14
\$\begingroup\$

We have guidance about instructing that we do not allow and will revert any edit that attempts to guess the system of the question and edit it into the question here. We also have related guidance for how to handle a specific case where the answer was a system agnostic answer to question whose system was unclear here.

But we have no corresponding guidance for what to do when an answer which guesses the system gets placed on a question (usually moments before the question gets closed as being unclear).

The issue

I see such answers as an issue for a couple of reasons at least. For one, having an answer to a question can be a disincentive for OP to update the question to meet community guidelines. There has been at least one case where an answer has been accepted by OP without them bothering to update the question to include the system.

Another is that answering when the system is unknown can create confusion for the asker as well as future readers and a guess, no matter how sure a guess, is never guaranteed to be correct. I have seen a question that looked very much like a 5e question turn out to be a custom system where the DM was combining 5e, 4e and Pathfinder rules which means that any answer would have been useless that had guessed before. I've also seen questions where OP was confused about the system and using the wrong resources.

Thus, to me, guessing serves only to create confusion and disincentivize the proper use of our systems here.

Past actions

For the last several instances that I've seen, the answer has gotten deleted either by a diamond mod or by the community. This is sometimes combined with heavy downvoting. I can't recall specific examples right now, but I know in the past some answers have simply been downvoted and left undeleted.

What is the consensus from the entire community about how these answers should be handled?

\$\endgroup\$
1
  • \$\begingroup\$ ♦ Marked as dupe to current guidance to help redirect users (either who missed the change or found this before the FAQ) \$\endgroup\$
    – Someone_Evil Mod
    Commented Nov 26, 2022 at 22:29

4 Answers 4

Reset to default
11
\$\begingroup\$

Don't answer questions that need a system but do not specify one

Guessing the system leads to confusion in several ways. Since many users that have this issue are new there is the issue that in addition to not knowing how the site works and having their question put on hold they have no idea why their question has only one (or a few) answers that are highly downvoted.

Sometimes, the asker doesn't even know what system they are asking about or they are using the wrong materials and give the impression they are using one edition but are using others. We've even had a case where the asker was playing a homebrew system that was a combination of 5e, 4e, and Pathfinder. For all these cases, guessing the system in an answer is going to result in an answer that could be useless to OP. However, since they could be confused from the get-go it may also give them an explicitly wrong answer under the guise of being an expert answer to their problem.

Additionally, it is often a waste of effort to answer these questions anyways since OP seems to be very likely to not come back and ever check on/update the question.

Finally, this is about training OP in the proper way to use this site. If they get an answer that may seem useful to them (even if it might not be) they may have no incentive to update their question to meet site guidelines. This could leave us with orphaned questions with useless answers and a confused and un-helped OP.

Wait for OP to clarify and then answer and we should have none of these issues.

Downvote & Delete (and comment)

Simply downvoting is not enough in this case. It leaves a confusing and potentially wrong answer up which might not even answer the question at all. Being a new user, they might not understand or care about what downvoting means on this site and accept the answer as the best answer they are going to get regardless. We should work to avoid this situation.

Deleting the answers makes sense because it removes all the confusion for the asker and forces them to address the issue with their question without the distraction of potentially incorrect answers to muddy everything up.

When an answer like this is seen, the best thing to do would be to downvote immediately and then those users with enough rep to delete should vote to do so as well. A comment should be left explaining our policy and encouraging self-deletion, though we should not wait for the user to do it themselves.

\$\endgroup\$
3
  • 2
    \$\begingroup\$ As a general rule mods decline “NAA” flags if the post is at all attempting to answer. (That’s the guidance on our side of a NAA flag: it’s for gibberish or chat or other categorically not-even-trying-to-answer posts.) For material that is trying to answer, but still needs removing for some reason, a custom flag is useful to explain the situation. \$\endgroup\$
    – SevenSidedDie Mod
    Commented Sep 10, 2018 at 14:44
  • 1
    \$\begingroup\$ @SevenSidedDie good point. I thought about that as I wrote it which is kind of why I leaned on the previous guidance for general answer to specific questions. However, I can see that logic is strained (and I'm not so sure that there was ever a super-consensus behind that idea anyways). So I'll probably just remove that section. \$\endgroup\$
    – Rubiksmoose
    Commented Sep 10, 2018 at 14:48
  • 4
    \$\begingroup\$ Answers that are for the wrong game are one of the exceptions that has built up over time into a recognised exception. NAA isn’t a perfect fit for those, but it’s close enough that such flags “work” in practice. In theory a custom flag for those is better, especially if it’s not obvious how it’s for the wrong game, but in practice most of the time it’s clear enough without a custom flag. \$\endgroup\$
    – SevenSidedDie Mod
    Commented Sep 10, 2018 at 14:54
8
\$\begingroup\$

Vote down and comment, and let the community deal with it.

It seems to me your question is a token of a broader type of problem. I would think we should, in general, discourage any answer that guesses at the meaning of a question. If the question so wants for clarity that one's answer must take the form "Well, if you mean X, then Y, or if you mean A, then B, or..." potentially ad nauseam, that's a bad question. By discouraging guess answers, we encourage better questions.

As for how to discourage guess answers, I think the answer is already baked into the Stack's voting system, as reflected in this answer to one of the meta questions you linked to. If the community as a whole sees a guess answer as not valuable, it will get voted down into the negatives and should be deleted by the community. Moreover, downvotes should at least help to communicate to the answerer that they ought to know better. Comments to the querent requesting clarification, and also to the answerer to hold off for said clarification, would likewise help.

There shouldn't be a need for a diamond mod to take action in these cases. Policing this sort of answer-quality issue is exactly what voting is for.

\$\endgroup\$
0
-4
\$\begingroup\$

I'm entirely new here, but considering that I have seen this particular thread by way of a question that was apparent enough what was being asked through context of the question itself I may have at least an outsider's perspective.

As with what happened in this question (note, accepted answer now also deleted), if OP selects a particular answer as correct that also specifically states or iterates what system is being referred to, then clearly OP was inquiring as to that particular system.

I have been here less than 6 hours already at time of posting and have noticed that mods (or whomever) are more than willing to edit titles of posts if they feel it could contribute to overall clarification of intent of question, and don't see a reason as to why rather than putting a question on hold just before or just after an answer and having everyone downvote until deletion is accepted as more reasonable than editing the title to properly reflect what is expected.

I get that one-off OPs that won't come back or won't even bother clicking accept on an answer can be troublesome, but if the effort is already being put in by the mods to edit posts then why not let ones that have already been asked and accepted as answered live on for future reference so the question may not need be asked again?

\$\endgroup\$
7
  • 3
    \$\begingroup\$ Just to clarify, most of the edits on RPG.SE are made not by diamond moderators but rather by regular users like you and me. (Note: This comment was made before I became a mod :P ) \$\endgroup\$
    – V2Blast
    Commented Apr 28, 2019 at 7:56
  • 4
    \$\begingroup\$ One is several big issues with this suggestion is that it assumes that OP understands what edition they are playing and that they know there are differences between editions. We get many many questions by posters who don't know this and thus then selecting an answer is not a good indicator. The best option for everyone is to wait and get OP to clarify. \$\endgroup\$
    – Rubiksmoose
    Commented Apr 28, 2019 at 14:14
  • 1
    \$\begingroup\$ @Rubiksmoose - Your preferred approach has that same problem. If the OP doesn't understand what edition they are playing, then how can you expect them to state the information they aren't even aware exists? As so often happens on this stack, you're making the perfect the enemy of the good. \$\endgroup\$
    – Dave Sherohman
    Commented Apr 29, 2019 at 8:02
  • 6
    \$\begingroup\$ @DaveSherohman the method I describe above has no such problem. We've worked many confused people through the process to find out via the comments. Often it's as easy as them texting the DM to find out. We've never had a case we couldn't work out when OP responded. If they don't know and can't find out, we literally cannot answer. We are not making perfect the enemy of good here. System information is essential to get right otherwise we cannot answer the question. If there's one thing we need to get right, this is it. \$\endgroup\$
    – Rubiksmoose
    Commented Apr 29, 2019 at 11:05
  • 1
    \$\begingroup\$ @DaveSherohman in other words, making them aware that the problem exists is actually one of the main goals of my proposal. And teaching users this has helped many questions and users. If you have a better idea for how to handle this, please do put it in an answer. \$\endgroup\$
    – Rubiksmoose
    Commented Apr 29, 2019 at 11:21
  • \$\begingroup\$ @Rubiksmoose - Done. Now the race is on to see whether my answer gets more downvotes than this one, as I openly reject the presumption that an imperfect answer is worse than no answer at all. \$\endgroup\$
    – Dave Sherohman
    Commented Apr 29, 2019 at 14:02
  • 2
    \$\begingroup\$ @DaveSherohman regardless of how it might end up getting received, I'm happy you have contributed an answer! It is good to hear and consider all viable ideas. \$\endgroup\$
    – Rubiksmoose
    Commented Apr 29, 2019 at 14:04
-9
\$\begingroup\$

If it appears to be a clear and useful answer, upvote it, even if the system guess ultimately turns out to be incorrect.

Such an answer can still be useful to those who use the guessed system, regardless of whether it is useful to the OP or not. Other posts on this meta site have made it clear that, in the collective mind of the RPG stack, the primary purpose of the stack is to produce answers will be useful in the future, not to merely answer the question which was asked.

Regardless of whether you are the OP or a future reader, a blank space containing no answers is less useful than an incomplete or otherwise imperfect answer, so long as the imperfect answer is not directly misleading.

\$\endgroup\$
10
  • 4
    \$\begingroup\$ “the primary purpose of the stack is to produce answers will be useful in the future, not to merely answer the question which was asked.” It's both at once: provide answers to the question asked so that future visitors with the same question will find answers. Answers must always answer the actual question asked, meaning providing it should be appropriate to the game and edition asked about, otherwise it is considered Not An Answer and removed. Meanwhile querents are responsible solely for asking a clear question, including specifying game+edition when necessary. \$\endgroup\$
    – doppelgreener Mod
    Commented Apr 29, 2019 at 14:20
  • 3
    \$\begingroup\$ In other words, a question about how combat works in D&D 5e should collect exclusively only answers explaining how combat works in D&D 5e. There should be no answers about how combat works in other editions or other games, nor about how ice cream (but not combat) works in D&D 5e, because they would be Not An Answer to how combat works in D&D 5e. You appear to be advocating upvoting those other answers because they may be useful answers to someone sometime about something, but they decrease the signal-to-noise ratio and make it harder to find the actual answers to the question. \$\endgroup\$
    – doppelgreener Mod
    Commented Apr 29, 2019 at 14:37
  • 3
    \$\begingroup\$ If the querent doesn't specify or know what game and edition their question is about then we put the question on hold and leave it alone until they do so. If you want to guess we'll probably have to clean it up later because it won't be an answer, and people will be upset by that, and we'll probably clean it up sooner than that to encourage the querent to clarify their question. Do yourself and the community a favour and don't guess at the system. \$\endgroup\$
    – doppelgreener Mod
    Commented Apr 29, 2019 at 14:40
  • 1
    \$\begingroup\$ Once Game A is specified, an answer for Game B is not even minimally useful to future people: it’s filed under the wrong game and won’t be found by people looking for it. Yes, the point of the site is to be useful into the future, but you can’t forget the “well ordered database” part of that mission. \$\endgroup\$
    – SevenSidedDie
    Commented Apr 29, 2019 at 16:44
  • \$\begingroup\$ ...and all the criticism completely ignores the point that an imperfect answer is better than a complete absence of answers. "We will give you no answers because the question is incomplete" will help absolutely no one. "The question is incomplete, but here's the best we can do at the moment" may (or may not, true) help someone, either the OP or another reader. I am continually baffled by this stack's insistence on making the perfect the enemy of the good. \$\endgroup\$
    – Dave Sherohman
    Commented Apr 30, 2019 at 8:31
  • 2
    \$\begingroup\$ We accept and welcome imperfect answers all the time. Answers which guess the system are not merely imperfect, they are critically flawed and useless and confusing to everyone. This site's mechanisms are made to filter out bad, useless answers and to encourage good questions. Allowing these types of answers does neither. That's why we are able and encouraged to close questions that are unclear or otherwise flawed. And a question missing key, essential information is the definition of unclear. We don't want answers on any unclear question, not just ones missing a system. \$\endgroup\$
    – Rubiksmoose
    Commented Apr 30, 2019 at 11:22
  • 2
    \$\begingroup\$ @DaveSherohman The difference is we don't think “unclear question got random guessing answers” is a good outcome; we think it's a bad outcome. We also optimize for pearls, not sand, which means we're seeking good questions that will produce good answers, not poor questions which will produce poor answers. This in turn means we reject a situation of taking unclear questions and throwing in random guesses at answers because we feel it makes our site worse, not better. \$\endgroup\$
    – doppelgreener Mod
    Commented Apr 30, 2019 at 12:36
  • 2
    \$\begingroup\$ No answer is actually better than a bad, upvoted answer. An upvoted answer tells the site that the question is maybe done, so it takes it off the “unanswered” list and reduces its visibility. This criticism is not ignoring it, it’s actively disagreeing that your suggestion has desirable outcomes. \$\endgroup\$
    – SevenSidedDie
    Commented May 1, 2019 at 23:27
  • \$\begingroup\$ Note that my suggestion included "clear and useful". If the question asks, "How does a to-hit roll in D&D work?", then one person can answer "In 5e D&D, it works like this..." while another answers "In 3e D&D, it works like this...". Even if the OP meant 4e, both answers are clear about which edition they apply to and are useful to anyone who looks in the future for how to hit rolls in D&D work and are playing those editions, despite the OP's flaws in formulating the question. I remain unconvinced that this is somehow worse than having no information on to hit rolls at all. \$\endgroup\$
    – Dave Sherohman
    Commented May 2, 2019 at 7:04
  • 2
    \$\begingroup\$ @DaveSherohman Because it interferes with fixing a question that needs closing, because they don’t need to fix it to get answers. Because it potentially misinforms an asker who can’t tell a wrong from right answer. Because it is contrary to the basic point of the site. Because it encourages asking bad questions, on the hope that a fast answer will be faster than the close. Because it encourages posting answers to unclear questions. All these things harm the mission of the site. Basically all your reasons for rewarding bad answers with votes/reputation are contrary to what votes exists for. \$\endgroup\$
    – SevenSidedDie
    Commented May 3, 2019 at 3:18

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