15
$\begingroup$

Puzzling is unique across the SE network for allowing "partial answers" -- that is, answers that aren't full answers to the questions, but just make some progress towards answering it. But recently, I've seen several answers that just give tentative answers to a small piece out of a large puzzle. This generally happens when a puzzle (or part of a puzzle) has a "list of clues" format, with a lot of small pieces that can be answered independently. Do these -- should these -- count as "partial answers" at all? When have I made enough progress to post something as a partial answer?

$\endgroup$
13
$\begingroup$

In my opinion, answering only a small part of a large puzzle should not be a "partial answer", particularly if these answers are unconfirmed. What counts as a "small part" will of course vary from puzzle to puzzle, but here are some guidelines I recommend (and use myself), to decide when -- or whether -- to answer:

  • (1) A partial answer should clearly make some sort of confirmed progress.

If large parts of the 'partial answer' could be wrong, then it's possible that it couldn't be developed into a full answer, and therefore it's not a partial answer. This confirmation could be by words interlocking in a crossword, or by noticing that your clue answers all have some property in common (which is progress in itself!).

  • (2) Partial answers should make significant amounts of progress.

Answering a few cryptic clues isn't enough by itself for a partial answer, even though the answers may be 'confirmed' by the structure of the clue. The fewer clues in your answer, the less likely they are to be a significant enough fragment of an answer that someone else can develop an answer off of yours. If you've solved, say, 5 or 6 clues of a 50-clue crossword, there's no meaningful sense in which a full answer is "based off of" yours, even if the person who posted it did refer to your answer.

(You wouldn't post a Sudoku puzzle with three numbers filled in as a partial answer, so you probably shouldn't do something similar with other puzzle types.)

  • (3) Partial answers should be as much progress as you can feasibly make on the puzzle (at time of posting).

Posting a partial answer very soon after a puzzle comes out, and then updating it several times immediately after with more and more bits of progress, seems to me like "rep-grabbing" with an intentionally rushed and incomplete answer. Even if this isn't the intent behind it -- and I'm sure it's usually not! -- it gives that impression to onlookers. And I believe promoting this sort of "post an answer as fast as you can" attitude harms the health of the site.

(If you're actively working on a large puzzle but still want to show that you've made progress, why not leave a comment with your progress? If I only have time for a few potentially-useful thoughts, that's what I generally do.)


So, for example:

  • In Sp3000's Signature Puzzle, solving most of the crossword but not figuring out the next step would be enough for a partial answer. There's clear significant progress there, confirmed by the words interlocking and filling most of the grid: so (1) and (2) are satisfied. And if you've solved the entire grid, you could still be stuck on the extraction step - it's reasonable that (3) could also be satisfied in this case. However, doing only the region in the top left corner, or a few random clues scattered around, would not be enough to warrant posting. It would likely fail (3), and even if you really couldn't make more progress it would still fail (1) and/or (2). This much progress might be worth a comment, but probably not an answer.
  • In jafe's 5x5 Image Grid, answering 4 or 5 rebuses would likely not be enough to be partial-answer-worthy -- I would just give the ones I had in a comment if I couldn't continue from there. (But answering a majority of the rebuses, even if you aren't able to continue with the second and third steps of the puzzle, would be worth posting as a partial answer.)

If we want Puzzling to be an archive of high-quality puzzles and solutions, I think very small partial answers are ultimately harmful. Jokes about "sniping" other solvers when two answers are posted very close together are good fun, and a bit of friendly competition in solving certainly never hurt anybody. But I do worry that when "racing to fully solve the puzzle" becomes "racing to post whatever progress you've made", the overall quality of answers sharply drops. Prioritizing speed over quality and thoroughness can't be beneficial.

If you haven't made much progress but still want to share what you've made to help other people solve, there are a few ways to do that that are better than answering. You can just post what you have so far in a comment (possibly rot13-encoded, or behind an Imgur link). You could also create a chatroom for a group solve and link that in a comment on the question. Or you could post your answer as a community wiki -- this isn't done as often as it should be here on Puzzling, but collaboration is the whole reason the feature exists across the network!

Partial answers aren't the only option for sharing your progress so far, and in situations where you don't have very much of it, they likely aren't the best one.

$\endgroup$
  • $\begingroup$ @JS1 I think that an answer with only 5 solutions out of 24 likely wasn't enough to count as a partial answer, at least under my guidelines -- if I only had that much, I would have posted it as a comment. (Though I didn't mean to target any particular user with this -- edited to not single anyone out!) $\endgroup$ – Deusovi Sep 8 at 21:27
  • $\begingroup$ Users on this site learn by seeing what other people do, at least that's how it is for me. And as in your original example of Stella and the president's secret notes and other many other puzzles, what I see is people posting answers that don't quite measure up to your guidelines 1-3. I'm not saying that your guidelines are bad, but it's just not what I see happening. Is there some official "readme" somewhere to explain it all? $\endgroup$ – JS1 Sep 8 at 21:56
  • $\begingroup$ @JS1 There's no official "readme" besides the help center (which is largely written for the network in general, and so less helpful in particular here on Puzzling.) Though that was proposed recently, and I'm considering writing one! These guidelines are not official at the moment -- I'm proposing that they should be official, or at least something like them should be. $\endgroup$ – Deusovi Sep 8 at 22:01
  • 2
    $\begingroup$ I feel that if people are too afraid to post partial answers, then difficult puzzles will simply vanish off the front page and not have any answers posted. A lot of times I see a fairly weak partial answer lead to a lot of activity and brainstorming by other users (possibly because it bumps the question to the top?). $\endgroup$ – JS1 Sep 8 at 22:05
  • 1
    $\begingroup$ @JS1 I mentioned posting the answer as community wiki, and making a separate chat room for that particular puzzle - both of those are good ways of sharing your progress and attracting attention without just posting a regular partial answer. Much more often than your scenario, I see someone post a partial answer with only a few obvious clues, shortly after the puzzle is posted. That just seems like low-effort rep-farming, especially if the answer is edited several times immediately after. $\endgroup$ – Deusovi Sep 8 at 22:10
  • 1
    $\begingroup$ @js1 They were used more often a few years ago -- that's exactly why I'm trying to promote these as an alternative! $\endgroup$ – Deusovi Sep 8 at 23:43
  • 1
    $\begingroup$ What do you propose be done with incomplete answers? And would it be all incomplete answers or only those which don't rise to the level of a "partial answer"? My inclination is the former and promote the nice options you mention if people want to collaborate. I often will see an incomplete answer and will either move on, or add my contributions in comments, not wanting to (or appear to) use the effort of other people. $\endgroup$ – SteveV Sep 10 at 20:16
  • 1
    $\begingroup$ I see the same problem as you with some partial answers which (to a cynical eye) look like "an epsilon part of this is easy to solve, so I'll churn out a quick answer". However, we should be careful not to discourage multi-answer collaboration with a policy like this. In particular, I think we should be pretty lax with (2): what's "insignificant progress" for one person might be "just the stepping stone I need to help me solve the rest" for another. Also, how would (3) work with "I've done this much, could do more but I need to go AFK now, I'll leave this for someone else to use" answers? $\endgroup$ – Rand al'Thor Sep 11 at 14:39
  • 1
    $\begingroup$ I'm still not sure what to think of this. On the one hand, the dividing line between "helpful partial answer" and "not enough to be worth it" is inherently subjective, and such subjective issues should usually be handled with downvotes rather than deletion. On the other hand, I do see the same issue as you, and your answer is quite persuasive. So ... I guess I'm all in favour of this as guidelines for voting on partial answers, but more leery about using it as criteria for deletion. $\endgroup$ – Rand al'Thor Sep 11 at 15:26
  • 2
    $\begingroup$ I'll add a comment or two here. PARTIAL ANSWERS have often appeared as someone wants to stake their early claim to (at least part of) a solution to a puzzle with many small bits to solve, with the intent to fill in as they make headway. Sometimes another answer preempts that further development by posting essentially if not actually a full solution to at least the puzzle stage that uses the bits. These situations are a bit awkward (note how often we see disclaimers about "I got these on my own" or "I didn't read Xyz's answer") as the more complete posting likely was entirely independent work. $\endgroup$ – Rubio Sep 20 at 7:37
  • 2
    $\begingroup$ ... This specific nascent policy aims largely at curtailing this type of answer, as an answer that is incomplete merely because the poster was racing to Just Post Something feels like gamification rather than actually trying to make significant progress "that seems to lead forward". Such answers, if posted and then abandoned when made irrelevant, should (in my opinion) be deleted as Not an Answer because they're really not, nor were they a stepping stone in the progression of other solvers toward a full solution. $\endgroup$ – Rubio Sep 20 at 7:41
  • 2
    $\begingroup$ However, answers that are incomplete but (a) made a demonstrable effort to solve a significant portion of at least the initial stage (as in, all the immediately visible "bits") of the puzzle, and (b) provided a starting point for subsequent answers to build upon, should of course remain (and probably upvoted) despite being incomplete, even if they are never fleshed out further—they served their purpose of advancing the solution. $\endgroup$ – Rubio Sep 20 at 7:45
  • 2
    $\begingroup$ I think the time factor is pretty important in how we consider these answers. Someone in the "I need to AFK now" scenario who posts a fragment of an answer very early in the puzzle's life, realistically shouldn't have even started answering; the unexpected happens, but if it's a pattern with a poster, this should be recognized and spoken to. The pattern of multiple solvers posting early progress piecemeal is anti-collaborative as SteveV mentions; if they want to collaborate, Deus's suggestions on how are far more on point. $\endgroup$ – Rubio Sep 20 at 7:52
  • 2
    $\begingroup$ Ultimately, I think we would of course not move to delete any answers while a puzzle is still in that early phase where more readily handled "bits" are actively being worked through. I would hope we see people (1)&(2) only post when it's (admittedly subjectively) "enough" and that they really should be posting when they've hit the point (3) of "this is as far as I am able to go; here, PSE, see what you can do with it from here"—and then of course be prepared to see someone else run with it from there on, using their partial answer as the stepping stone to the next (or last) step of a solution. $\endgroup$ – Rubio Sep 20 at 7:59
  • 2
    $\begingroup$ And when the dust settles, if we're left with fragmentary answers that ended up subsumed by others, even if (shortly) later, which ended up not being one of the necessary stepping stones, those we can (and I think should) cull after the fact. So - that was a few more than "a comment or two", but ... them's my thoughts. $\endgroup$ – Rubio Sep 20 at 8:02

You must log in to answer this question.

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