'll be fun to see what replaces their role
Now, speculations about a deceased mans possible viewpoint on the digital technology is about the most important input any discussion can get
The problem with biometrics is, that they are symmetric in nature, this allows anyone you ever authenticated against to effectively take over your id.
Of course, that's harder with biometrics than with passwords, but why take the chance?
I'd much rather use challenge based authentication, and the use my own, bio-metric and password protected, device to store the information (usually private keys) for answering the challenges.
As i understand, you DO need a web-broser to interface with the box, so why not just connect the PC directly to the stereo?
...
If you are worried that will feed the sound when you're watching pr0n into the stereo add one more soundcard.
I got a P3-700 sitting under my video, hooked to the stereo and TV. It does PVR, mp3-play, routing, dhcp server, NAT, print-deamon, wireless access point,
The whole thing set me back around $190, one year ago.
I severely doubt that any positive effects on memory and reaction times is _caused_ by wireless phones.
...accidentaly, young persons are documented to have better memory and reaction time than average people :)
Actually (at least in europe), there are demographic differences between cell-phoners and the rest.
One difference being that younger people are over-overrepresented in the set of cell-phoners
A: "You stole my watch"
B: "No, i bought that"
A: "No, you stole mine!"
B: "Okay, show me somthing that identifies your watch"
A: "Can't do that... it's a secret watch!"
B: "uuuhhmmm, if you really think it's your watch, I already GOT the secret!"
If it's already in the linux kernel, it's publicly available. The only thing you cannot see publicly is which parts are actually SCO's.
For me to even consider whether SCO source is in the kernel, SCO should:
1. point to which source they mean are infringing
2. give reasonable evidence, that they actually have the rights to that source.
Should a judge decide that the source is SCO's, they will have a good propection for that source in the future.
This places everyone who ever accepts a signature from anther person in a position to man-in-the-middle attack that person, pretty clever.... DUH!
What is needed in computer-systems is zero-knowledge proofs, or at least a challenge/response method influenced by both parties.
The real use for bio-metrics is to verify relations between identity and private-key possesion in a trusted 3rd-patry setting, not for authentication itself.
As a defensive step, I have e-mailed SCO, stating that I am cercerned about their claims that I am infringing on their property, asking for an indication of exactly how I'm infringing on their property, asking (politely!) that they show me the offending code.
:)
:)
Of course, I am not in a position to license Linux from SCO without that information. So I'm just waiting for them to mail me back
(BTW: I live in a country where the LAW is actually reasonable, I don't think SCO could convince any Danish judge with their outrageous claims
googling for "rechageable faq" yeilds lots of hist, for example: http://www.buchmann.ca/faq.asp. But here's some advice anyway:
* Basically, you want rechargeable batteries whenever chances are your thingy with the batteries will not get nicked, since that defeats the investment in rechageables.
* Choose batteries with the highest mAh rating, which says how much energy you can draw from the battery.
* Choose NiMH for camera's and other power-suckers.
* Buy a charger with seperate charging cicuits for each battery.
* NiCD batteries have some problems with memory, but good chargers with "decharge" can usually fix that. (actually, NiCD batteries are hard to come by these days)
From the patent:
So, it would _seem_ a bit like:
1. build matrix:
A B C
D E F
G H I
2. to cipher up the letter F which is at row 2, col 3 send (2,3).
3. mutate matrix, goto 2
So the real "crypto" lies in the mutation of the matrix... how that is done is not described... maybe it's just x-or'ed onto itself or whatnot.
The way the key is found has nothing to do with the value of the crypto, so don't even begin to critisize how easy it must be for an attacker to guess which file is being used as key.
An optimized version of that algorithm would require only 1*len bits ;)
The trust model of X509 Cerifitates is fundamentally flawed, in that it does not mimic the trust model applied in "the real world", but an authoritarian one.
:)
In the real world, you trust someone if enough "peers" that you trust trust that someone, and probably a bit less
Hey wait, that's PGP's model!
So if Microsoft ultimately wins a major share of the market, then using a Microsoft DRM client may be necessary in order to buy music.
The solution to that problem is already being effectively implemented in the Computer Programs area: generate a vast industri, that doesn't pay taxes or royalties, providing people with unprotected material.
Restrict people enough, and they will turn rebel!
The difference between encoding and encryption is, that encryption is parametrised with a key.
... is an encoding
:)</fun>
ROT-13,
ROT, is an encryption, since it's parametrised.
<fun>Some encodings are not surjective, which makes them kind of in a grey area, since the interpretation of those elements which are not the image of any input is basically a key... This nicely explains why different text-encodings are so god-damn cryptic to get working in applications
I would venture the hypothesis, that any strong encryption has a mathematical model. Otherwise reasoning about the encryption would be rather hard.
--
Helge
The important thing, is to not provide Free (as in beer) training to one OS vendor, radically unbalancing the competition in the OS market.
The danish goverment spend millions of dollars each year on "teaching the people to use IT", which basically boils down to giving users a training course on all M$-OS and Office products.
I suggest having a mix of OS'es, so that the students have different experiences and learn from comparing those.
I myself is a student at DAIMI where machines with SunOS, HPUX (well not that many anymore) IRIX, GNU/Linux and Windows (Using vmware), and yes it's a pain with the differences between computers but:
1. You can just select to use the same OS every time
2. You learn a lot by seeing different solutions to the same problem
Here in Denmark, nobody, except the police, carry guns that are now for sport. Police guns are weak, and VERY RARELY (once every few years) kill.
1-10 people are killed each year by guns, fom a population of 5.5million.
Draw your own conclusions....
If Americans have problems protecting themselves from "villans" with guns, maybe they should start thinking a bit about how those villains got their guns.
Whenever "usability experts" start working on a program, they usually remove all the features that made it usable by me, dare I say GNOME2 :)