Not really. Sure it is for traditional "search ai". But this is hardly even a blip towards this "strong ai" that is going to replace everyone's job and make humanity its bitch. It is however a very interesting approach, using a combination of 2 different methods.
We have. The shit cheap chip cards struggle with any real security anyway. Yea who knew that if you be cheap you get cheap security. I worked on some 12 odd years ago, i was on contract with a different system a month or so ago, and you know what. The performance hasn't increased at all. However the memory had, and you know what, McEliece is faster than even ECC. So ECC was out, too slow. In the end we went for a shared secret key. Not much choice for how fast the thing was suppose to work.
And well it is irrelevant anyway. That is your choice, or go invent a public key system that uses mere bytes for keys and sigs, but somehow requires 2^256 operations to break without the secret. since that seems to be the only thing your interested in. And like i said shit chips give shit security. They are often very insecure because the vendor insists on the 50c per card option rather than the $2 per card. Like the US is just only moving past mag stripe. Clearly none of these decisions are about real security.
Yea cus having a human at the end of all that complex mechanical machinery we already have is somehow magic and make failures not serous. Like when a human has a tire blowout. Or if the breaks fail, human cognition will intervene and change everything. NOT. We already have life critical software/machinery everywhere. Traffic lights, fly by wire, brake systems, aircraft hydraulics, air traffic control systems, medical equipment,.. etc.
Oh for FUCKS SAKE, get your head out of your shit hole. NO SCIENTIST HAS EVER PREDICTED THE DEATH OF HUMANITY FROM AGW. EVER. Sure some dumb fuck enviro alarmist have. But no one reputable.
It really wasn't that stable over that time frame. Lots of people had to move around, drown and lots of other things. We have just become more accustomed to keeping all our stuff now.
Thank god someone on/. is not totally ignorant of AI. Just because the media goes on about AI finding cats on the web doesn't mean much in terms of "strong AI". We had this AI fever a few decades ago. Sure there has been some great progress lately (deep learning mainly, and big datasets), but these are specific algorithms that do a very narrow range of things, and even more interesting is that for these new methods to work well, very specific tuning is done *exernally* to help with the particular problem at hand.
Ahh yea the back in my day programmers were real programs and we didn't have lots o ram and stuff. And it like ran at 1Hz and we still made it go faster than you young wiper-snappers go on about. And my seat was covered in broken glass and we worked outside in the snow!
Get whatever grandpa. The good old days weren't. The source code is around to prove it as well.
I think your being irrational because you believe that human cognition is somehow magic. It is not. Machines will be better drivers than 99.9% of current drivers in the near future, and better than all humans a little while after that.
And its like most of the things that run out in front of cars get run over. Whats your point. That the magic of human cognition will somehow change the laws of physics? That they can somehow see into the future if the force is strong with them and despite reacting thousands of times slower somehow change the cars direction through sheer willpower alone?
Basically even on a geeks tech website we have a bunch of Luddites that are irrational when it comes to car they can't drive. Guess what your a shit driver and the machines can't replace you soon enough.
Fork out for the Enterprise drives. Then you will see what it costs. I infact do fork out for enterprise drives for my backups. I have 2x4T and 2x2T dirves. they are quite a bit more expensive.
I also wonder how your suppose to stay at someone else house anonymously, or hop in someone else car while they are driving it? I guess you could were a paper bag or something.
Anonymity doesn't make sense for somethings. Airbnb is one of em.
Its not 1980s anymore. We are not on 9600 baud rate modems. Hell i was offered 10GBit fiber for my domestic internet just yesterday, the server farm i am looking uses 100Gbit! I have over 10T of disk space on my desk. Also there are schemes to reduce key size and sig size in Lamport. Finally McEliece with Goppa codes is old and hasn't been broken and lots of people have been trying recently (lots of papers in the last few years) as well, there is even now a signature scheme using it. Not sure where you keep up with this stuff. Since there have been broken variants of both DH/RSA and ECC we should just assume they are also weak? Yea ya not making much sense.
They don't have a great history of remaining either unbroken very long, unencumbered by patents or having key sizes that are reasonable.
Yes they do. Lamport signature and extensions (merkel etc) are totally secure as long as the hash function is secure. And McEliece has been around a long time and not been broken. Neither has patents. So no idea what your talking about.
No he hasn't. If the error rate is a *constant* then error correction can work, with a *lot* more qbits i mite add. That is if the total "error" is simply proportional to the number of qbits. However that is *not* how the physics works, in fact there is quite a lot of evidence that the error rate goes up faster than proportional to the number of qbits, so if i add a qbit i need to add 2 more error correcting qbits which requires 4 more error correcting qbits..... This makes a quantum computer over a particular size/number of operations impossible in this universe. I know quite a few people working on this problem that believe this is the case, there is some upper limit below 1000qbits. There are of course others who don't think this is the case. But right now there is no proof either way.
First note that quantum computers that can even think about cracking the current crop of encryption are *thousands of qubits away*. Not 5. Quantum computers are exponentially difficult to build. And a 1000bit quantum computers is 100% totally absolutely useless for a job that requires 1001qbits. Also that 1 extra bit makes it about 2 times harder to build. We are talking about a rock abacus compared to a modern 15nm process CPU/GPU here.
Next note that it has no real effect on symmetric encryption. Also there are signature schemes that are also 100% based on hash functions, that quantum computers are no better at dealing with. Finally there are public key methods that quantum computers cannot break. They have large keys, but we are not on 9600baud rate modems anymore.
The main threat to security will be the same as it is now. Bad implementations and sloppy processes.
Slashdot Mirror
With solar power! You can't not have solar power somewhere for a mdsolar story.. oh and something about everybody being dead from a nuclear!
Not really. Sure it is for traditional "search ai". But this is hardly even a blip towards this "strong ai" that is going to replace everyone's job and make humanity its bitch. It is however a very interesting approach, using a combination of 2 different methods.
We have. The shit cheap chip cards struggle with any real security anyway. Yea who knew that if you be cheap you get cheap security. I worked on some 12 odd years ago, i was on contract with a different system a month or so ago, and you know what. The performance hasn't increased at all. However the memory had, and you know what, McEliece is faster than even ECC. So ECC was out, too slow. In the end we went for a shared secret key. Not much choice for how fast the thing was suppose to work.
And well it is irrelevant anyway. That is your choice, or go invent a public key system that uses mere bytes for keys and sigs, but somehow requires 2^256 operations to break without the secret. since that seems to be the only thing your interested in. And like i said shit chips give shit security. They are often very insecure because the vendor insists on the 50c per card option rather than the $2 per card. Like the US is just only moving past mag stripe. Clearly none of these decisions are about real security.
Yea cus having a human at the end of all that complex mechanical machinery we already have is somehow magic and make failures not serous. Like when a human has a tire blowout. Or if the breaks fail, human cognition will intervene and change everything. NOT. We already have life critical software/machinery everywhere. Traffic lights, fly by wire, brake systems, aircraft hydraulics, air traffic control systems, medical equipment,.. etc.
Oh for FUCKS SAKE, get your head out of your shit hole. NO SCIENTIST HAS EVER PREDICTED THE DEATH OF HUMANITY FROM AGW. EVER. Sure some dumb fuck enviro alarmist have. But no one reputable.
It really wasn't that stable over that time frame. Lots of people had to move around, drown and lots of other things. We have just become more accustomed to keeping all our stuff now.
That one developer is the guy costing the team time with their shit code. They could be replaced with a turnip.
Thank god someone on /. is not totally ignorant of AI. Just because the media goes on about AI finding cats on the web doesn't mean much in terms of "strong AI". We had this AI fever a few decades ago. Sure there has been some great progress lately (deep learning mainly, and big datasets), but these are specific algorithms that do a very narrow range of things, and even more interesting is that for these new methods to work well, very specific tuning is done *exernally* to help with the particular problem at hand.
Absolute shit. I am over 40 and i get very lucrative offers precisely because i am more experienced and older. Not just pure programming years either.
Ahh yea the back in my day programmers were real programs and we didn't have lots o ram and stuff. And it like ran at 1Hz and we still made it go faster than you young wiper-snappers go on about. And my seat was covered in broken glass and we worked outside in the snow!
Get whatever grandpa. The good old days weren't. The source code is around to prove it as well.
I think your being irrational because you believe that human cognition is somehow magic. It is not. Machines will be better drivers than 99.9% of current drivers in the near future, and better than all humans a little while after that.
And its like most of the things that run out in front of cars get run over. Whats your point. That the magic of human cognition will somehow change the laws of physics? That they can somehow see into the future if the force is strong with them and despite reacting thousands of times slower somehow change the cars direction through sheer willpower alone?
Basically even on a geeks tech website we have a bunch of Luddites that are irrational when it comes to car they can't drive. Guess what your a shit driver and the machines can't replace you soon enough.
Fork out for the Enterprise drives. Then you will see what it costs. I infact do fork out for enterprise drives for my backups. I have 2x4T and 2x2T dirves. they are quite a bit more expensive.
I also wonder how your suppose to stay at someone else house anonymously, or hop in someone else car while they are driving it? I guess you could were a paper bag or something.
Anonymity doesn't make sense for somethings. Airbnb is one of em.
Its not 1980s anymore. We are not on 9600 baud rate modems. Hell i was offered 10GBit fiber for my domestic internet just yesterday, the server farm i am looking uses 100Gbit! I have over 10T of disk space on my desk. Also there are schemes to reduce key size and sig size in Lamport. Finally McEliece with Goppa codes is old and hasn't been broken and lots of people have been trying recently (lots of papers in the last few years) as well, there is even now a signature scheme using it. Not sure where you keep up with this stuff. Since there have been broken variants of both DH/RSA and ECC we should just assume they are also weak? Yea ya not making much sense.
They don't have a great history of remaining either unbroken very long, unencumbered by patents or having key sizes that are reasonable.
Yes they do. Lamport signature and extensions (merkel etc) are totally secure as long as the hash function is secure. And McEliece has been around a long time and not been broken. Neither has patents. So no idea what your talking about.
You have absolutely nothing scientific to back that up. I should know. I have colleges working on the real science.
Nutrition and diet. Where every internet idiot can regurgitate bullshit and sound like they know what they are talking about.
As long as you don't do *anything* it is portable. Use a tcp socket, open a window, Use threads... and BANG, no longer portable.
Yea he is so incompetent to get the idea to execution first and a billion on the way. Your just Jealous
Or any other system for that matter. Or just a plain exe file or .sh on unix? STUPID USERS. As always. PEBCAK
Well i think over hyped and misunderstood are the problem. Just look at mainstream articles on quantum teleportation...
No he hasn't. If the error rate is a *constant* then error correction can work, with a *lot* more qbits i mite add. That is if the total "error" is simply proportional to the number of qbits. However that is *not* how the physics works, in fact there is quite a lot of evidence that the error rate goes up faster than proportional to the number of qbits, so if i add a qbit i need to add 2 more error correcting qbits which requires 4 more error correcting qbits..... This makes a quantum computer over a particular size/number of operations impossible in this universe. I know quite a few people working on this problem that believe this is the case, there is some upper limit below 1000qbits. There are of course others who don't think this is the case. But right now there is no proof either way.
It is also horrendously cumbersome and impractical for many real world cases.
There are quantum resistant signing and public key methods. So no. It won't be pre DH days.
First note that quantum computers that can even think about cracking the current crop of encryption are *thousands of qubits away*. Not 5. Quantum computers are exponentially difficult to build. And a 1000bit quantum computers is 100% totally absolutely useless for a job that requires 1001qbits. Also that 1 extra bit makes it about 2 times harder to build. We are talking about a rock abacus compared to a modern 15nm process CPU/GPU here.
Next note that it has no real effect on symmetric encryption. Also there are signature schemes that are also 100% based on hash functions, that quantum computers are no better at dealing with. Finally there are public key methods that quantum computers cannot break. They have large keys, but we are not on 9600baud rate modems anymore.
The main threat to security will be the same as it is now. Bad implementations and sloppy processes.