Slashdot Mirror


User: Karhgath

Karhgath's activity in the archive.

Stories
0
Comments
167
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 167

  1. Re:Probably a dumb question, but on Improvements in Teleportation · · Score: 1

    (English isn't my primary language, so sue me, hehe. Seriously, sorry about the typos, I usually try to spellcheck my posts.)

    Yup, most of the calculation involves complex number(like the amplitudes of qubits), difficult calculus and such. However, there is usually much less variables and they are all well defined. I would much rather solve a very complex, 5 variable system of 3 equations than one somewhat complex with 24 variables and 12 equations.

    Ok, this is very fishy and all, I know. Just forget it =)

    Also, I never said I was an expert in QM, most of my info comes from lectures and classes from Gilles Brassard, and I must say he is more into the theory and computation side than the physics side.

    So, in Quantum Computation, it's simple. I won't comment on the whole quantum mechanic field. Is that better? =) I'm pretty sure the physics of quantum mechanics are very complex and maybe huge, but I always pictured it as much smaller and consise than classical science because it plays with the microworld, which is much more precise and focused than the macroworld.

    So, I was probably just talking out of my ass. Sorry.

  2. Re:Does this... on Improvements in Teleportation · · Score: 1

    Yup, this is the Weirdest concept in QM, as there is no equivalence in classical science. As Gilles Brassard once said in a class : "If you begin to accept this and understand it, you are as fucked up as I am. Welcome to the wonderful worl of quantum theory." =)

    I am in no way an expert, but I can tell you the following:

    Everything in quantum mechanics is about probability. A qubit is a surimposition of many classical state, and each state has a certain 'amplitude'. For example, if it has only one value, it is either 1*|1> or 1*|0>. |0> and |1> are the representation of classical bits 1 and 0.

    A qubit is represented as follow: [a*|0>] + [b*|1>] = qubit, where 'a' and 'b' are the amplitudes of both values. They are complex numbers, and (a^2) + (b^2) = 1. (which is quite logical, since amplitude is kind of a %) It is very simple algebra actually, and only the concept behind is hard to grasp.

    However, when you look at the qubit, it will assume one of the 2 values, destroying the other value. So, if the amplitudes are equal [1/sqrt(2)*|0>] + [1/sqrt(2)*|1>], there is 50% of chance you will get a 0 and 50% that you'll get a 1, you have no way of predicting it. So, if you look at it, it could become 1*|0> or 1*|1> with an equal probability. This is not very practical. One example Brassard presented is a 'magic page'. This page contains the whole novel, however, as soon as you start reading the page, it will choose a random page of the book and show it to you, for example page 365, destroying all other pages. You now have one page of the book, and you 've lost everything else forever, the rest of the book is gone. This is totally impractical to convey classical data, and is in no way capable of replacing classical data at all. However, it has a lot of other uses (encryption, teleportation, speeding many algorithms, parallelism, etc.)

    Actually, when I said 'look at it', I mean using a Measurement Gate. Thing of it as a classical gate like AND or OR, however, the purpose of the gate is to give you either |0> or |1> according to the amplitudes. There are not a whole lot of quantum gates, 3 that I know of: Measurement(M), Walsh-Hadamard(H) and XOR/Controlled Not. A quantum gate ALWAYS has the same number of input and output[since you cannot copy or clone data].

    Measurement Gate:
    http://www.msri.org/publications/ln/msri/20 02/quan tumintro/brassard/2/banner/17.html

    Walsh-Hadamard Gate:
    http://www.msri.org/publications/ln/msri/20 02/quan tumintro/brassard/2/banner/18.html

    XOR/Controlled Not Gate:
    http://www.msri.org/publications/ln/msri/20 02/quan tumintro/brassard/2/banner/19.html

    Was that helpful? I hope so =) It is basic algebra with complex numbers. Now, think about the possible values of 2 qubit together =)

  3. Re:Nevermind material transportation... on Improvements in Teleportation · · Score: 1

    Well, quantum teleportation is useful when you start thinking about network repeater. Since qubit cannot be cloned or copied(since looking at them disturb it), but the signal still 'weaken' over distance, introducing noise and all, a practical repeater is needed, and without quantum teleportation, it would be impossible to do it.

    QT opens might become a nice way to build a practical quantum repeater. However, 'might' is the keyword, as it could prove infeasible or too costy.

  4. Re:Probably a dumb question, but on Improvements in Teleportation · · Score: 1

    However, quantum configurations are INFORMATION. There are not matter. The spin of an electron is a qubit, the angle of a photon is a qubit, a photon in a special reflective chamber is a qubit, etc.

    By the very definition of what a quantum information is, we can't just "find something else under it". We can add more things considered quantum information, yes. It won't change anything in the theory. However, if we discover something else, it won't make Quantum Mechanics wrong, it will just be a new science completely parallele, alongside classical science and quantum mechanics.

    It is a LOT easier to proove quantum mechanics than even the simpliest thing in classical science. Quantum Mechanics is mostly small, straightforward and self-consistant.

    The small hic is that, we need classical science to proove Quantum Mechanics... so, classical science principle used to proove it could be wrong, but everyone highly doubt it. There are a LOT less ambiguity in QM, and there is no 'hidden variables or concept' we could uncover. If we do find something like that, the whole QM will fall, it cannot just readjust like classical science does everyday, but since we have so much practical proof of QM, it's highly unlikely.

    Really think of quantum mechanics as something totally different alongside classical science.

  5. Re:teleportation repeaters and quantum cryptograph on Improvements in Teleportation · · Score: 1

    My error: this is the intercept/send attack I described above. This paper talks about a Collective Attack, and proved quantum encryption is secure against this one.

    Sorry. Anyway, the same applies, it's a mathematical proof against another kind of attack. Just replace my example above with a collective attack example.

    Sorry again, should have read it more carefully.

  6. Re:teleportation repeaters and quantum cryptograph on Improvements in Teleportation · · Score: 1

    I study at the University of Montreal where Gilles Brassard works, and I think in a class he talked about that specific article by one of his student. I don't have time to read it all, but here is what I remember:

    When sending the quantum key, a spy can 'intercept' it, thus disturbing the key. So, Alice and Bob must find a way to detect those errors. They proved a way that accurately gives the % of error, thus knowing if someone disturbed the quantum information or not, and what % of information the spies has on the key, thus, making it a TOTALLY SAFE encryption method, as you accurately know if it worked or not. If it didn't, exchange another key, until the % is at a safe level. Even if the spy has 5% of the information, it's not enough to get anything useful out of the partial keys he has. To get the % of error, Alice and Bob compare a proportional portion of their keys. If there was no spies, it will be nearly 100% exact. If a spy eavesdropped, it will be disturbed a lot. They just found a fast, small, secure and mathematically proven way of comparing parts of the key, giving an accurate % of error and knowing if the key was disturbed or not.

    If you do NOT check, the spy could intercept the key Alice sent to Bob, it would be disturbed, then Bob would think he has the correct key, then Alice send a message to Bob with the key, and the spy can read it, but Bob cannot. A VERY smart spy would, when he intercepted the key, create ANOTHER key, send it to Bob. Then, when he interecepted the message, he reencode it with the key he sent to Bob, and Bob would get the message correctly, thinking it wasn't intercepted. The paper proves that you can know if the Spy attempted this or not, making quantum encryption TOTALLY secure.

    Ok, the above wasn't 100% clear:

    Stupid spy:
    ------------------
    A send quantum key X to B.
    E intercept X, thus X is now Y.(qubit was distured)
    B receives key Y.
    A send encrypted message MX to B.
    E intercepts MX. Decode it with X.
    B receives MX, but cannot decode it with Y. He know someone intercepted the message, but it's too late. A and B stops exchanging info.

    Clever spy:
    ------------------
    A send quantum key X to B.
    E intercept X, thus X is now Y.
    E create quantum key Z and send it to B.
    B receives key Z.
    A send encrypted message MX to B.
    E intercepts MX. Decode it with X.
    E send encrypted message MZ to B.
    B receives MZ. Decode it with Z.
    Now, B thinks no one intercepted the message and key, so A and B continues exchanging info, and the spy E gets ALL the information.

    This paper mathematically proved it's possible to know if Eve intercepted the key and send a fake one to Bob. Alice and Bob just have to compare a subportion of their key to know if there was a spy or not.

    Hope this helped =) As fo the exact info in the paper... I couldn't explain it to you, it's not my field of expertise.

  7. Re:This can produce interesting results on Improvements in Teleportation · · Score: 1

    Nope, you cannot clone or copy quantum information. When you mesure a qbit, it is disturbed and lose information. There are no way to 'xerox' qbits.

    Also, here we talk about qbit, quantum information, not matter or anything. So, you can't be hacked while teleporting, because you CAN'T be teleported with quantum teleportation.

    A 'spy' could intercept the transmission, but he would get NOTHING, as he needs the other pair of the EPR, which he doesn't have. He only received parts of classical information that makes no sense.

    http://www.research.ibm.com/quantuminfo/teleport at ion/

    Here for an illustration.

  8. Re:Does this... on Improvements in Teleportation · · Score: 1

    No, this analogy is wrong.

    A better analogy is the following:

    Still the cats but without the poison treat(such a variable might be tempting to be used to illustrate the concept, but there are NO such variables in EPR and the analogy will be wrong. It was proven.)

    You have 2 Cats in a box, the only thing you know is that one is dead, the other alive.

    You put each cat in a seperate box. You move one box to pluto, the other to mercury. Until THAT point, BOTH cats are DEAD AND ALIVE. Cat A is DEAD and ALIVE at the same time, and Cat B is DEAD and ALIVE at the same time. This is the superposition of information in quantum information.

    If you open Box A, Cat A will IMMEDIATELY choose to be either DEAD or ALIVE, at the exact moment you look at it. This is random, you cannot predict what state Cat A will choose. And he wasn't dead or alive before hand, he was BOTH.

    If you then open Box B, CAT B's state will be the opposite of A. Always. If A choosed to be dead, B will choose to be alive. They didn't determined this beforehand while together in the box, according to the current EPR theory(not the Einstein explanation). I don't remember the name of the one who proved this, sorry =( It was said to be something philosopher should discuss, but that it could never be determined if Einstein was right or not. He was proved to be wrong, in the 70's I think.

    Even if you open the boxes at the EXACT same time, or so close apart that you couldn't have time to propagate the state of the Cats at the speed of light, the same will apply: Cat A will choose a state, Cat B will choose the other. However, you have no way to know what it will be.

    This is still not perfect, but not blantatly wrong like the above one =)

  9. Quantum Teleportation on Improvements in Teleportation · · Score: 5, Informative

    Ok, before everyone freaks, Quantum Teleportation isn't what most think it is. I've had a class and attended a few lectures by renowned Gilles Brassard from the University of Montreal, one of the founder of the field and especially quantum encryption, along with Charles H. Bennet from IBM and many others.

    First, "teleportation" only teleports "DATA", quantum information, like the spin of an electron. You won't see any beam me up scotty, despite how much people wants to and how wrong reporters are in artciles =)

    Second, here's a VERY brief info page on Quantum Teleportation on IBM's page:

    http://www.research.ibm.com/quantuminfo/teleport at ion/

    For more in depth info, try to find articles in magazines and books, especially one written by Charles H. Bennet and/or Gilles Brassard.

    One lecture by Brassard can be found online here, there's even a PDF:

    http://www.msri.org/publications/ln/msri/2002/qu an tumintro/brassard/2/

    They will explain this much better than my understanding will do. It's MUCH funnier and interesting when Brassard presents it, and it's MUCH harder to understand too. The few pictures and bits at the begining of the lecture are what Quantum Teleportation is NOT. Even renowned scientific publication are fooled by bad journalism, and even IBM went over it's head with this, it's kinda funny =)

    Anyway, "Beam me up Scotty" will never result from Quantum Teleportation, so don't hold your breath =) The article briefly states this tho, but only seems to gloss over it and even says "maybe", which is completely wrong.

    Also, Brassard stated MANY times that is does not violate ANYTHING, and especially not Heisenberg uncertainty principle. The original DATA must be destroy, then it is "rebuilt" on the other side, and because of a property of EPR, "entanglement", you never mesure the quantum information completely, thus not violating Heisenberg uncertainty principle.

    One last note, the following bit on the article is probably the most simplistic non-explanation of what is Quantum Teleportation:

    "A message is combined with the encoding light particle. That combination goes to the recipient, who uses the decoding photon to decipher the message. Because no one else has the decoding photon, no one else can decipher the message."

  10. Who watched more than the 2-3 first episodes? on Firefly Likely to be Cancelled · · Score: 1

    Ok guys, a lot of people are claiming high and clear they dislike the show, it sucked, this and that is stupid, show X is better, and so on.

    How much episode you watched? Frankly?

    Do you REALLY expect the first 2-3 episodes to be any great? In fact, I love the show, but the first 2 episodes weren't that great, the second one is really the worst yet, in fact, very sub-par. Same happened with X-Files(watch the pilot and the few first episodes... ugh! And I bought that on DVD, hehe), same for Buffy, same for Farscape, or heck, even Babylon 5. I couldn't stand the first few episodes, and in some case, even the first complete season. However, I've watched X-Files for many seasons, until Mulder disapeared. Buffy until she died, farscape for a long while, etc. These show grew on me.

    If you only judge a show on the first few eps, well, sure you will be disapointed. It's like listening to the first few seconds of a song. A normal pop song, 10 seconds is enough, but something like electonica, progressive, and so on, the first few seconds are really not like the entire song, often and intro and such. Same thing here. Firfly isn't a pop piece you can judge on the first few eps, you need to listen to it to the end, and sometimes more than once for it to grew on you. You must be willing to give it a chance.

    Look, Firefly isn't a mainstream show I reckon. You might not like it, but it doesn't mean it sucks. Buffy wasn't popular, Farscape wasn't either. Their first few eps suck because they HAVE an storyline to follow, each episode is linked to the other. They can't do like any new show lately and start strong on the first ep, because each eps are different and unrelated, they must show them in order with slow character progression and development. It's not star trek.

    Some talked about John Doe. I personally don't like that show at all. The first few eps were 'strong', because... well, every eps is the same thing, with a different theme, and with the background story of him wanting to know who the heck he is. Same A->B->C formula over and over. MUCH worse than X-Files who had many other quirks to complement that. Sure, if you like one episode, you will like them all, because they are all basically at the same level. Since it's after Firefly, I watched nearly all John Doe eps, trying to see if it grew on me, but it's more a love/hate kind of show that you can check one eps and have an opinion. Totally different from Firefly.

    However, Firefly is quite different. There is an ongoing storyline, lots of character development for the big crew, etc. Each ep builds on the previous ones. It's slow, but when it starts to pick up at around the 4-5th episode, it's really great. It's better than Buffy, it's better than Angel, IMHO. However, if you don't like those shows, you might not like it and I can understand.

    But you have no rights to call a show sucky and ugly and all that because you watched the first 2 episodes. Firefly is not the kind of show that is strong from the start, it takes a while.

    If after seeing the pivotal eps of Out of Gas and the latest Ariel one, you dislike it, fine. But it's not because you don't like chocolate cake that you can attack people enjoying it and say that chocolate cake should stopped being baked and sold.

    Give it a chance if you want, and then you can say you really dislike it. Just don't go further, any comparisons and 'flaws' or 'incongruity' you state is just meaningless most of the time. You don't like it, you don't like it, you don't have to write a thesis on why, unless you want to discuss it and give it a chance and maybe, god forbid, give it another chance afterward.

    Oh, and I saw a few people saying they saw the ad on TV and didn't watched it because of it... Well, I tell you, give it a chance, the show is TOTALLY NOT what Fox shows you on the ad. Fox don't know shit about advertising shows. A lot of people were turned off by the ads(understandably), an Fox is responsible for that. Just try to give it a chance, you might be pleasantly surprised.

    So, the key word here is: give it a chance. If not, just move on... but, hey, it's slashdot here, so I'm writing that all for nothing.

  11. Re:Cooling System on Nvidia GeForceFX(NV30) Officially Launched · · Score: 1

    However, how much space do you have BEHIND your computer? 1/2 feet? 1 feet? 5 feet?

    It would probably be effective if you had lots of space behind your computer, which isn't the case most of the time.

  12. A new copy protection scheme! on New Audio Disc Formats and Copyrights · · Score: 5, Funny

    The RIAA is currently, and has been for quite a while, working on a new foolproof copy protection mechanism. With this system, no one will want to even try to bypass the copy protection mechanism and rip the CD, it will be futile and useless to even try.

    I was at their secret meeting on a undisclosed island in the pacific[shaped as the head of Hilary Rosen] and I was floored to hear of this new mechanism. They were working on it for many, many years, they were quite visionary. Slowly it is entering the unsuspected consumer's home even now. They call this new scheme JPCM: Just Produce Crap Music.

    No one will ever want to rip any CD anymore... Enter the complete DRM system for music!

  13. What it basically says... on Stealware: Kazaa et al Stealing Link Commissions · · Score: 3, Informative

    It's totally illegal. What the EULA actually says is :

    "By signing this contract you allow us to steal from your neighbor."

    This is the same thing, period.

    First, it asks the permission to someone not related to the contract's target, which is illegal. (You cannot have a contract that says: By signing this, you agree that your friend X owes us XX bucks.)

    Second, stealing is illegal.

    So, it doubly illegal!

    This is just sick.

  14. Let me get this straight... on Stealware: Kazaa et al Stealing Link Commissions · · Score: 1

    So, you agree to an ELU that basically says: Whenyou buy from an affiliate, we will be the one who receives the affiliate cash.

    How can this even be legal? A lot of people are spouting 'semi-legal', moral ethics, P2P is illegal in itself so the EULA is too... but, have you considered what the EULA says?

    This is equivalent to the following:

    "By signing this contract, you allow me to steal from your neighboor."

    So, if this is even remotely legal, wow, that's great, please, I just need one person to sign this contract declaring that all money transfers around the world should now be forwarded to me from now on. Anyone? I just need one signature...

    -Karhgath

  15. P2P in gaming. on DOOM 3 will use P2P System? · · Score: 3, Informative

    P2P architecture has existed way before any P2P file exchange system.

    Previous iD's games used a client-server architecture. Now, they changed it to use a Peer to Peer 'protocol' and architecture.

    What does it mean? Since it won't use a client-server protocol, you won't be able to join a game that has already started(that was stated at QuakeCon). The game is 'hosted' on each player's computer that exchange data about the current state of the game. There is no central server that handles all the load. Each player communicate in peer to exchange the information, hence the name.

    Peer to peer architecture is what is used in most Real-Time Strategy(RTS) games like Starcraft, Warcraft, Command and Conquer, Age of Empire, etc.

    So, there's absolutely no relation to P2P file exchange like kazaa and such, just he architecture that has been used extensively before.

    Now... the question is: why? Also, won't that allow hackers to create better hacks? Usually, games go from P2P to Client-Server because of security concerns, even if Client-Server is usually 'slower'. They rarely go the other way around. But that's another completely different topic.

  16. Re:Medieval American accents... on BioWare Has Neverwinter Publisher · · Score: 1

    And beside, Bioware is a canucks developer, so you'll have those awesome canadian accents =)

  17. Re:I have my doubts about the Linux release on BioWare Has Neverwinter Publisher · · Score: 1

    Both =)

    Only the toolset won't be released under linux... unless Borland C++ gets released one day for Linux.