Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
Closed 8 years ago.
This question appears to be off-topic because it lacks sufficient information to diagnose the problem. Describe your problem in more detail or include a minimal example in the question itself.
This question does not appear to be about programming within the scope defined in the help center.
Improve this question
I'm playing around with text transformations - ciphers. From all that I have surveyed it seems that all of these algorithms either break even in terms of transformed message length, or get larger. Are there any known algorithms/text transformations that when applied to a message actually make the message smaller (not counting the key, of course)?
For instance, RSA, when you encode the message, makes the encrypted message quite a bit larger than the original. Is there any such thing as that only the message becomes smaller, instead of larger, after (encryption, transformation, etc whatever you want to call it)?
I'm not doing this as part of security, so whether or not it's hackable is not of any interest to me.
P.S. I've done a lot of research in this area already through search engines (google, wikipedia, etc) but I have found no results. I don't want to say that such a technique doesn't exist without at least posting the question publicly first.
Thanks!
Compression tries to make input smaller. Obviously lossless compression will not make every input smaller, since that's impossible.
You can encrypt the compressed input if you want that. In principle compression and encryption are orthogonal concepts, but in some situations the length of the compressed text can be used to attack the system.
At first I thought about language transformation. Some English phrases translate to a single Chinese symbol. That's not a rigorous, mathematical example, but I suppose it qualifies.
Alternatively, from a bit-wise perspective, it wouldn't be possible to cipher/encode 2 bits of information in 1 bit.
Related
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
This question does not appear to be about programming within the scope defined in the help center.
Closed 10 months ago.
Improve this question
One of the biggest struggle with ML research is the creation of objective functions which capture the researcher's goals. Especially when talking about generalizable AI, the definition of the objective function is very tricky.
This excellent paper for instance attempts to define an objective function to reward an agent's curiosity.
If we could measure intelligent behavior well, it would perhaps be possible to perform an optimization in which the parameters of a simulation such as a cellular automaton are optimized to maximize the emergence of increasingly intelligent behavior.
I vaguely remember having come across a group of cross-discipline researchers who were attempting to use the information theory concept of entropy to measure intelligent behavior but cannot find any resources about it now. So is there a scientific field dedicated to the quantification of intelligent behavior?
The field is called Integrated Information Theory, initially proposed by Giulio Tononi. It attempts to quantify consciousness of systems by formally defining formally the phenomenological experience of consciousness, and computing a value Phi, meant for a proxy of "consciousness".
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 5 years ago.
Improve this question
I have a plain text and its cypher text. I know that the algorithm used was MD5. I want to break all cypher texts that are produced using the same algorithm.
Is there any way to do so?
Kerckhoffs's Principle applies here. Knowing the mathematics, and the interaction between the plaintext and ciphertext, will not let you break the MD5 hashing algorithm.
This is due to Shannon's principles of cryptography, outlined in 1945, "Confusion and Diffusion". In simple terms, this means that any even reasonably good encryption algorithm does not show a clear relationship between the cleartext and the ciphertext.
The short answer to your question is no, there is no way to break MD5 purely by knowing a cleartext and a ciphertext. There's no key, so you can't reverse engineer it like a simple XOR cipher.
However, **as MD5 is a very quick, processor-light algorithm, it has been (and is still) possible to simply bruteforce a vast array of cleartext strings, then compare your target ciphertext to the resulting **rainbow table.
This site can help you do this: MD5 Decryptor
I will mention, however, that it's generally rare that there is a use for this outside of computer misuse, which I will strongly caution you against.
I hope this was helpful.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 7 years ago.
Improve this question
What is the current state of serialisation-to-binary in Rust?
I have some large (1-10MB) data structure to be sent across a network, and don't want to encode them as JSON or hex (the two serialisers I have found).
I have found #[repr(packed)]. Is this what I should use, or is there something more portable?
#[repr(packed)] only makes your data small. It does not offer any format guarantees or serialization help.
You have a few choices here (ordered by my opinion from best to worst solution):
You can use the Cap'n proto implementation for Rust
https://github.com/dwrensha/capnproto-rust
It's not really serialization, more of a forced format for structs that are then sent over the network without any conversion
fast
You could write your own Serializer and Deserializer.
you have full control over the format
runtime overhead for every single datum
you need to implement lots of stuff
You can transmute your structs to a [u8] and send that
probably the fastest solution
you need to make sure that the compiler for the program on both sides is exactly the same, otherwise the formats don't match up.
Someone evil may send you bad data. When you transmute that back, you get buffer overflows and stuff
references in your data-structure will cause wild pointers and undefined behaviour
Don't use references
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
I'm looking at the various bcrypt implementations across several languages and noticing the character limitation across most - specifically, the 72 character maximum that node-bcrypt, php's bcrypt, and py-bcrypt all exhibit.
What are the advantages and disadvantages if an application were to run user input through, say, a SHA-256 or SHA-512 checksum beforehand to enable longer inputs for bcrypt?
The CLI application found here is limited to 8-56 "characters" inclusive (it's C, so a character can be anything I suppose). Heaven knows why you would create limits for something that you feed into a password based key derivation function afterwards (which almost certainly will take unlimited input).
An additional secure hash with sufficient strength and output size will not do anything to degrade security.
Encode the result to hexadecimals before feeding it to a bcrypt library, which is almost certainly expecting a String (don't get struck by the "odd" 00h byte). You might as well use SHA-256, I don't think a few bits more or less will make a difference if you feed it into bcrypt afterwards. Otherwise you may be forced to use base64.
Finally, try not to get into this situation, performing non-standard cryptography is almost certainly a bad thing.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
This question does not appear to be about programming within the scope defined in the help center.
Closed 6 years ago.
Improve this question
My company's web-site is all about online games, so accessibility is not high on our priority list. SEO best practices, however, are. Searching on the net we couldn't find any discussion of whether or not ARIA is such a best practice (which is a kind of answer already :-). I found this surprising because using ARIA roles seems like a natural thing to do - they contain a lot of SEO-relevant meta-data (overall page structure, what parts of the page are the "main" as opposed to "service" navigation area, what parts contain "actual" vs. "related" vs. "independent" content, etc.). What's more, given they effect the user interface (screen readers and so on) they would tend to be pretty accurate when they exist.
Does anyone have specific knowledge about whether any search engines actually use this data, if it exists in a page?
Search engines like Google are pretty smart no matter how badly you set up your page, SEO relevant meta data or not.
The main thing is to make sure your page is marked up properly, that it validates and that you don't employ any "black-hat" techniques that could cause search engines to black list your page.
As for ARIA, I'm not sure if it's really going to make much difference one way or another.