3
$\begingroup$

So, we have this question, which currently has 4 close votes. All 4 are votes for "Requests for analyzing or deciphering a block of data are off-topic here, as the results are rarely useful to anyone else."

I'm not sure that really fits. The OP says "I do not want a solution to this, but I want an approach to solve such open ended problems."

I feel like we get a lot of these, where the poster says they don't want an answer, just a general approach to solving the problem. The thing is, I'm not sure such a thing exists.

Is that close reason right for these cases? If not, what should be done?

Edit

Here is another, maybe better, example. The OP is not asking anyone to analyze or decipher their data. Furthermore, they have a real need/desire to figure it out.

$\endgroup$
6
  • $\begingroup$ I think part of the problem is: We can't actually do a lot to help those people, so we don't even try. Modern crypto relies a lot on kerckhoff's prinicples (-> publishing algorithms) which get violated by such questions so we don't consider it crypto and close it right away. I doubt we can do better except for excusing for being unable to help. That makes me kinda sad :( $\endgroup$
    – SEJPM
    Commented Sep 8, 2015 at 20:10
  • $\begingroup$ @SEJPM, I agree. I just think we should do a better job in the close reason. If I were a new user and I posted a question like that and it got closed for the reason specified, I'd think the people on this site were unintelligent for not realizing that I am not asking them to analyze or decipher. I'm asking how to do it myself. Maybe the best thing to do would be to start a question "How do I go about analyzing ciphertext that looks like a classical cipher?" A kind of, "what are the basic steps I should take" sort of question. Then close these as duplicates? $\endgroup$
    – mikeazo Mod
    Commented Sep 8, 2015 at 20:22
  • $\begingroup$ @SEJPM I think your comment on the second question was very good. $\endgroup$
    – mikeazo Mod
    Commented Sep 8, 2015 at 20:24
  • $\begingroup$ We do have this answer and its associated question. Probably the closest thing we have currently. $\endgroup$
    – mikeazo Mod
    Commented Sep 8, 2015 at 20:30
  • $\begingroup$ Sounds like a good plan. We should link this question in our help center if possible so our members being really fast about closing have a better chance to find it. Or actually: We could create some more of this sort of question also handling the other types of classical ciphers and link them all and maybe create a "master-question": "How can I decipher data that used an unknown classical cipher?" that gives a brief overview and then redirects to the more detailed answers. $\endgroup$
    – SEJPM
    Commented Sep 8, 2015 at 20:33
  • $\begingroup$ @sejpm I wrote it up as an answer to give the community the ability to decide if that is a good approach. $\endgroup$
    – mikeazo Mod
    Commented Sep 8, 2015 at 20:39

4 Answers 4

Reset to default
3
$\begingroup$

Is that close reason right for these cases?

No it isn't. I voted for that reason, but shouldn't have. The correct reason here would have been 'too broad' as I mentioned in the comments. (But it was too late to change my mind.)

If not, what should be done?

Well, not close them for that reason. It's probably a good idea to edit out (example) blocks of data from such questions so that those with light close-finger don't just vote out of reflex. Assuming it isn't necessary for the question to have it, of course.

If the question should be closed anyway for another reason (that can't be fixed), I don't think there's much use worrying about it, though.

$\endgroup$
3
$\begingroup$

Is that close reason right for these cases?

I would say the answer to this is no. That is not the right close reason.

If not, what should be done?

Ilmari's answer on a question about cracking simple substitution ciphers is pretty good. At the very least, we should link users who post this type of question to that answer.

We could also edit the question that Ilmari answered to make it less about "simple substitution" ciphers (since it isn't clear from the OP that the cipher was in fact a simple substitution cipher) and more about "How can I decipher data that used an unknown classical cipher?".

Another option would be to start a new question that asks "How can I decipher data that used an unknown classical cipher?"

Then we can close these types of questions as duplicates.

$\endgroup$
2
$\begingroup$

Meanwhile, OP has deleted his/her question 2 hours ago (at time of writing this). But after checking its history I agree that the correct reason would have been “too broad”.

Reason:

I do not want a solution to this, but I want an approach to solve such open ended problems.

OP clearly did not target the decriphering a blob of data, but instead targeted to gain more general information about potential cryptanalytic approaches.

On a side-note: the question also hints at something that I’m seeing appear more and more lately…

This is more of a puzzle than an Industry problem.

An alike effort was recently discussed in our chat… leaving the open problem that neither Puzzling.SE, nor ReverseEngineering.SE offer a safe harbour for alike questions – while we generally tend to treat them as off-topic.

Since cryptanalysis very well is part of cryptography (maybe even one of the most important parts), something tells me that we should look into this and try to find a way to offer answers based on cryptanalytic options… instead of bluntly running for that close button to make it easy on ourselves.

$\endgroup$
3
  • 2
    $\begingroup$ Deletion is another thing that annoys me. There have been a number of times that someone will post a question like this, get an answer, then try to delete the question (since they don't want to be "caught"). But that is a side rant for another day :) $\endgroup$
    – mikeazo Mod
    Commented Aug 18, 2015 at 13:52
  • 2
    $\begingroup$ I agree with the cryptanalysis argument, but I believe OP question (and others) are not "legitimate" cryptanalysis problems. 99% of these questions don't show any prior effort, and are very close to "plz decrypt this thx". I don't think that a quality question regarding cryptanalysis would be blindly closed here. $\endgroup$
    – cygnusv
    Commented Aug 18, 2015 at 14:48
  • $\begingroup$ @mikeazo Echo that... $\endgroup$
    – e-sushi Mod
    Commented Aug 18, 2015 at 16:02
1
$\begingroup$

I completely agree with the close votes. I think that the solution to that particular problem is of no use to anyone else. It was covered here and turned out to be just the hash of an e-mail address (for recruiting) and some garbage. I doubt that this problem and its solution could be generally useful in the future.

$\endgroup$
2
  • 1
    $\begingroup$ FWIW, I agree with the close votes, but not the reason given. $\endgroup$
    – mikeazo Mod
    Commented Aug 18, 2015 at 11:47
  • $\begingroup$ I see. Anyway, as you said in the question, I don't think either that there is a "general approach to solving these problems", and, in case that such approach exists, I don't believe that this should be encouraged in this site, IMHO. I think such questions are more a puzzle than a serious crypto problem. $\endgroup$
    – cygnusv
    Commented Aug 18, 2015 at 11:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .