It looks like you're new here. If you want to get involved, click one of these buttons!
I was about to upload some stuff to dropbox.
I was wondering how it detects whether the files are synched so quickly. Obviously, they don't send all the data over, either way. Of course, I'm aware that checksums exist, and figured that they probably used them.
I was curious as to how checksums worked. Obviously, not all checksums can be unique, if you have an infinite number of potential inputs to a checksum function, but only a fixed number of outputs. So I went to Wikipedia.
By now I've seen the following articles:
* Cryptographic hash function (I got here because I forgot the term "checksum" and instead looked up "hashkey", and got this as a suggested result)
* Birthday attack
* Collision resistance
* Cryptographic nonce
* Nonce word
I've also learned about preimage resistance and collision resistance, which are the guiding principles behind the design of a cryptographic hash function. I've also confirmed that, just as I suspected, such functions can never be perfect and there will always be at least two inputs that have the same output, but designers just strive toward the ideal of perfection by making it difficult for two inputs to have the same output--or at least difficult to find two such inputs.
Now I'm wondering what the difference is between second-preimage resistance and collision resistance. Can you help?
And now I've seen "Pseudoword" ("Logatome") and "Accidental gap". And "Gostak". Now I know what distims the doshes.
Comments
Once I was reading about New Coke and ended up reading about autoerotic asphyxiation.
A second-preimage attack is a collision where one of the plaintexts is a fixed, given value; for example, a particular message, and we wish to find another message that has the same checksum. This would allow you to replace, say, a cryptographically signed message with a different one with a valid-seeming signature.
A collision attack does not necessarily have any restrictions on the plaintexts except that they be different. Because of the logic behind the birthday paradox, collision attacks are much easier to find, and thus much harder to defend against.
So basically, the difference is whether you want to find something that has the same checksum as a particular thing you're interested in, versus whether any two things you may or may not be interested in have the same checksum at all.