How to Improve Quality

Question

In recent discussions, we've seen there are two major suggestions for improving our site, Quality (better questions) and Quantity (more questions). Obviously, we all want both, but some of us (full disclosure, me) think we need to bring in more questions first, and then improve quality, others think that if we improve quality first, that will drive more questions.

THIS QUESTION IS NOT ABOUT THAT DEBATE!

Instead I want to ask, what would it take to improve the quality of this site, and how should we go about doing it (independent of any other strategy to directly increase the quantity of questions)?

Answer 1

Close questions that do not have enough information to be answerable

This is a very big problem currently that questions that cannot be answered are open and people just guess on them. Instead questions should be closed until the have enough detail to be answerable specifically for ID question, or for other questions that aren't clear close as unclear. If the question is fixed, it can be reopened.

Answer 2

Edit noise out of questions and make them findable

I do think we are getting better at this one, but we still could use a lot of work on this. We should make sure questions asked have a good title, correct tags, the body is readable with correct spelling and grammar, and any excess noise (like signatures) are removed from the question.

Right now if you do a search for What song is this, we have several questions where that is pretty much just the title.

Answer 3

Have a format and requirements for questions and answers that typically generate low quality responses.

One of the bigger problem with answers to ID questions was we were sometimes just getting a song name and a band as the entire post. We would also get questions that were just links to a song and no more information. Both are not high quality posts.

Having some kind of minimum requirements for these and enforcing them will show these questions and answers need structure to be asked here and will help improve quality.

Answer 4

If we can't care for something, let it go

I'm not going to explicitly talk about current scope here, but if we can't handle answering or caring for questions of a certain type we should reexamine how to handle them. Specifically if the community as a whole tunes out the questions, or does not want to deal with the questions then it's best to just disallow them. There are many in-between steps and standers we can impose, but we can't just have something because people ask and not care for it.

This is where having a scope shaped up helps and we can shape it as the community.

Answer 5

Vote with purpose for what it means

I'm not saying we currently don't, but this goes back to the guessing game post. Misplaced votes (both up and down) have consequences. I'm not going to say exactly how we should vote, but please keep this section in mind from the reputation & moderation help page:

Voting up a question or answer signals to the rest of the community that a post is interesting, well-researched, and useful, while voting down a post signals the opposite: that the post contains wrong information, is poorly researched, or fails to communicate information. The more that people vote on a post, the more certain future visitors can be of the quality of information contained within that post – not to mention that upvotes are a great way to thank the author of a good post for the time and effort put into writing it!

Answer 6

Locate pre-existing, high quality questions, and work to give them good answers

After several years, we have a large backlog of questions, and some of the best ones don't have answers, or have insufficient answers. Even without a strong supply of fresh questions, we can improve the effective quality of our site by working better with the best of the questions we already have.

Answer 7

Stop catering to guessing games

Specifically ID questions, but other questions also have a problem where they turn into guessing games. One thing may fit their description, but isn't what the OP wants. Leaving the answer solely in the OP's mind, but the voting doesn't always go that way.

Leaving a guess there that is incorrect that gets an upvote (which happens all the time) sends a bad message that any guess is tolerable and we don't actually care about answering the question and trickles in incomplete answers that are total shots in the dark.

As a site, we need to figure out how to handle this. Leaving wrong answers would be fine if it wasn't for the upvote factor. We could have a delete policy for this, but that's another question for another day. Either way we can't leave it the way it is as it signals we don't care about the questions asked.

Answer 8

Create areas of on-topic scope that don't require real 'expertise'

(FWIW I'm not suggesting that SE sites have to be 'non-expert'; clearly SE does have some good expert sites. I'm trying to work out how we get to a functional scope from the point we're at. I have also wondered elsewhere whether we actually can do that.)

For any SE site to work well, there needs to be one or more clear areas of on-topic scope within which people want to come here and ask questions, and for which other people want to come here and answer them (and are capable of doing so). My opinion is that this site didn't really achieve that. Let's look at what's on-topic:

• The history of songs and musicians/bands/artists/composers/songwriters/etc.
• Song meanings and the origins of lyrics
• Terminology used in the music industry
• Use, explanation, and history of music technology, devices, and media
• Information about music memorabilia and the collection of it (e.g. “How rare is X?”)

All these seem fair enough - we had lots of good questions along these lines, and there's undoubtedly potential for many more. The problem here though is that for it to be worth coming to this site after googling and failing, there needs to be a good, high-participation community of experts here who know stuff that google doesn't. The only time we had anything approaching that was at the start, driven by the initial excitement of site opening; things quickly faded, and I don't see how it's going to build up again the current state of things, based on the above on-topic scope.

The site has tried identification questions with mixed success, and I can see the logic in throttling those back. But the question then still remains - what kinds of question do we want that will attract and maintain a user base of people who can capably answer those questions?

One possible area of investigation would be opening up to more questions that aren't really 'expert' questions. I'm thinking of the success of sites like parenting, the workplace, and interpersonal skills; most of the answers there aren't 'expert' in nature at all, just someone's experience. So perhaps we could brainstorm and see if we could find some music-related types of question that would be similarly open to 'non-experts'.