10
\$\begingroup\$

A certain (currently on hold) question was retagged from to with the reason given

Because the challenge is asking for the smallest possible solution, this is a [puzzle]

I was unconvinced, so I had a look at the tag's summary and wiki entry:

A puzzle is a problem or enigma that tests the ingenuity of the solver. Solutions may require recognizing patterns and creating a particular order.

and

Puzzles are often contrived as a form of entertainment, but they can also stem from serious mathematical or logistical problems.

This doesn't seem to make it very clear which questions the tag should be applied to. On a broad reading, it could be applied to 90% of the questions on the site.

The obvious next step is to look at questions which have the tag, but that doesn't clear it up either. Some of them appear to have it because the question is a programming puzzle: e.g.

Others are about a puzzle: e.g.

What should mean? Do we need to split this tag? If so, what should we call the new tag(s)?

\$\endgroup\$
  • \$\begingroup\$ This is even further complicated by the new Puzzling site ;-) \$\endgroup\$ – Doorknob May 16 '14 at 14:07
9
\$\begingroup\$

Tags have auto-complete, so the tradeoff between a shorter tag vs a clearer tag is definitely in favour of the clearer tag. With that in mind, I propose eliminating the tag entirely in favour of and .

undergroundmonorail's proposed definition suits perfectly. 's short description would be something like:

A puzzle is a problem designed to test ingenuity. Questions in this category ask for a program which can substitute for that ingenuity.

\$\endgroup\$
  • \$\begingroup\$ Since this has moderate support and no objections, I will start retagging slowly. Please don't help me: I'm trying to avoid filling the front page with uninteresting changes. \$\endgroup\$ – Peter Taylor May 31 '14 at 7:43
  • 1
    \$\begingroup\$ Retagging completed. \$\endgroup\$ – Peter Taylor Oct 31 '14 at 12:10
4
\$\begingroup\$

This question seriously needs an answer. is simultaneously meaningless and far too broad on this site. It came up today in chat briefly but nothing really happened with it, so I'm throwing a proposal into the ring.

In my opinion, most questions on this site aren't "puzzles" (which is why I'm kind of sad that this never went anywhere). "Write a program to solve this problem in the least bytes" is not a puzzle. Neither is "write a program that solves this problem and gets the most upvotes". Those are challenges.

A puzzle has you working in a smaller scope. It isn't just "solve a problem", it's "solve a problem that's been set up specifically to be difficult for you". You aren't only working to solve the problem, you're also working against the question. When is a giraffe not a giraffe? is a programming puzzle. The javascript game Untrusted is composed entirely of programming puzzles.

Without further ado, my formal proposal for a meaning for :

A includes a goal, a partially completed program, and rules outlining how program can be modified. The program is specifically designed to make achieving the goal difficult. An answer to a takes the program and modifies it only in ways specified in the rules, so that the goal is achieved.

I'm sure this could be worded better, but I hope it's fairly clear. As an example: This question posted today meets these criteria.

  • Goal: Get the flow of code into the outer catch block.
  • Partially completed program designed to make the goal difficult: The program has an extra try/catch block inside the first, which makes achieving the goal difficult because...
  • Rules on how the program can be modified: You're only allowed to add code between the 5th and 7th lines.

Under these rules, it would be a proper .

This isn't the only possible meaning of the tag, but I wanted to say something to hopefully get a discussion going.

\$\endgroup\$
4
\$\begingroup\$

Puzzles and objectivity

This is related to but separate from my other answer here. I have that one in mind while writing this, but there's no reason that ideas from this answer (or at least, slightly modified ones) couldn't apply if we go a different direction.

How do puzzles mesh with the requirement for an objective winning criteria? I can see three ways we could take this. That isn't to say there can't be more, but this is what I've thought of. Please vote in the comments to this post to decide which one to use.

1 - Keep it the way it is.

doesn't have to interact with the rule at all. Any question tagged with still needs one of , or .

2 - becomes a tag with rules meaning.

means "lowest bytecount wins". means "most upvotes wins". Today in chat, Doorknob said this:

I'm assuming the \[puzzle\] tag implies "first correct answer [wins],"

That makes sense for a puzzle. Once it's solved, it's solved. However, some puzzles are more open-ended than others, and the OP may want to judge based on something other than speed. In my opinion, the best way to handle this would be to say "On it's own, implies 'first working answer'. If combined with another win condition's tag, that tag overrides ".

3 - is an exception to the rule.

(I don't like this option, but it's here for completeness.)

Why do we have to have a win condition for a ? In real life, people don't usually "win" a puzzle, they just have fun solving it. We could do the same thing here.

\$\endgroup\$
  • 1
    \$\begingroup\$ Upvote this comment if you agree with option number 1 - Keep it the way it is. \$\endgroup\$ – ace_HongKongIndependence May 27 '14 at 1:26
  • 5
    \$\begingroup\$ Upvote this comment if you agree with option number 2 - [puzzle] becomes a tag with rules meaning. (Note: +1 vote from me) \$\endgroup\$ – ace_HongKongIndependence May 27 '14 at 1:27
  • \$\begingroup\$ Upvote this comment if you agree with option number 3 - [puzzle] is an exception to the rule. \$\endgroup\$ – ace_HongKongIndependence May 27 '14 at 1:28

You must log in to answer this question.

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