12
\$\begingroup\$

Both of them has a common theme, "how can I get the most of these characters, given these boundary conditions", but the treatment is very different:

Both questions provide a frame to work in, like classes, levels and races.
Neither asks for an easily comparable metric, like DPR.
The closed one was from a beginner, the other from a valued veteran1.
Both answers lack any calculations (because the questions were not about DPR, and other things are hard to calculate), so can be deemed opinion-based.

I like answering this type of question very much, I would like to know how to guide the querents to an unclosed question. Without getting too specific on details2.


1) I really hope this is not the main reason for the different treatment.
2) "Best ranged weapon using Bard archetype" is a helpful question for many people, while "Best ranged weapon using Bard archetype for a 7th level full Dragonborn pary" is probably just for one.

\$\endgroup\$
9
\$\begingroup\$

They’re only superficially similar

The two questions are only really similar is asking for build advice in their current versions. They’re different in ways the Stack cares about:

  • The one that was closed needed editing and was closed for good reason.

  • The scope of what askers (and voters) are being asked to consider is dissimilar

And if the closed question is suitable for the site now, it just hadn’t been long enough to get reopened.

It was correctly closed as Primarily Opinion-based

The original question literally asked for opinions and was correctly closed as such. One of the five votes was for it being unclear, and comments asked for clarification.

The other question was well-asked from the beginning.

The scope of the closed question is a drag on its reopen chances

The following is an analysis of the psychology of the reader (i.e., potential voters and potential answer-writers), not precise analysis of scope complexity. Few readers will do more than a rough intuitive estimate of how complex the scope is — which as we’ll see, is the point.

So now it’s having a bit of a hard time getting reopened due to, it seems to me, its larger area of possibilities for potential answer-writers to grapple with before coming up with a good answer, or even considering whether to answer. It’s asking for which archetype to pick for an unknown future up to 20th, which is 18 levels + feats or ASIs that answer-writers have to at least briefly account for for each archetype option (else someone is going to come along in comments and say “no, they shouldn’t pick Elephant Mesmerist, it’s a terrible choice because at level 17 it can only hypnotize two elephants, when Trapeez Artist can befriend three by 12th”), with optimised interactions with two other unknown-future PCs. The answers aren’t even being asked for a full build, just the best archetype, but they still need to think about this.

A full analysis might be necessary to answer the question, or it might not. It’s not required for an answer to be on-topic, but it might be needed for a “best” answer. And the thief could potentially multiclass at any level, exploding the possible full-level builds answer writers might consider before recommending just an archetype.

That doesn’t make it necessarily too broad — a lot of those possibilities are useless branches that can quickly be pruned — but analysis paralysis is real and this question asks for answer-writers to take on a lot of it before even deciding whether to answer. And same for voters deciding whether to reopen.

The linguist question by comparison is cut-and-dried: it is a given set of 19 levels with a wildcard 20th level. Its still a lot of work, but the scope is well-defined and not a matter of reader (or voter) judgement. Deciding whether it’s well-scoped is easy for voters. It had way, way fewer future unknowns for readers to analyse before deciding “I can answer this” or “this should stay open”. Its easy to judge.

And all that is stuff that answer writers probably don’t need to put in their answers, but maybe need to work out to make their answer best. It’s a lot of up-front work even before starting to write. Or voting.

The point, again, is not whether I think it should be open or closed. (Especially since you didn’t ask that; you asked why they’ve been treated differently.) The point is that analysis paralysis appears to be slowing its reopen votes.

There’s no one-size-fits-all help you can give

Every optimisation question is different. Useful advice depends on the unique circumstances — not just for what it’s trying to do, but as a question here on a Q&A site. Similar questions may be very different Stack-wise.

To help a held or closed charop question with advice, pay attention to the circumstances of its closing. Look at the edit history and comments. Leave a comment asking what’s wrong, if nobody has said. Bring it to meta asking what’s wrong with it and how it can be fixed, and assume good faith, rather than assume it has been closed wrongly.

After all, closing is normal here and even when it should be reopened, it’s just normal process in a vote-based system, not a flaw.

\$\endgroup\$
  • \$\begingroup\$ There are 14 (potentially 15) levels of Ranger without a set archetype, one of the tasks was to pick one. It is not much different from 17 levels of Rogue archetypes. And you had to pick feats, and a level that could go to any class. So the Linguist was actually broader, if you compare answer size. \$\endgroup\$ – András Jan 8 at 16:24
  • 5
    \$\begingroup\$ That’s not quite the point I’m making. Let me revise the post. Also: that’s only half of what I said was the problem. \$\endgroup\$ – SevenSidedDie Jan 8 at 16:47
4
\$\begingroup\$

The Monk question originally straight-up asked for opinions and that is what 4/5 votes to close were for.

Do you guys think its worth it? Should I just go straight Monk since I don't get Sneak Attack, and plan on getting Mobility anyway? Would it be better to just go with Swashbuckler and skip the Mobility feat?

That was 100% valid. The Max-languages question had no such issue.

The only way the questions are similar are that they are optimization questions. One was a well-formed question, the other wasn't and was correctly closed. Simple as that.

Side note: You can never have too many details in an optimization question

I did want to address something you said though:

I would like to know how to guide the querents to an unclosed question. Without getting too specific on details2. [...]

2) "Best ranged weapon using Bard archetype" is a helpful question for many people, while "Best ranged weapon using Bard archetype for a 7th level full Dragonborn pary" is probably just for one.

The best way to help an optimization question to be better is to encourage OP to post as much information related to their goals and circumstances as they can provide. This meta is a good starting point at least.

In general, our goal is to solve the problem of the person asking the question. If that means that we need details such that it narrows the question to being only useful to that person, that is the best thing to do. Better to have a Q&A that solves OP's problem fully and completely but really is specific only to them then one which is broadly applicable but does not help them at all. We shouldn't try to keep questions broadly applicable if it hurts the question (nor should we unnecessarily and artificially make a question more specific).

Optimization questions are really tricky and require a lot of information to do correctly. In general, every detail that we can get helps us immensely because of the vast number of options available for builds. I've never seen one get "too specific" and I have no idea what that would even look like. I've seen many, many optimization questions that aren't specific enough (some of them are far from it).

Also worth noting that no matter how many details the question contains, answers are free to answer as broadly as they would like as long as they directly answer the question.

In short, we should never shy away from having a question get "too specific" in pursuit of getting OP the best answer to their problem and I don't think that such a thing is even possible insofar as it would ever become an issue.

\$\endgroup\$
  • \$\begingroup\$ I think I have demonstrated quite well in both answers that not a lot of information is required to do optimization correctly. \$\endgroup\$ – András Jan 8 at 16:14
  • 3
    \$\begingroup\$ @András I respectfully disagree. Optimization in general only benefits from having additional details and your answers do nothing to prove otherwise. \$\endgroup\$ – Rubiksmoose Jan 8 at 16:16
  • \$\begingroup\$ @András As a note, since we seem to disagree here you may want to take a look at this meta (that I wrote) and if you disagree with my conclusions feel free to comment or write an opposing answer. I wrote this as a way to help guide CharOp question askers to a good post so I appreciate input. \$\endgroup\$ – Rubiksmoose Jan 8 at 16:32
  • \$\begingroup\$ I upvoted it a long time ago, and I agree these make the answer shorter and easier. But not better. Leaving out more parts makes the answer useful for a broader audience. \$\endgroup\$ – András Jan 8 at 16:42
  • 2
    \$\begingroup\$ @András like I said in my answer above though, leaving questions as broader question just to be more generally applicable is not what we do. We do the best for the person asking the question and their problem regardless of how broad or narrow it makes the question. The number of people the question is useful to is simply just not a factor in judging how good a question is for answerability purposes. \$\endgroup\$ – Rubiksmoose Jan 8 at 16:45
  • 2
    \$\begingroup\$ @András as an addendum do note that answerers are free to answer as broadly as they like. If they want to make an answer that applies to every build they are welcome to (as long as they also directly answer the question at hand). \$\endgroup\$ – Rubiksmoose Jan 8 at 16:48
  • 1
    \$\begingroup\$ @András It does make them better. The site is explicitly designed to have more questions that are concretely specific than fewer that are more general. We don’t artificially broaden questions by leaving details out because it objectively harms the site’s mission. \$\endgroup\$ – SevenSidedDie Jan 14 at 21:50
  • \$\begingroup\$ @SevenSidedDie and this is why we have about 30 questions concerning multiclassing and spell slots in 5e, combining different class names. \$\endgroup\$ – András Jan 14 at 22:58
  • \$\begingroup\$ @András That’s 1) an exaggeration, which kneecaps your argument, 2) only a problem if they’re actually the same question and not getting closed properly as duplicates. The fact is, you’re arguing that those should be closed, but you don’t get to decide that alone. You have to use your votes to collaborate, and accept it when the result of everyone’s input isn’t what you like. You’ve been told this before. If you don’t like how the site works, you’re free to leave. \$\endgroup\$ – SevenSidedDie Jan 15 at 0:26
  • \$\begingroup\$ @SevenSidedDie, 1) you made me count, I stopped at 30, not counting the closed ones. We should have 2, one for Warlocks in the mix, one without. 2) You are right, I cannot decide alone. That is why I am trying to convince all of you to go broader, and edit out at least the level from the questions. \$\endgroup\$ – András Jan 15 at 9:57
  • 1
    \$\begingroup\$ @András there are certainly a lot of multiclassing spell slot questions, I'll give you that. Perhaps we can open a meta on what to do about that (if anything). However, this case doesn't prove anything except (at most) we should have been more aggressive at closing the questions as duplicates. Forcing questions to go broader would not have any effect on that issue. We definitely should not be removing information from questions because that actively hurts our ability to answer the question as asked or figuring out if it's a duplicate. \$\endgroup\$ – Rubiksmoose Jan 15 at 12:20
4
\$\begingroup\$

Just as a bit of food for thought, you say,

The closed one was from a beginner, the other from a valued veteran1.

1) I really hope this is not the main reason for the different treatment.

and I think, to an extent and from a certain point of view, it’s a major reason, though certainly not the sole reason, or the main one.

But what I mean is, there is an element of voters looking for cues that a question will or won’t work, and one of those cues is whether or not an experienced user thinks it’s going to. So because it was my question, implicitly, I was vouching for it, and that seems to matter—I have seen similar effects when I have stood up for others’ questions that were garnering close votes.

And I think that’s fairly reasonable in a lot of cases. Certainly, no one around here is taking my opinion as gospel writ, nor should they be, but it matters when a veteran user champions a question, and I think that’s fair. It is difficult to adjudge whether or not certain questions are going to be problematic or not. Having someone whose judgment you trust and who stands up and say “I think this will be fine” can be a powerful reason to hesitate on clicking that Close button.

It’s also not as though I didn’t get push-back; at least one user questioned whether it was valid to have an optimization question without a distinct, measurable quantity to optimize. But I was able to address that point, and I was able to lend my expertise with this site to my claim that it would be fine. If pressed, I could have cited other questions that were similar and had worked.

Ultimately, though, I’ve had questions closed. I have had questions I argued against closing closed. I’ve even decided I was wrong and gone back to vote to close something I’d argued in favor of. I’m not perfect. In this case, it hinged on the fact that the answer space was tightly limited—with race, background, a feat, and 19 out of 20 levels predetermined, there was a really limited answer-space here. Also, fundamentally, I think this site is far better at “optimization” requests that are more conceptual—“how do I leverage and build upon what I have?” “how do I emulate this approach or tactic?” and so on, than it is at literal “how do I have the most x at level y?” Those kinds of questions tend to be handled better by discussion forums—which can do a dedicated iterative process, improving upon the best results, to try to really establish the “best” possibility. Stack Exchange is good at reporting and documenting those results, but not so much at determining them independently.

\$\endgroup\$
  • \$\begingroup\$ I should mention here that I haven’t really read the other question, so I’m really only discussing my own. \$\endgroup\$ – KRyan Jan 14 at 3:57
  • 2
    \$\begingroup\$ This matches my experience too; established users tend to get the benefit of the doubt a lot of the time out of faith that they probably know what they're doing. It's a reasonable thing to do. So while it's definitely not the only reason the questions have been treated differently, it's probably not a non-factor either. \$\endgroup\$ – doppelgreener Jan 14 at 12:21

You must log in to answer this question.

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