I am currently working on a series of mini-puzzles I’ll be posting on the main site, and I have an idea for a mechanism that utilizes features of users all across the Stack Exchange network, specifically their avatars and user IDs. Is this allowable? This meta post suggests that it is fine to use specific usernames for story purposes, but doesn’t provide any insight on using the users themselves.

  • 2
    $\begingroup$ Not an answer, but keep in mind that (and I see you're already aware) users can change their avatars and usernames, which may impact the "self-contained-ness" of your puzzle. $\endgroup$
    – Alconja
    Commented Dec 12, 2019 at 0:38
  • $\begingroup$ @Alconja The puzzle won't involve the usernames at all, just the IDs and the avatars. And there will be other clues besides the avatars to help figure out the users depicted $\endgroup$
    – HTM
    Commented Dec 12, 2019 at 0:40
  • 2
    $\begingroup$ Avatars change too, and users can (and too frequently lately do) request deletion. And while I’m sure you can find a way to point at an arbitrary user(id) on some other non-Puzzling SE site, I’m not sure you should, but that’s a lesser concern. $\endgroup$
    – Rubio Mod
    Commented Dec 12, 2019 at 1:53
  • $\begingroup$ Note a related past meta post here. There's one user in particular whose response to this question I am sure will follow... :) $\endgroup$
    – Stiv
    Commented Dec 12, 2019 at 9:22
  • $\begingroup$ I’ve figured out a better puzzle mechanism for what I want to clue that does not involve searching up users. But I would still appreciate a canonical answer to this question as the previous meta post does not address this issue fully. $\endgroup$
    – HTM
    Commented Dec 13, 2019 at 7:58
  • $\begingroup$ @Stiv Hello, is it me you're looking for? $\endgroup$ Commented Dec 15, 2019 at 15:39
  • $\begingroup$ @Randal'Thor You know what, it actually wasn't - but it's nice to see you anyway :) $\endgroup$
    – Stiv
    Commented Dec 15, 2019 at 16:55

1 Answer 1


I would highly recommend that you don't use specific users as part of a puzzle, for two main reasons:

Preservation: One of the many goals of this site is to be an archive of high-quality puzzles that can be solved at any point in the future. That's why we have the general guideline that puzzles not rely on external resources. Users can change their display names or avatars, or even delete their entire account! (And this isn't just a hypothetical - I can think of an instance of each of those from a prominent user off the top of my head.)

Obscurity: Having puzzles that rely on knowing particular Puzzling users may be problematic even if that information is still unchanged. The people who you refer to may not be anywhere near as prominent in the future. And additionally, people who aren't members of Puzzling will have very little chance of being able to solve the puzzle. You'd be creating something that is only solvable by people in this particular clique of active members right now. I think we should try to encourage participation from new people, and this seems to me to do exactly the opposite.

So even if you technically could do so, I think it would not be a good idea, both for the quality of your puzzle and for the goals of this site.

  • $\begingroup$ Thanks for the informative answer! I have one point of clarification though: when you and the other mods say “puzzles should not rely on external resources,” do you mean that the mechanisms cannot use any knowledge that requires outside resources e.g. web search or dictionary? Or is there more nuance than that? $\endgroup$
    – HTM
    Commented Dec 13, 2019 at 19:48
  • $\begingroup$ @HTM The guideline isn't exactly precise, but it boils down to "we don't want a puzzle to depend on one particular resource, because if that one resource goes down the puzzle is unsolvable". A puzzle that depends on the fact that the word "vexillology" exists is fine, even if solvers have to look it up. A puzzle that depends on Merriam-Webster's example sentence for vexillology is bad: if that goes down or changes, the puzzle is broken. $\endgroup$
    – Deusovi Mod
    Commented Dec 13, 2019 at 19:55
  • $\begingroup$ OK, that’s starting to make a little more sense. But I still have quite a few questions about that policy (which I may ask in a separate meta post actually). For instance, would a book cipher that relies on a specific printed book be off-topic? How about puzzles that rely on information from the SE network? How long does the puzzle have to be future-proof? Etc, etc $\endgroup$
    – HTM
    Commented Dec 13, 2019 at 20:09
  • $\begingroup$ @HTM I would advise against any information that is subject to change. The guideline isn't clear-cut -- the boundary between "okay" and "not okay" is sometimes fuzzy (though there are definitely things that fall on one side or the other). But I would personally say that the book cipher is not a good idea (because different sources may have different editions, and some might be paywalled); other information from the SE network is definitely subject to change; [...] $\endgroup$
    – Deusovi Mod
    Commented Dec 13, 2019 at 21:28
  • 1
    $\begingroup$ [...] and ideally, the puzzle should be future-proof indefinitely (with the knowledge of when it was written, of course: there's nothing wrong with a puzzle that depends on (for example) a set of sports teams, even though teams may form or dissolve. The issue is access to information becoming impossible, not the information itself becoming outdated.) $\endgroup$
    – Deusovi Mod
    Commented Dec 13, 2019 at 21:31
  • 3
    $\begingroup$ I think any question that depends on vexillology should be flagged. $\endgroup$
    – Gareth McCaughan Mod
    Commented Dec 18, 2019 at 15:22

You must log in to answer this question.

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