What happened in Area 51:
- In Area 51, the example questions have no bodies, so they must be entirely self-contained only in the title.
- Most puzzles and riddles are impossible to be crammed to a single title. At least for me, I discarded some questions simply because I found no way to reduce them to only a title.
- In order to cram the most content in a single title, a lot of coordinate effort was necessary from many people. That was not an easy task afterall.
- The reason that driven the improvement of the titles was that if they are not self-descriptive enough, a diamond moderator might suddenly close them as "too broad" or "not a real question". Again, since they have no bodies, there was nothing to redeem the question otherwise.
Put simply, what driven the improvement in the titles was a combination of a good and a bad reason: The good reason was community coordination and collaboration. The bad reason was fear.
Could we do the same here? Perhaps, but this will be not easy. Being moved by fear is not good, and if you take out the fear, the questions would probably just have the original titles. When people elaborate questions, they are normally lonely, so no coordination and community collaboration would happen. What we could do here is to incentive people in improving the titles. And by improving, it would mean "make then as self-contained as possible and the most descriptive of the question as possible". In fact the title is not a title anymore, it becomes a one-line summary of the question instead. And again, for many questions, this is simply impossible.
So, what is an useful directive to take in writing titles? I propose this:
The title of the question should in fact be a one-line summary of its contents.
Creating a question is easy. Creating a good one is hard. This is somewhat that happens universally in StackExchange and elsewhere. Here on puzzling, anyone can invent a crappy puzzle or riddle in just 5 minutes and post it. But to invent a good one, it is necessary some creativity, talent, writing skills, time, practice and experience, and not everyone has all of these and not everyone knows that he/she is lacking any of these.
Bad questions will inevitable appear. Further, they rarely are bad in the mind of their creators, they are bad only to everyone else. People who create bad questions tends to sincerely believe that their questions are great, or at least good enough, so simply telling people "do not post bad questions" will not change anything.
Bad questions are full of problems as vagueness, broadness, ambiguity, errors and a lot of more problems. See this question and especially its answers for a further discussion about how to evaluate if a question is good or not.
To show what are happening, I will incorporate a bad question writer creating an awful question, but thinking that it is great:
Ok, I want to create a good murder mystery! Lets see... Ah, there would be a millionaire that has died. Who could kill him... Ah, the buttler, of course. Why he would do that? To get his money obviously. Who else is around? Hmmm.... Ah, he has a neighbour who owns a dog and... The dog don't likes the millionaire. And... there could be a lawyer somewhere, this might be cool. What the lawyer was doing... Oh, his testament, of course! And the buttler knew that, so it killed the millionaire before the testament gets done. Hmm... I will need somebody else to this be really cool... Oh, a gardener that is dating a blonde drug-addicted woman would be funny, haha. And the girl likes the dog, how cute! That is it, its perfect! Now, it is time to write the question:
Who killed the millionaire?
You found a millionaire died in his mansion. His lawyer were just preparing his testament when he died. His neighbour has a dog that don't like him, but the girlfriend of the gardener, a beautiful blonde girl, likes the dog. You know that she has some problems with drugs. And there was a buttler in the mansion too. Who killed the millionaire and why?
And then, yet another awful crappy question is posted! The author sincerely believes that this is a good question and it makes perfect sense in his/her own head. But in fact it has no logical solution that leads to the conclusion that the author were pretending that was the correct.
When still unsolved, many people may think that the question is in fact good, and sincerely starts to make his/her brains works to found a solution. When they found something that might be an answer to the question they post it, in hope that it might be the correct answer or at least get some advice of the correct solution. Other people, post ridiculous trollish answers as a way to subtly tell the question author that his question is crappy.
I disagree that lateral-thinking and PPCG's code-trolling are somewhat related. The purpose of code-trolling is/was to ridicule awful gimme-teh-codez SO questions (and I know this because I am the inventor of code-trolling). Lateral-thinking has nothing to do with that, the author has an intended correct solution, and the tag is used to tell that there is an unknown context unclear in the question that needs to be guessed in order to find the correct answer. About the answers, this is very different too. In code-trolling you are/were supposed to post answers trolling the fictional bad question OP. In lateral-thinking, answerers are supposed to try to find the correct and serious answer, and most people do (or at least try to do) that. What happens is that some people just do wild/strange guesses or post trolling answers ridiculing the actual question itself or the process needed to find the answer, and not because they are supposed to do that.
The problem with lateral-thinking is that in order to correctly solve the question, you need to correctly guess what is the context of the given question. If the question requires unreasonably guessing, it is probably a bad one, because there is probably an infinitude of radically different possible but equally valid ways to solve it, and only one of them is considered by the author as the correct. And here is the root of the problem.
So we may get back to this question. Although it is negatively scored at the moment that I am writing this and is indeed broad, it goes directly to where the problem lies and should be better evaluated IMO. I really recommend people to read the answers and to try to post additional answers to that question, regardless the fact that it is broad or not (this could and IMO should be fixed by editing and/or making the question community-wiki and/or migrating it to meta).
So, to put simply, the problem is not the lateral-thinking tag. The problem is the infinitely equally valid and radically different ways to solve the question, regardless of the way it is tagged. In fact, I think that it is indeed possible to write a valid good lateral-thinking question that do not suffers from this problem, although at this moment I don't know how.
Back to Area 51
The proposal at the Area 51 still stands. Why? Because we still don't know what would or not be on-topic here! I still did not found an unambiguous, clear and direct official statement that puzzle challenges are welcome and that riddles challenges are equally welcome. When we get a firm and definitive position in favor of that, I am sure that everybody would be pleased to abandon the proposal. If we get it to the opposite, everybody would be sure that the way is to work to make it reach the commitment phase and then beta.
How to avoid writing bad questions and bad answers?
The majority of people here have no clear directive about how to write a good question, it is entirely based on intuition and/or imitating. I didn't saw anything clear about that in meta and this question (again) that is asking about this is just a few days old and negatively voted at the time that I am writing this.
As I mentioned before, it is necessary to have some creativity, talent, writing skills, time, practice and experience, and not everyone has all of these and not everyone knows that he/she is lacking any of these. So what to do?
Discuss more about question-writing here in meta (or even in the main site). We currently are lacking a good debate of how a question writer could improve his/her skills in question writing. I will not discuss it in detail here because this is not the purpose of this question, but we should really create some (meta-)questions dedicated for this. Without these, we will not have any improvement in questions quality. And the debate need to be much more focused in writing questions than in disputes about if spoilers are good or not or if deciding that X or Y are or not on-topic. We need to train the community about how to write good questions and not about how to downvote and close crappy questions.
Provide feedback to question writers. When you write a question that you think that is good (even if it is not), it is really frustrating to get downvotes popping out from nowhere without a single word telling you why. It is still worse if your question is suddenly closed without anyone telling that before its closure (and remember, people without enough rep can't see the close-votes). If you think that the question has some problem, even if it is glaring obvious for you, post a comment. This way the author will be able to learn from it instead of only getting frustration that could even result in he/she going away from the site forever.
Good questions attract good answers. People naturally are more zealous on posting answers than questions. We have more directives on meta about posting answers than posting questions. Bad answers are quickly defeat by better answers due to the voting system and this happens because the SE QA format was designed to naturally put a selection pressure for better answers. If the question is great, an eventual answerer rarely will post an stupid answer because he/she knows that this would only attract downvotes or make his/her answer be ignored. OTOH, no such pressure exists for questions, and if a particular question is bad, people are unwilling to put work in trying to answer, or may end posting answers that are as bad as the question, or may post trollish answers, overly creative answers, or just strange/wild guesses in hope to find the answer by luck. So I conclude that the key to improve the answers is to improve the questions.
Be nice, especially to newcomers. No one born knowing how to write great questions and answers. It is almost impossible to write a good question or answer without some mistake that will needed to be fixed by editing later. Most people don't know how to write great questions or answers, but they are invited to participate and try anyway, otherwise we would create an elitist and doomed community. Everyone starts by writing crappy or mediocre questions and improve with time, so we should not discourage them.
We could measure quality in an infinitude of ways, but lets consider two (generic and vague) particular ways: The real content quality and the perceived content quality. Although I will not discuss for now what the vague and generic "content quality" is, lets see the difference between the "real" and the "perceived":
- The perceived content quality do not accounts for closed, deleted or downvoted content.
- The real content quality accounts for closed, deleted and downvoted content.
Aggressive downvoting, close-voting and deleting is not the way to improve the real content quality. This might perhaps only improve the perceived content quality.
To improve the real content quality, what we need is feedback, incentive, training and clear directives and instructions in meta and elsewhere. With that, people will tend to learn faster on how to post content of better quality, the overall quality itself improves, and this results in less need of editing, downvoting, close-voting, deleting, complaining, flamewars, rants and frustration.
Riddles are a special case. They have obscure and unclear meaning on purpose by design, and thus are subject to wild/strange guesses in most of the cases. People might argue that "oh, then riddles are too broad and/or unclear what you are asking. Close them. End of story.", but this is not really the case.
Writing good riddles is harder than writing good puzzles (or even non-puzzling questions) on any other category, but it is indeed possible and have been done here before anyway. But again, we are severely lacking meta-questions and directives about how to write good questions (and especially riddles), and without these, the quality will never improve.
In fact, there is still some debate about if they are or not on-topic, so we are still way behind the point of defining how to create good or bad riddles, but I think that we should run for it anyway and declare them "clearly and undoubtly on-topic". Otherwise, we will need to go to Area 51 or sites outside SE and we really do not want this.
And, as suggested by xnor in a comment, the fact that riddles are on-topic does not implies that every riddle is on-topic. Further, bad riddles are bad questions, and this do not means that all the riddles are bad nor that bad riddles are acceptable. In fact this could be generalized to every category of question in this site: if foo is on-topic, this do not implies that every foo is on-topic; Bad foo questions are bad questions, and this do not implies that every foo question is bad nor that bad foo questions are acceptable.
How to evaluate a question?
That is a one-million dollar question! It is very hard to tell what is or not a good puzzle or riddle if you don't know the solution. This way, people don't know how to vote on questions, only on answers. We have a question about that here. And again, this question is very recent, so we are just starting to debate the issue, a very large and important issue. Without knowing how to evaluate a puzzle or riddle without knowing the answer, it is very hard to improve the quality and we are just starting to debate this.
A little note (just to be pedantic)
Doorknob, you said that "... even though it's actually one of the better quality riddles we've had ...". And I hope that you just had forget to write the "not" in that statement, otherwise I vehemently disagree, IMO it is either a crappy or no more than a very mediocre riddle and we have plenty of better ones. But it is just a single question between many many others, so I will not debate much about it.