0
$\begingroup$

Frequently, I stumble upon questions that are purely hypothetical and assume a specific technical base which does not yet exist. What I mean are questions going somewhat like this: "…let's assume that everyone has a quantum computer…" or "…is this or that algorithm theoretically secure in case technology advances in this or that way…".

From my point of view, such questions may fit better into philosophy or science fiction, as they are not really about "cryptography" but - at it's best - about "a possible future of cryptography in case technology should evolve a specific way". Simpler said: I have some doubts that such hypothetical questions based on assumptions really belong on cryptography.SE because any answer to such a hypothetical question based on assumptions can only result in hypothetical answers based on more assumptions.

Which takes me to my question: How much science fiction should be handled here "in theory"?

In other words: I have a hard time finding that small line between "they fit in here" and "they don't fit in here". If such questions fit cryptography.SE, what's the reason they do? And if they don't fit in here, how should we handle such questions? Simply by flagging them with an "off-topic - The question is based on assumptions about a theoretical future which is not guaranteed to become reality" or something like that?

| |
$\endgroup$
  • 1
    $\begingroup$ We have a good mathematical model for quantum computers. It might not be possible to build such a QC, but asking about the difficulty of a problem in that model is normal complexity theory. If such a question relates to crypto, it should be on topic here. $\endgroup$ – CodesInChaos Jul 30 '13 at 11:16
  • 1
    $\begingroup$ @CodesInChaos That much I understand and I know there are many questions (especially in quantum crypto) that do make sense. But there are also those questions that tend to leave the path of logic an translate to "Assuming that ABC exists, would XYZ be possible?" We can't answer "Well, let's assume DEF is invented to prevent just that problem" and be done with it. I mean, it would be a correct answer — but wouldn't it be as useless as the question itself? In a sense it would be like trying to "predict the future"; which would be asking for "personal opinions", or wouldn't it? [confused] $\endgroup$ – e-sushi Jul 30 '13 at 11:53
  • $\begingroup$ Many security proofs are based on a kind of speculation – that a mathematical problem is hard to solve. Do you want to forbid them too? $\endgroup$ – Paŭlo Ebermann Jul 31 '13 at 7:27
  • $\begingroup$ @PaŭloEbermann You got me wrong. I do not want to "forbid" anything. I asked my question for "orientation" purposes, as I was bit confused on how far questions could take "theory", "assumptions" and "speculation"… sometimes even beyond the borders of common sense. I think it's clear that most questions make sense even when they contain assumptions, but others seem to simply render themselves non fit for cryptography.SE due to the fact that the whole question is completely based on fiction instead of facts. Therefore, I asked where to draw the line so, since I don't want to make mistakes. $\endgroup$ – e-sushi Jul 31 '13 at 10:48
  • $\begingroup$ @PaŭloEbermann If you read my comment here, you will quickly recognize that my general goal is to be "more constructive", instead of more restrictive (as your comment implies). $\endgroup$ – e-sushi Jul 31 '13 at 10:51
3
$\begingroup$

I think it will greatly depend on the question, but for the most part I am okay with having well thought out "what if" questions as long as they specifically state their assumptions and it is clear that they are assumptions. The second point is important as we don't want to confuse people new to cryptography with something like "If SHA2 is broken, what do I use?" Someone unfamiliar with cryptography might see that and think that SHA2 is broken.

What I am not okay with are questions, for example, that the assumption can never be true. Like "If I devise a way to break the one-time-pad, what impact will that have on cryptography?"

Of the "science-fiction" questions I can remember seeing on here, I haven't had any big problems with them.

| |
$\endgroup$
  • $\begingroup$ Thanks for the feedback. As said: I'm simply not sure what to do with some questions. So I guess I'm understanding it correctly that - if there are any doubts about the possibility of an assumption - it's more constructive (aka smarter) to try to get the OP to reformulate the question more clearly? $\endgroup$ – e-sushi Jul 30 '13 at 12:27

You must log in to answer this question.

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