Perfect Encryption

[Kezil]

Though there will never be perfect encryption (that we know of), there are some ways to supplement existing systems to make it much more effective. One of these is to create your own language (and possibly even a separate script for it). By doing this, not only do they have to break the encryption, something with a key that has to be transferable in some physical way, but also interpret an entire language without any clues as to what even the simplest word is. The language doesn't even need a physical medium, as the two reading the message can hold the language in their head.

[darkes]

Originally posted here by ac1dsp3ctrum
Encryption is a process that is made to be reversed, and it will (legally or not) Even if it takes that person a year trying to break the encryption, it will be done... When people make stronger and stronger keys, other people make faster and more powerful computers to break that encryption.... So in other words, there is no such thing as a perfect encryption

That's not quite true - most encryption algorithms are devised so that it is as difficult as possible to reverse it (otherwise there wouldn't be much point!) - to reverse the process usually involves some mathematical weakness that was not known about, or more commonly a weakness with the way the algorithm is implemented in practice. Brute force alone won't work most of the time.
As was pointed out by Bithway714, the advent of quantum computing will put a whole new spin on this (sorry - bad joke). It has two possible applications - the first is to generate real random numbers, and the second is to use it as a secure communications method, as intercepting the message causes the spin on the particle to change, and both ends of the link know this has happened. The first prototypes to do this are very close, and I wouldn't be surprised if this was commercially viable in the next 5-10 years.
But, of course, nothing is perfect ....

[Tuskin]

I dont know if they do this but if you wanted... you could create space in the encrypted file and put 'noise' in it so that its even more a jumble if you dont get the key right... I think using a self made encryption program would do the job cause you would not only have to have the key but the program that would remove the 'noise' from the file... the key would also generate that "crazy data" :-) ... only downfall is an increased file size... nothing is perfect though

[Terr]

Well, even a one-time cipher isn't necessarily secure. The content of the message matters too. For instance, simple character-replacement could be easily foiled through some intelligent brute-forcing. Just look for the most common character. It is probably an "e". (And if there are a lot of three letter words ending with the e-character, they are probably "the"...)

[darkes]

In reply to Terr's comment about a one-time cipher, this really does come down to how the key was generated. If a one-time pad is used with truly random numbers, then the first time the letter 'e' will be changed to 'z', and the next to 'b' say totally at random. This destroys any pattern in the original text (so it doesn't matter what language is used!).
However, if the key used is not random, then some patterns will persist in the encrypted text.
Which is why using a quntum based computer to generate real random numbers could be very significant.

[cwk9]

That’s why you write it in Romanian or Japanese before you encrypt it. Who ever is trying to brute force it might assume its English and start looking for common patterns in the English language.

[The Old Man]

Originally posted here by Kezil
Though there will never be perfect encryption (that we know of), there are some ways to supplement existing systems to make it much more effective. One of these is to create your own language (and possibly even a separate script for it). By doing this, not only do they have to break the encryption, something with a key that has to be transferable in some physical way, but also interpret an entire language without any clues as to what even the simplest word is. The language doesn't even need a physical medium, as the two reading the message can hold the language in their head.

Interesting. In wwI the Allies used some soldier commo types from eastern US indian tribes quite successfully as military "code talkers". Between wwI and wwII the axis quitely enrolled a few "students" into schools with our natives and supposedly were able to gain a reasonable knowledge of those languages. Come wwII, the idea of native languages came up again, but this time the army used Navajo soldiers as "code talkers". But not only did the Navajos speak a language that no outsiders understood, it had a totally "foreign" sound with no ralation to any other language. The Navajo soldiers further modified their language with code words that meant military terms.... like bomber, observation plane, submarine, infantry company, and a bunch more i can't remember right now.
So Kezil's idea has merit. Maybe the next successful code language will be .... Klingon?

[The Old Man]

Oh, i forgot; if you're going to write something to a friend (probably not a proper business message) don't use proper english sentence structure (or whatever language you speak ) but use slang, poor sentence structure, mizspel wurdz, leev ov th ind littirz uf shrt wurdz an us phniticz,... Get inventive, use code words in your original before the substitution that will be used by the encryption program... Anyway, this is not meant to be even close to scientific in a short space, just a few ideas that will slow down the decipher process. A "secret message" usually only has a specific valuable life cycle, if you can delay the decipher process beyond that point, the effort spent in breaking it becomes worthless, and a drain on resources. Anyway, that's the way some of my friends see it, just thought i'd pass it on...

[Scorp666]

Wow, always trust a veteran for good ideas!

I suggest using animal speech, like whales, dolphins or even cockroaches lmao

Are you guys encrypting emails to your mom?

[jcmcb]

About those code talkers (as they were called) there is a new and presumably bad Nick Cage movie coming out this summer called "Wind Talkers" about it....just food for thought...

As for so callled "perfect encryption", humans have an incredible capacity for self delusion (this is not a personal attack on anyone, just a broad generalization) that often extends to the supposed power of encryption. We have thought we had perfect encrpytion before (engima comes to mind) but it has always only been as strong as its weakest link, which in this case is the human end user. If they refuse to follow the often simple safety protocols of password creation, how on earth will they use quantum cryptography correctly???

Just another thought from everyones favorite optimist....