8
$\begingroup$

By "prisoner" puzzles I mean those in which several different entities (conveniently perfect logicians) are locked in separate rooms and given various amounts of information. The goal is for one of them to be sure that a specific condition is true, usually through manipulating lightbulbs, switches, or other communication methods. (This tag would not be a subset or a superset of .)

This is a fairly common puzzle genre:

So, what do you all think? Would this tag be helpful? If so, what should we name it? ?? Something else?

Edited:
I've added a few more examples of the genre, and am creating answers with suggestions for what this tag should be called, or for tag-not-needed. Please vote for the option you think should be taken.

$\endgroup$
5
  • 3
    $\begingroup$ captive-strategy? $\endgroup$
    – Gordon K
    Feb 6 '16 at 9:43
  • 3
    $\begingroup$ I think prisoners is a good name, and that this would be a helpful tag. $\endgroup$
    – Mike Earnest
    Feb 14 '16 at 21:17
  • $\begingroup$ What distinguishes these puzzles from the hat-guessing puzzles? We (or at least I) may need a clearer description of exactly what type of puzzle is being described here. $\endgroup$
    – Rand al'Thor
    Jan 25 '18 at 15:05
  • $\begingroup$ @Randal'Thor In hat guessing puzzles, the goal is for people to guess information known to everyone else but not known to them. In "prisoner puzzles", information is not necessarily known to all but one person, or even associated with a person at all. $\endgroup$
    – Deusovi Mod
    Jan 25 '18 at 15:33
  • $\begingroup$ (Oh, I just noticed that the top-voted answer here is for a tag which would be a supernym of hat-guessing ...) $\endgroup$
    – Rand al'Thor
    Jan 25 '18 at 15:36
7
$\begingroup$

cooperation-game

First suggested in this answer to Purpose of [prisoner], and then formally proposed in another meta post, Tag proposal: [cooperation-game], both by @A. P. as follows:

The tag description would be something like

A puzzle about finding a strategy involving several individuals cooperating to achieve a common aim. Examples are prisoners trying to avoid an execution, a group of people trying to guess the colors of their hats, or a magician and his assistants.

The tag would be a bit more specific than just because strategy also tags questions where a single person has all the information and optimizes the strategy on this.
On the other hand, the tag would incorporate all questions and a part of questions, like this one.

The use of such a tag is of course an easy search for people interested in "$A$ knows that $B$ knows that $C$ knows..." questions. But besides from that I can also imagine that for example a RPG gamemaster will have fun making his group solve one of these puzzles "live".

It has been noted that

This makes a lot of sense, and probably addresses the spirit of what was originally intended for the "prisoner" tag. — Phylyp

 

______

(This suggests the new tag is a superset of . I'm not convinced that it necessarily is, or should be. --Rubio)

$\endgroup$
1
  • 2
    $\begingroup$ The "co-operation" part makes sense, but not a fan of "game". Maybe cooperative-strategy? $\endgroup$
    – Alconja
    Jan 24 '18 at 23:10
4
$\begingroup$

captive-strategy

Suggested in a comment by @Gordon K

$\endgroup$
4
$\begingroup$

limited-communication

Suggested by @Deusovi in the original posting of this question.

$\endgroup$
0
$\begingroup$

Perfect Logician Strategy

Since I started this issue (I have since removed the tag from the original question, since it does not actually have to do with any prisoners), I would like to throw my hat into the ring. I think that there is a common theme to the mentioned puzzles, in that the subjects are generally perfect logicians and usually have time to discuss how they will resolve the puzzle (My puzzle is not this type of puzzle FWIW).

With a perfect logician strategy tag, we could easily search on anything involving prisoners, three mathematicians in a room, etc.

$\endgroup$
-2
$\begingroup$

We don't need such a tag

Possibly taking liberties by excessively removing his answer from its original context, namely the Purpose of [prisoner] question, @Rand al'Thor makes this argument against such a tag at all:

I don't think we need this tag.

Although I'm willing to be convinced if you can elaborate on this:

it does class a particular type of puzzles

What type of puzzles? As far as I can see, it describes a particular context in which puzzles can be set, which doesn't equate to a type of puzzle. Flavour text is most often irrelevant, and puzzles about prisoners could be (e.g. Four prisoners wearing black and white hats) or (e.g. The Honest Guard) or ("you are a prisoner and must solve this riddle to escape") or (e.g. Stargate escape) or ... well, almost any type of puzzle. The fact that the puzzle is about prisoners doesn't really seem to be useful information.

Or, to put it another way:

This [tag does] not help to define the topic of the question. Tags should help to describe what the question is about, not just what it contains.

$\endgroup$
1
  • $\begingroup$ I think I'll probably downvote this (my own words! :-o ) because while prisoners doesn't describe a specific type of puzzle, the idea of the tag suggested by Deusovi does seem to be a specific type of puzzle (even if I don't quite understand its definition yet!) $\endgroup$
    – Rand al'Thor
    Jan 25 '18 at 15:08
-3
$\begingroup$

prisoners

Suggested by @Deusovi in the original posting of this question. Subsequently used, in the singular, as a new tag for Who killed the indiscriminate warden? - but the plural form is probably better.

This tag (in the singular) has been discussed in Purpose of [prisoner], in which @boboquack notes:

Though it does class a particular type of puzzles, the setting is rather restricted (since similar puzzles can be set without prisoners and still be equivalent). My concern is that with the current name the tag doesn't mean more than could be achieved by searching for "warden" or "prisoner".

$\endgroup$
1
  • $\begingroup$ -1, for the reasons here :-) $\endgroup$
    – Rand al'Thor
    Jan 25 '18 at 15:03

You must log in to answer this question.

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