7
$\begingroup$

First, to be clear, I have no gripe and I do understand that "too broad" is a rather hot debate topic. I do have a question concerning this puzzle. It was very much on my mind as I composed it, that it was treading very close to "too broad" territory. I tried to counteract this by adding some hints that, I thought, would make it more obvious that my question was a fair one. It was pointed out to me in a comment that there was a discussion here on meta which ruled that hints were not the way to avoid a "too broad" classification.

Adding the hints hurt the puzzle, in my opinion, and it turns out that they were not the way to achieve my ends. I'm happy to be better informed now. I do have a question which I hope can be answered in a constructive way and it has to do with the framing of the puzzle.

To my mind the keys to the puzzle were in the title and the question that I posed. The question, of course, was "Which name does not belong on the list?" (emphasis added). The title was a clue that the wording of the question should be looked at closely. The question is not which person does not belong on the list but rather which of the names differs in a significant way from the others. I think this ought to cover it but I would like to hear other opinions.

I am happy with the outcome. No one gave me a hassle in any way which is really more than I deserve. The puzzle actually originated in my desire to take a playful poke at people who go search-engine crazy in an attempt to solve the puzzles here. I thought it would be kind of a cute practical joke to send them out combing wikipedia for facts about such diverse people as Ignatius of Loyola, Edward Albee and Charles Darwin only to find that the puzzle was really about the spelling of their names.

So, way too much detail, but the question is: are there any sort of guidelines concerning a puzzle that looks as if it's too broad but is relying on a trick to stay technically within the boundaries?

The other course I was considering was to ask for a hearing of some sort before any vote to put on hold. I don't know if a mechanism for that sort of thing exists.

Thanks for your time.

$\endgroup$
10
$\begingroup$

Disclaimer: this post contains subjective personal opinion, not a definitive answer...

I think the issue with questions like this (your linked puzzle provides a good example to discuss, but it's much broader/general issue than just this one post), arise when there are too many possible answers that are all equally correct. Even though your intended solution is arguably the "best", there's technically nothing "wrong" with the other possible solutions. This is my personal measure of whether to VtC as too broad.

That being said, there is nothing wrong with a puzzle having multiple correct answers, so long as the intended one is the one that people look at and say "Aha! That's clearly the solution!".

However, as you observe, that can be challenging with "trick" questions (and generally with riddles, or lateral thinking puzzles), since trying to narrow the possible answers via hints, tends to ruin things a bit. The way to avoid this, I think, is to provide, confirming clues - that is to say, fragments that don't necessarily hint at the solution very much, but help to prove beyond doubt that it was your intended solution, once it's found.

It's probably easier to demonstrate via example, so tweaking your puzzle:

Someone gave me this list of six names, but said that one didn't belong. I've been scratching my head all afternoon, and can't see the common thread. It's really bugging me...

Which name does not belong on this list?

Charles Darwin  
Anthony Burgess
Edward Albee
Alice Munro
Ignatius of Loyola
Phyllis Schlafly

I don't think adding this little bit of flavour at the front harms the puzzle too much, and it doesn't really provide anything that would be read as an actual hint, but once you do solve it, you'll likely realise the references to six, scratching my head and bugging me, confirming your answer.

$\endgroup$
  • 3
    $\begingroup$ I'm cringing really hard at the idea of adding those particular hints (in this entirely hypothetical and not at all real example), since, $\mathbb{IMHO}$, they would have made the puzzle far too easy, but otherwise I think this is a great statement of the issue. We take for granted the fact that every puzzle potentially has multiple answers and that the accepted answer just happens to seem like the only possible answer in the absence of others. $\endgroup$ – question_asker Mar 4 '16 at 13:03
  • 5
    $\begingroup$ @question_asker - yeah, probably laid it on a bit thick in my slightly cheesy example, but hopefully still illustrates the point (that being said, it's easy to over estimate how easy something is when you already know the answer). $\endgroup$ – Alconja Mar 4 '16 at 13:23
  • 1
    $\begingroup$ Makes sense to me. I'll buy that. Adding the correct frame to the puzzle could do a lot. $\endgroup$ – Hugh Meyers Mar 4 '16 at 18:03
  • 1
    $\begingroup$ I had my answer to this question written up last night, but forgot to post. Putting just a bit of framing material makes for a much more satisfying puzzle. $\endgroup$ – Jon Ericson Mar 4 '16 at 21:23
13
$\begingroup$

I think the hints are necessary parts of this particular puzzle. The accepted answer showed two other possible solutions. Another answer showed a viable solution and my guess was that the answer was Edward Albee because it was the only name where first and last are not ordered alphabetically. But none of those answers fit the second hint.

People are pretty good about finding patterns, so puzzles that rely on them can be difficult to specify:

The Hatter opened his eyes very wide on hearing this; but all he said was, 'Why is a raven like a writing-desk?'

'Come, we shall have some fun now!' thought Alice. 'I'm glad they've begun asking riddles.—I believe I can guess that,' she added aloud.

'Do you mean that you think you can find out the answer to it?' said the March Hare.

'Exactly so,' said Alice.

'Then you should say what you mean,' the March Hare went on.

And people did find answers to the intentionally nonsensical riddle—all different and equally plausible. The Hatter might have taken your approach and included more like objects and at least on unlike. That would have helped, but not solved the problem entirely.

Another strategy I've seen is to provide a hint in the setup. Instead of asking the bare list, perhaps tell a little story that includes a hint. Maybe something like:

'That's not a fair riddle. Let me ask one,' said Alice. 'Who is my favorite person?'

'How could we possibly know?' asked the March Hare, 'It could be anyone at all.'

'Alright then,' Alice replied after a moment's thought, 'It's the odd one out of Charles Darwin, Anthony Burgess, Edward Albee, Alice Munro, Ignatius of Loyola, and Phyllis Schlafly.'

The Hatter, who had scribbled the names on a scrap of paper, laughed, 'Easy enough. It's the one who shares your name!'

'No that's. . .' Alice started to say, but the Dormouse spoke up, 'There's something crawling in your mane.' Then he fell back to slumber.

At this, the Mad Hatter began tapping the list with his pencil and the Hare studied it sagely. Suddenly, the later looked up and announced the answer.

What was the answer and how did he know?

I don't think the puzzle needs to be closed necessarily but it would be good to polish puzzles that are a bit too open ended before they get too many alternative answers. If it takes some time to make the edit, closing seems like a useful tool in the meantime.


On a side note, the title, "Read and Answer the Question", is far too generic. If you have a framing story, that could help you pick a more interesting and descriptive title.

$\endgroup$
  • 2
    $\begingroup$ +1. Love your framing of the puzzle :c) $\endgroup$ – BmyGuest Mar 6 '16 at 17:37
5
$\begingroup$

I think maybe the best puzzle describing your problem - and teaching you a lesson - is the puzzle: "Find the letters that complete these five patterns" by Matt Malone. It created a lot of hot-dispute although it clearly is genius once you've seen the answer. It went to minus two-digits in rating and recovered to 54 and a nomination to be best puzzle of 2014!

This puzzle had the same problem. Initially, it was 4 lines of puzzle and there where really dozens of solutions which were valid and nothing to indicate one was better than the other. (The intended solution was only 'better' in the sense that it was harder to find and thus argueably more fun...)

Hints were added - and didn't turn the tide. Because the main problem remains: Why should one answer be better than the other?

What did make the puzzle unique and stress its quality was when additional content was provided. Suddenly there were five not one puzzle - and they required the same answer(method). This ruled out all - or at least nearly all? - of the 'simple' answers, making the "true" answer unique. Just another way of making the puzzle less broad.

I think one has to face it: Finding any interesting/obscure/fun way of connecting just a very small set of facts simply does not make a good puzzle. (Example: continue the number series: 1345, 2679, ... ?)

Adding (restrictive) hints is one, usually bad way of "fixing" this. Why bad? Because it is not a hint if it is required, but a required restriction, spoiling the fun.

Adding true restrictions is sometimes needed, though. f.e. I think it is okay to state restrictions like "The solution needs to be in English" or "You do not need the digital version of this image." etc. to rule out some crazy "alternative possibilities".

Providing more 'data' is the other, much better option. It still remains the puzzler's task to 'find' any restriction and he can 'test' any idea against all data - without the need of asking the author for aproval. And if he really finds an answer that fits all data? Congratulations! Time to be honest and say, "Wow, I haven't seen this. Now, can you find the other answer as well?"

So as far as your puzzle goes, you could have put up hints/restrictions right away - but strong enough ones to "rule out" other answers. However, hints like "Just the name is important, not the person" would obviously be a bad spoiler. Instead - more difficult to do, but that's what good puzzles require - you should have come up with a list of 20, 30 or even more names. Enough names, to falsify other 'ideas'.

$\endgroup$
  • $\begingroup$ Thanks for the thought-provoking post. I am trying this approach with my latest puzzle. In the case of the puzzle we are discussing here I don't think I edited it at all. I composed it with the "hints" included and with the express idea that they were more for seeing if your answer is right than to help you get to an answer. $\endgroup$ – Hugh Meyers Mar 6 '16 at 19:32

You must log in to answer this question.

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