Considering the resources intelligence agencies have available, they can probably crack anything anybody comes up with.

Except for the one-time pad. You can also find a link to Claude Shannon's article where he proves that the one-time pad can not only provide perfect secrecy IF implemented properly, but also that any perfectly secret encryption requires one independently random key for every symbol in the message. The one-time pad has been used during the Cold War, and the great majority of the messages sent have never been or will ever be cracked. The ones that were cracked were because the implementation rules were not followed, and thus the pads were two or three or more-time pads, rather than one-time pads.

Note that this type of encryption is only practical to use when sending an important message, not for storing data locally, like block cyphers of today. Imagine having to store a key as large as the HDD you were trying to encrypt. That would be quite useless, as you could simply lock the HDD inside a vault instead.

They have all the guns and all the encryption and the election process is tightly controlled to avoid giving alternatives to the current power structure. Just shut up and go along, its easier, and if going along causes problems for you, im sure some govt agency can find an appropriate drug and dosage.

Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery?
-- Patrick Henry March 23, 1775

These days a dangerous character like that would have been spotted in school and appropriately dosed or confined.

"The Constitution of the United States" doesn't mean a damn thing unless the people of the country constantly force it to have meaning. Plenty of banana republics have very nice Constitutions.

Never expect real people to have "terribly high" motivations when Simple Greed will get the job done. There is a tremendous (and secret) amount of m-o-n-e-y to be made by, for example, "convincing a government that it must Spy on Everyone and Everything." (Just imagine how many billions of disk-drives that this is selling right now, so that top men might get to look-at the information some day ... or, for that matter, even if they never get around to it.)

So far, 95% of the US Congress isn't even "allowed" to know what they're voting for: they don't have "security clearances." But, amid all that very contented grunting and slobbering, I don't notice any of the pigs complaining about that. Nor will they ever ... why on earth should they? "They've got theirs. You can just Eat Cake."

The real lesson of the Parable of the Prodigal Son is that people tend to drive themselves downward until there is simply nowhere else farther down to go, before they first start to think (if they ever do) about what it is that caused them to wind up in a sty, alone and friendless. Entire societies can have the same ailment.

Yep

SSL as always had issues, you don't need to attack the cypher itself, you can hit it with side channel attacks, such as BREACH which attacks the https servers compression algorithm, because of the way it compresses the requests one can inject clear text into the header and then solve for parts of the session key (if your really good at it, you can get the whole session key in 30 secs).

OpenPGP, will never be cracked in a very long time, (they have not yet been able to factor 1024bit keys, assuming they can and have, 2048 is miles away from being cracked..) What you are trying to achieve when cracking RSA encryption is solving n to solve n you need the prime number... if you can factor the prime number (the whole length of the key) then it is possible to solve for n which mean they could break it. the problem is (even with quantum computers) this won't be crackable in some time. Everyone seems keen on quantum computers, but have no idea how they work or why... a quantum computer will not solve RSA in our life times, if you try to crack a 4096bit key assuming you could build a quantum computer large enough (4 trillion quid bits) that are in full swing for about 90 years, then yes you could crack 4096Bit encryption (assuming there are no huge leaps in this field that is) and by 90 years from now we wouldn't be using 4096bit keys anymore anyways.

Also quantum computers do not solve problems any faster then a regular computer, because they can see whole sets of data and does not operate serially (like modern pcs do) doesn't magically give them the ability to solve problems any faster. You here people mention quantum and computer in the same sentence and it's like it's a magic box or something that can do anything.


So SSH/PGP/RSA unless you are using 1024bit keys are still safe, and even then there is no evidence that 1024bit keys have even been broken yet.

Remember that the NSA has computers designed specifically to do there kinds of operations very quickly.

The NSA will surely be able to break an encryption BEFORE there is any evidence that anyone can do so. I'm not saying RSA can be broken, but you shouldn't assume that it can't be.

See chapter 10:
http://www.cryptovirology.com/cryptovfiles/newbook.html

I thought AES was open-source?

A quantum computer should be able to solve factoring (and hence RSA) quite quickly, however, it might turn out to be like cold fusion: always 20 years away...

I'm pretty sure only 4096 qbits would be required.

Quantum computers can't magically solve everything faster, but factoring is one of the things they can solve faster given our current knowledge. It is also possible that factoring can be solved quickly on a classical computer if some mathematical advance is discovered.

I agree with this.

Extremely well said.

Yeah, but where did the constants used come from ?

Are the NIST Standard Elliptic Curves Back-doored?
http://it.slashdot.org/story/13/09/1...es-back-doored

Another reason for my recommendation that you look up the encryption on the net, on wiki. Search the page for "NSA", if the search comes up positive, choose another encryption.

Another reason to use twofish is because Bruce Schneier, Niels Ferguson, and others designed it, and at least these two seem more trustworthy.

A previous article:
http://cyberwarzone.com/did-nsa-put-...ption-standard

As for AES and Serpent also see:
http://en.wikipedia.org/wiki/XSL_attack

EDIT: My other concern is hash algorithms, because there aren't that many that weren't designed by the NSA (like the SHA). I'm currently using whirlpool because it is supported in the kernel, but would rather use something like:
https://blake2.net/

I know there are these sites for some info on hash functions that have been broken:
http://ehash.iaik.tugraz.at/wiki/The_Hash_Function_Zoo
http://en.wikipedia.org/wiki/Cryptog...ash_algorithms

Paradoxically enough, the man would -- in later years -- go on to purchase slaves!