33
\$\begingroup\$

We occasionally get questions which don’t specify which game system they are about, and which are about specific aspects of that game. Because the context and answers depend and vary with the system, and multiple systems include content by the same name, this means the system needs to be clarified for that system. Current praxis and policy is that the clarification must be made by the asker (in title, body, tag, comment, etc.).

This is first solidified here, upheld in 2018 and 2020; the policy as currently understood is summarized in this Q&A. This policy has been referred to in a few different ways, but for simplicity’s sake, let’s call it the Don’t Guess the System (DGtS) policy.

However, since the last revisit (in 2020) was active and was considered settled, ongoing votes on the answers changed the apparent consensus. (There’s discussion specific to that phenomenon here.) As a result, a revisit of the DGtS policy is warranted.

For everyone’s benefit, I’ll try to give a rundown of the other changes that have happened since the last revisit:

Following the discussion at How shall we frame our upcoming revisit of the Don't Guess the System Policy?, this question will be closed for one week. This will prevent answers from being posted during that time, allowing all answers to be ready to post around the same time when the question is reopened. (While the answer to the linked Q&A says to lock the post, none of the lock options prevent posting answers without also preventing voting and commenting on the question. This seems like a better compromise.) We’ll give people 1 month after the question is reopened, then take a look at the votes on the answers here. If there’s a clear winner and voting appears to have stabilized, we’ll proceed with that course of action; otherwise, we will re-add the tag and look to see if there are any other actions we need to take.

With all that out of the way, how should we proceed with regard to the DGtS policy?

  • Should we keep the overall policy as is?
  • Should we replace the policy with less absolute guidance? If so, what should this be?
  • Some kind of middle-ground approach, or something else not covered here?

(Don’t feel constrained to the listed options; we’re trying to be inclusive of every possible approach.)


Results as of September 3rd, 2021.

One month after opening this post to answers, Thomas Markov's answer is the top answer at +62/-5. It advocates for "Let experts make expert decisions", removing the policy and creating new guidance for users.

The other two answers (Pyrotechnical +29/-6 and nvoigt +30/-11) also advocate for removing the policy.

With no answers opposing the change and very positive votes in favour of the proposals this will form our policy (or lack thereof) moving forward.

Where to from here?

There are a number of posts that will require modifications, cleaning up or historical locks. The mod team have already started identifying these and will work to update them in accordance with this change. Give us time to make the changes, but if you feel we've missed something feel free to flag it with a custom message.

The community will need a new FAQ guidance question on how and when to make appropriate system edits. "Equip our experts with some community crafted guidance" as Thomas Markov puts it. This should be a collaborative effort to create the best guidance possible.

The guidance Q&A has been posted and it is ready for your input: What should I do when a question does not specify the game system being used?

\$\endgroup\$
17
  • 10
    \$\begingroup\$ I’m voting to close this question because we've decided to wait a week before allowing answers. \$\endgroup\$
    – Someone_Evil Mod
    Jul 24 at 17:38
  • 1
    \$\begingroup\$ This doesn't show up in the Featured on Meta section for me, despite having the featured tag - possibly because it's closed? \$\endgroup\$
    – A_S00
    Jul 26 at 15:16
  • 6
    \$\begingroup\$ @A_S00 <screams internally> Thanks for pointing that out. That's probably the reason. Which is pretty frustrating, but I don't know what to do about that now. We had even talked about this internally, but just assumed it would still be featured because why would closing the question stop it being featured? On the other hand, why would you want to feature a closed post normally? I'm not sure what's the best approach here, but I'll welcome bright ideas \$\endgroup\$
    – Someone_Evil Mod
    Jul 26 at 15:31
  • 9
    \$\begingroup\$ Ok, I've created a community event which should achieve the same end as featuring it. (Except it displays above the MSE posts, hah!) \$\endgroup\$
    – Someone_Evil Mod
    Jul 26 at 17:34
  • 3
    \$\begingroup\$ And, with a week passed, reopened the question. Let the answering commence! \$\endgroup\$
    – Someone_Evil Mod
    Jul 31 at 18:14
  • 9
  • \$\begingroup\$ "The community will need a new FAQ guidance question," should this be made by the mod team? I wanted to propose my answer to form the basis of that guidance. \$\endgroup\$ Sep 3 at 13:12
  • 2
    \$\begingroup\$ @Pyrotechnical There's no reason it needs to. Arguably, if it's to be community crafted it shouldn't be. If you have thoughts and words to get that Q&A post started, please do so! \$\endgroup\$
    – Someone_Evil Mod
    Sep 3 at 13:20
  • \$\begingroup\$ @Pyrotechnical I plan to get the conversation started next week, though I may be able to sit down tomorrow and gets something written out. \$\endgroup\$ Sep 3 at 13:43
  • \$\begingroup\$ With the question solved for the forseeable future, we should lock the question to preserve its status for posterity. \$\endgroup\$
    – Trish
    Sep 4 at 10:41
  • 1
    \$\begingroup\$ @Trish No. Longitudinal voting changes are one of the key metrics for evaluating the community’s thoughts about the policy. \$\endgroup\$ Sep 5 at 0:29
  • 1
    \$\begingroup\$ @Pyrotechnical I guess we could just loosen up those votes with an edit? \$\endgroup\$ Sep 7 at 18:40
  • 1
    \$\begingroup\$ @Pyrotechnical We have already seen the utility of a voting swing with the last DGtS iteration. Doppelgreener gives more commentary on that here. Why would we abandon such a useful metric of community opinion for the sake of being procedurally correct in how we use the edit button? \$\endgroup\$ Sep 7 at 18:51
  • 1
    \$\begingroup\$ It's worth noting that the vote totals from when this decision was made are actually in the question itself and preserved there (you can also use the answer timeline to find their totals on any other day). I'll also point out that previous revisits are not (and were not) locked \$\endgroup\$
    – Medix2
    Sep 7 at 18:53
  • 1
    \$\begingroup\$ Thomas's new guidance post (which I see is also now linked in the question): What should I do when a question does not specify the game system being used? \$\endgroup\$
    – V2Blast StaffMod
    Sep 8 at 18:56
59
\$\begingroup\$

Let experts make expert decisions.

This is the essential premise of Stack Exchange. People come here because they expect experts to be equipped to solve their problem - and they are right. We have an active community of RPG experts engaging with questions to provide expert guidance for solving a broad range of RPG problems. Our experts do not need a rigid policy to govern their decision making.

We should not artificially obstruct the problem solving process.

Our primary focus and objective is to provide our expertise for solving the problems presented to us. Using the site and learning how to most clearly communicate questions should be a secondary, or even tertiary, objective of ours. By saying "we cannot answer your question until you learn how to use tags, even though everyone knows what system you're playing", we are putting problem solving in the back seat and setting site mechanics as our primary objective for new users. We should never tell a new user that our problem with their lack of tags or explicit forthrightness about the system is more important than the problem they bring to the table. But that is exactly what we are doing when we know what they are talking about but artificially hold up the problem solving process because they didn't explicitly state what system they are playing.

We can teach users how best to ask questions without artificially obstructing our primary objective.

We can teach users how to use the site and ask better questions without putting problem solving on hold. This should be obvious. When we know what a user is asking about, based on our expertise, we can both answer their question and teach them about the site. That is part of what comments are for. This keeps our objectives properly aligned. We focus on answering their question and secondarily engage with them on how to ask better questions and use site mechanics.

We are equipped to handle situations where the question really is unclear, or we incorrectly identify the system.

Site mechanics are already designed to handle these situations - we don't need a policy to prevent them.

Community curation is equipped to manage unclear questions.

The present application of the "Don't Guess the System" policy is to close questions where the system is not clear. This will continue to be standard praxis without the policy. This is what we should and will continue to do anyway. When a question is unclear, a system unidentified, our experts will recognize this and vote to close the question until we have sufficient clarity.

Readers decide if answers provide valuable information.

The small portion of times someone is wrong about their system guess will be sorted out by readers' votes. Even when it is unclear in a question what system is being played, answers can be made sufficiently clear to still be useful. If a question says "I'm playing 5e" and it could be D&D or Shadowrun, answers can easily state which game the answer pertains to. If OP never clarifies, both answers provide valuable information to future readers, and community voting will reflect the value provided by these answers. To be clear, we should still close this totally hypothetical question that is equally likely to be two entirely different games for “needs details or clarity”, but if we do get answers before it is closed, we can salvage the usefulness of the answers by appropriately pointing out the answer’s attendant game system.

Community curation is equipped to handle obsolete answers.

Now, suppose we do get the system wrong. If OP does come back to clarify, the answer for the wrong system will be downvoted or deleted. This is just the site functioning as intended. When an answer is given, but the question's author clarifies that they are playing a different system, we simply flag that answer NAA, leave a comment to the effect of "OP has clarified which system they are using, so this answer does not answer the question being asked", and remove the answer. This is just fine, and for the answer's author, this is just the cost of doing business with an unclear question.

Policy exceptions have already eroded away some of the contentious applications of the existing policy.

The most contentious use of the policy is in closing questions where the system is abundantly obvious, but the name of the system isn't actually stated explicitly. We often see objections taking the form of "Well it's not a guess". To remedy this less-than-intuitive application of the policy, the community has already carved out exceptions to the policy to make room for these common sense edits:

This first bullet in particular carves out a large exception space to the existing policy that was the source of the most discord in my experience. Closing questions with direct quotes and citations of system specific materials but without an explicit system statement in the question body was the most contentious application of the policy, but we've gotten rid of it already.

The space that remains for applying the policy is mostly the same space where questions would be closed anyway without needing a policy. The policy remains in effect for those truly grey areas, and I have already explained that we are equipped to handle those with existing tools, without needing the policy to motivate us. I say “mostly the same space” because there is still some space for questions that are abundantly obvious, that just don’t fit into our exception spaces yet. But if community direction tells us anything, it is that if there is a way to codify such a question into an exception, we will and the community will support it. We don’t need a policy to motivate our curation anymore. Let’s let our experts make expert decisions about content and curation.

Let’s further equip our experts with some community crafted guidance.

If I haven’t made it clear yet, let me reemphasize: wild guesses and stabs in the dark should be avoided. The current state of the policy prevents this, but it does so with a heavy hand by treating wild guesses and well informed decisions as the same thing, when the question’s context is underspecified. Understandably, this is the primary source of contention with the policy. So in the event of a removal of the policy, we can and should craft some guidance to nudge users in the right direction when approaching untagged or under-specified questions. Some of our exception metas can serve as a good starting point for that guidance, as these metas exist because they carve out exceptions to the current policy for situations where we can be exceedingly confident of the system being used. More guidance is needed, but this post isn’t the place to explore that in detail. Should this proposal succeed, we should work together to construct an FAQ-style meta Q&A with some guidance for handling underspecified questions.

Conclusion

To be transparent, the problem I really want to solve here is the counterintuitive closure of questions where the system is obvious, but not explicit. To this end, I think the proposal outlined above will serve two purposes:

  • It will eliminate counterintuitive question closures.
  • It will mostly preserve the favorable outcomes of having the policy.

I really believe that repealing the policy, establishing some concrete guidance, and letting the community handle question curation will yield a state of affairs not dissimilar to what we have now, but without counterintuitive closures. Under the policy, ambiguous questions get closed. Under this proposal to repeal the policy, ambiguous questions will still get closed.


Acknowledgements

I would like to thank KorvinStarmast and doppelgreener for their contributions and improvements to this answer.

\$\endgroup\$
6
  • 5
    \$\begingroup\$ The only part I might disagree with is this: "Even when it is unclear in a question what system is being played, answers can be made sufficiently clear to still be useful. If a question says "I'm playing 5e" and it could be D&D or Shadowrun, answers can easily state which game the answer pertains to." If a question really is unclear enough that users have no idea what system it's about (though I suspect that that situation is very rare), it should probably still be closed as unclear (until clarified) rather than arbitrarily guessing one over the other. Besides that, I agree with your proposal. \$\endgroup\$
    – V2Blast StaffMod
    Aug 2 at 2:44
  • 5
    \$\begingroup\$ @V2Blast Right, my point there is less “we can still answer those it’s okay” and more “if these do get answered before they get closed we can salvage some usefulness out of the answers by labeling them appropriately”. \$\endgroup\$ Aug 2 at 4:42
  • \$\begingroup\$ @V2Blast I added a bit to that section, is it more agreeable to you now? \$\endgroup\$ Aug 2 at 11:52
  • 1
    \$\begingroup\$ Looks better now. (I think that attempts to answer a question that should be closed are still likely to get downvoted and possibly deleted, just to discourage attempting to answer unclear questions and the like, but that's not unique to DGtS.) \$\endgroup\$
    – V2Blast StaffMod
    Aug 2 at 18:37
  • 1
    \$\begingroup\$ Agreed. In same cases, the system can be very obvious (such as a question that just lists D&D but mentions advantage (must be 5e) or specific feats/books/etc. by name or abbreviation (must be the edition those items belong to), or a question that lists Pathfinder but mentions 2/3 casters (must be PF1) or ancestry feats (must be PF2), etc.)... \$\endgroup\$ Aug 10 at 22:35
  • 4
    \$\begingroup\$ ...It would be absurd to say that because a system is not explicitly stated by name, the question cannot be answered due to site rules, even when the system can be explicitly determined; that's the point where it changes from being about uncertainty, to punishing everyone for one person's improper tagging. \$\endgroup\$ Aug 10 at 22:35
24
\$\begingroup\$

Recommendation: Rescind the policy - Guessing the system is permissible, but you do so at-risk

Going forward, we will adopt a policy to allow experts to be experts, but in that vein, we will expect that they be correct. This recommendation puts the burden of guessing upon the respondent:

  • Don't edit the question tags: Do not edit tags into the question unless there are clear indicators on the querent's intent via comments, language in the question (including things like a link to D&D Beyond), stating the system in the title, etc. This is no different than prior.
  • State your assumptions: When guessing the system, it is important to indicate what your assumption is within your answer, so a header stating, "I'm assuming you're referencing 5th edition," would be adequate before proceeding with your answer.
  • Accepted answer: Assuming the querent accepts your answer, but does not edit the assumed system into their question, then it becomes appropriate to edit the original question's tags to match the assumption.

The at-risk element:

I do not anticipate that this will occur very often, but in the event that the assumed system is incorrect, the answer must be deleted REGARDLESS OF SCORE. This is necessary to prevent confusion for others who may find the question at a later date.

An answer is considered wrong if the querent has edited in the system and it doesn't match the assumed system.

Deletion can occur via any of the following methods:

  • Deletion by the respondent is preferred, but not always viable.
  • If the score on the answer is negative, deletion can be handled by the community.
  • If the score on the answer is positive, deletion must be handled by the moderators. Members of the community can assist by raising a the, "In need of moderator intervention," flag option and citing that the assumed system was not correct.

Overall, this proposal creates a slight increase in the moderator's workload, but I don't get the impression this will be more than a handful of additional tasks for them every year.


The need for this has mostly passed

This is somewhat speculative history, but if my understanding is correct, the original need for the DGtS policy stemmed from the release of 5th edition D&D, which continuously had answers to questions that didn't align with system edition. This was likely further complicated due to 5th edition's extended playtest period whereby it was attracting folks who were playing 4th edition and enjoyed the strong balance of that system coupled with folks coming from 3.5 edition who enjoyed the strong thematic elements of the rules.

This situation was unique in gaming, because when 4th edition came out, a lot of folks used to 3.5 were adamant about not playing it describing it as being too much like an MMO. Conversely, 4th edition brought new folks to tabletop gaming in part because it was very similar in nature to the MMOs that they enjoyed playing. Now with 5th edition, coming out, these two groups are engaging with each other and may've been oblivious to the other's viewpoint prior to this point in time.

As a result, a question about Color Spray could have serious overlap with no less than 3 different systems and all are using similar terminology. Furthermore, because 5th edition was new, a lot of these questions were coming in very quickly.

Stack Exchange has previously indicated that they wish to maintain a focus upon keeping things easy for the experts. To that end, there are baseline rules whereby a querent should indicate the research effort that they have already engaged in prior to seeking assistance. The underlying basis for this theory is that there are endless half-considered questions, but the number of experts whom will be willing and/or able to answer those questions is likely far smaller in number. Thus, it is prudent to cater to those experts because once they leave the site, that's when the stack dies.

I had thought that this might also happen with the release of Pathfinder 2e, which is why a few months back I was supportive of maintaining the policy, however, this premonition does not appear to have occurred. Given this doesn't seem to be repeating issue, I am inclined to rescind the policy.

\$\endgroup\$
1
  • 3
    \$\begingroup\$ I was considering making a separate answer proposing a concrete workflow like this, but you've got everything I had plus even more thoughtful guidance. Thank you for your contribution. \$\endgroup\$ Aug 2 at 20:00
19
\$\begingroup\$

I present exhibit A:

Do Kaladesh Dwarves get +4 in Constitution?

The most recent example of how we make this site a worse place by having this rule.

This question was not unclear. It could be answered. And yet, we deleted the answer (don't know what actually happened, but our community acted in a way that in the end as a result we had a deleted answer). Why? Was it wrong? Misleading? Guesswork? No. Simply because we have "the rule". Regardless of whether "the rule" makes sense here or not, it apparently has to be enforced.

We are here to answer people's questions, not send them away because they made a mistake when filling out the right set of bureaucracy approved forms.

I can understand that people vote to close as unclear, when it's actually unclear. I can understand people not answering when they don't really know if their answer will be correct. I appreciate people asking for clarification. But deleting a perfectly fine answer and closing the question, because of some rule we made up, is really horrible.

If someone asks how hitting an Orc with their human fighter works, then yes, please, close it as unclear, because it is. D&D, DSA, Earthdawn, Warhammer, 40K, Shadowrun, I could find tons of RPGs that would fit that description.

But can anybody point me to at least one other RPG that has a Kaladesh dwarf subrace?

One?

We alienated two good people, just because we made up a rule and enforced it with no (or very few) exceptions. This nonsense has to stop. If someone has a question and someone else has an answer to it, I do not want to see people removing knowledge from this site in the name of "rules".

\$\endgroup\$
31
  • 9
    \$\begingroup\$ While I realize I’m biased, it seems to me that this is a really terrible choice of example: the OP of that question has literally not been logged into the site since asking it. They couldn’t have been alienated by the question being closed, because they never saw it get closed. They asked a question and left, apparently never to return. Which is really an excellent case-in-point for why we should not be expending effort and energy on this question and others like it. (Also, yes, this post is insulting to those who disagree with you, and your arguments to the contrary are disingenuous.) \$\endgroup\$
    – KRyan
    Aug 1 at 0:27
  • 11
    \$\begingroup\$ I don't even know how to reply to that. If you think that a question asked here deserves it's answer to be deleted because the OP does not respond to questions or logs in, then I cannot find a common ground with you. That is so far from the standard SE model, I don't even know how to discuss this with you, sorry. \$\endgroup\$
    – nvoigt
    Aug 1 at 11:51
  • 3
    \$\begingroup\$ To ask the obvious question, what are you advocating for in this answer? This seems more like a rant than a solution. \$\endgroup\$
    – NautArch
    Aug 1 at 13:34
  • 7
    \$\begingroup\$ I'm advocating for dropping this rule completely. There is no need for it. If a question is unclear on it's own, because it's ambiguous and tags are missing, close it. We don't need an artificial rule to close it and delete valid answers, if the question is indeed not ambiguous at all like in this case. \$\endgroup\$
    – nvoigt
    Aug 1 at 13:38
  • 4
    \$\begingroup\$ @nvoigt About the answer deletion specifically, a question is closed because it shouldn't be answered in its current state, which sometimes means removing well-meaning and probably applicable answer (because that shouldn't have been posted in the first place). And we want this to avoid the messes of dealing with invalidated answers from bad assumptions. Now, if you want to argue that those are rare enough around system statements and not worth the cost of preventing them (ie. DGtS), please do that. \$\endgroup\$
    – Someone_Evil Mod
    Aug 1 at 13:40
  • 5
    \$\begingroup\$ But would you agree that it is not ambiguous at all and there is no other Kaladesh Dwarf Race? If we did not have this rule, there would be no reason to close or delete anything. We did not close because it's unclear, we closed because of "the rule". Without the rule, it could been a pretty little "someone asked, someone answered, knowledge was built up" event, helping the OP an everybody in the future googling for it. \$\endgroup\$
    – nvoigt
    Aug 1 at 13:49
  • 3
    \$\begingroup\$ Also with the implicitness (albeit fairly clear one) of the Abolish the Rule stance, there's something of an issue. I think much because of how long we have operated with this rule, it would be very helpful to clarify what you think we should return to as an operating mode. I would assume the Treat it Like any other edit, but I'd rather it be clear what exactly this answer is advocating. Because I don't think any of us want to go out of this and discover we don't agree on what we've agreed on. \$\endgroup\$
    – Someone_Evil Mod
    Aug 1 at 14:00
  • 9
    \$\begingroup\$ I'm not sure where the problem without the rule actually is. The same problem statement exists across all sites of the network. For example on The Workplace, people omit their locality and it becomes almost impossible to craft a good answer until the OP clarified where they are and which labor laws apply. But that's it. We don't have rules for it. If it's unclear, you may get a few, generic answers of no value, and if someone just makes a wild guess they get commented on and downvoted. I'm not even sure what problem the rule fixes, so I don't know what would be confusing about abolishing it. \$\endgroup\$
    – nvoigt
    Aug 1 at 14:12
  • 9
    \$\begingroup\$ @ThomasMarkov I reviewed all of them and found not a single factual example of harm prevented by that rule, or harm caused when the rule was not in effect. It's all theories about how it would go all horribly wrong. Meanwhile other SE sites share the same problem, whether it's location in the workplace or prgramming language on SO or which timeframe someone is talking about in History. People ask for clarification. People vote to close as unclear. Nobody would ever suggest to delete a correct answer. \$\endgroup\$
    – nvoigt
    Aug 1 at 15:50
  • 4
    \$\begingroup\$ @nvoigt So what you're suggesting is not that the question be fixed and properly categorized, but just allow answers to guess at which system it's about? That seems kinda awful from a curation PoV and I don't think everyone would understand that from just getting rid of the rule. \$\endgroup\$
    – Someone_Evil Mod
    Aug 1 at 16:05
  • 6
    \$\begingroup\$ Well, it does not happen on other SE sites, do you think our community is so awful that we would drown in bad guesswork? Because other sites manage just fine without this rule. And I have not seen any answers being mod deleted upon closure there either. Since this site wasn't conceived with this rule in place, can you link me to some actual events that took place? I have linked an actual event and I have the feeling I have to argue against fictional future events that everybody can just twist and turn and imagine different until they fit their own narrative. \$\endgroup\$
    – nvoigt
    Aug 1 at 16:45
  • 4
    \$\begingroup\$ @nvoigt I'm pretty sure other stacks have similar rules to deal with specific quality issues. IIRC, arqade requires an image or similar for game id, SO needs questions to include code. The different stacks deal with different questions, different problems with those questions and thus end up with different practices and rules. And I strongly object to the categorization or implication that I think the community awful. I'm having a real issue taking that as a good faith argument. \$\endgroup\$
    – Someone_Evil Mod
    Aug 1 at 18:22
  • 3
    \$\begingroup\$ As for examples of things. The real ambiguous cases the rule was established for would be back in the 3.PF days. That predates my time here solidly, so someone else would have to help you with that. As for guessing, we absolutely do get answers which guess (implicitly) the system and get it dead wrong. The cases I can point would primarily be answers answering as though it's 5e for questions that clearly say they're a different system. I absolutely accept that it's rare and handleable, but I want to see that made and not just dismissed because that isn't useful. \$\endgroup\$
    – Someone_Evil Mod
    Aug 1 at 18:28
  • 5
    \$\begingroup\$ @Someone_Evil yes, what I see on SO are calls for code inclusion and, where necessary, calls to specify what language is used. What I don't see is the sort of policy-driven, rules-based, methodical insistence that all questions pass a particular muster, or they get closed. A good close analogy would be python 2 vs python 3 questions, which aren't always obvious by inspection, where it does sometimes matter a lot, and which don't often specify. And yet I still don't see a bunch of closures. \$\endgroup\$
    – Novak
    Aug 1 at 20:45
  • 11
    \$\begingroup\$ There is a lot of better content here in the comments regarding what you're trying to say, nvoigt. I'd very much recommend editing your answer with what you'd like to see and why. \$\endgroup\$
    – NautArch
    Aug 2 at 14:00

You must log in to answer this question.

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