How nuclear do we want to go on low-quality posts?

Question

It is often the case that posts which don't directly answer the question, are weirdly rambling and indirect, or otherwise don't really fit the high standards we are looking for here are flagged and end up in the Low Quality Posts Review Queue.

When reviewing, the options are "Delete" or "Looks OK". (Or, not to be forgotten, "Skip".) If the post is gibberish, spam, or really, really, really incorrect, I have no qualms about voting to delete. But what about otherwise? How trigger-happy should we be?

I often leave a comment, occasionally downvote (although these posts usually come in with a negative or at least 0 score already), or upvote existing comments noting problems, and then press "Looks OK", even though the post doesn't really look okay to me. The UI seems strongly geared towards encouraging reviewers to vote to delete these borderline posts. Is that what we want?

Answer 1

Good question. I think the Delete option should only be used if the answer is unsalvageable, which is rarely the case if it's an attempt at an answer and there are comments to help them make the answer better.

Delete if it has almost zero chance of remaining as an answer, otherwise Looks Ok (and comment, downvote, edit, upvote a helpful comment etc)

I leave the ones you describe - long, rambling, not fully addressing the question, or just plain wrong - to be downvoted.

There's some discussion of it here:

https://meta.stackoverflow.com/questions/266428/low-quality-review-queue-meaning-of-looks-ok

Answer 2

I think:

• Does the answer kinds of look good? Is it grammatically kind of correct? Is it relatively sanely formatted? If not, it is not a well-formatted answer, so it is not an answer, so Delete it. (People will see this visual noise and not dig through it). If you allow for poor grammar and formatting others will be encouraged to follow that practice and we will have to keep cleaning those up.
• Is it relatively trying to answer the question? (Does at least half of it answer the question?) If not, Delete it.
• Otherwise Looks OK (which means: I can read it and it kind of looks an answer.)

The quality of the answer will be decided by up- and downvotes.

Answer 3

I think if there is potential for an answer, we should try to encourage that the answer be edited and improved. There are certainly clear and obvious "deletes", but otherwise, I think chances should be given.

There should probably be a time limit to how long we wait for improvement. Some people drop by, answer, then leave and never return, and if an answer cannot be improved, and the community cannot improve it, the answer should probably be deleted.

Answer 4

The only guidance that I've seen about how to review in the Low Quality Posts queue is in How aggressively should we maintain and improve very popular questions? which deals with the quality of answers in a post.

In particular, the recommendation is:

• Extend the benefit of doubt to a correct though weak answer from a user with a significant amount of reputation and is still participating on the site.

  Whereas if it is a weak answer by a user with 1 rep, or a user with 50 rep who hasn't been seen in a year, I'd be much more inclined to delete it outright.

• Combine duplicate answers (this requires going outside of the review queue). If one answer brings up a, b, c, and d, and another one brings up a, and c - look how to combine the second post into the first, and then delete the second post.

• If they cannot be combined and are bad by concrete metrics, delete them.

  We know that quality equates strongly with length (TWSS). Therefore, answers that are strong candidates for deletion:>

  • belong to low rep or anon users with no real commitment to the community

  • are provably duplicate, that is, were added well after (30+ mins later) other answers that contained the same exact information

  • are short in length

  • do not explain much of anything

And there you have it.

The key value of a Stack Exchange site over a forum is that the information is curated and high in the signal to noise ratio. If it becomes too noisy with non answers, duplicate answers, low quality answers this value proposition is lost and when someone does a search on google they may find that the less noisy answer is on photo.net forums or preview. To encourage people to come here we need to have better and more accessible answers - that fourth equipment recommendation for how to use an F mount on some arbitrary body from an unregistered user doesn't contribute to the signal in the site.

Do not worry about hurting the feelings of an unregistered user who hasn't been active for a month. Instead worry about what a new user coming to the page from a google search would see and what types of behavior that question and its answer encourages. If you see half a dozen non-answers, whats the harm in adding one more (and the answer to that is that it makes it harder to find the actual answer in the chaff).

Answer 5

The downvote test of "Is it useful?" seems useful.
I'm ongoingly amazed but not surprised at the posts that get downvoted, despite their adding usefully to the sum total of knowledge on the subject. Often enough in life I find gems lying in unexpected places - maybe one sentence comments that add substantially to ones understandings. If it's on topic and 1/2 or better understandable AND useful I'd be happy to see it stay.

Answer 6

Crap is crap. Nuke it and walk away. Then you can spend your time writing a good answer somewhere else, which will leave the site much better off than if you spent the time trying to fix a bad answer.

And no, don't worry about the feelings of the poor little dweeb that wrote the unfortunate crappy answer. It should feel unpleasant to them. We want them to go away. If not, they'll just do more of the same.

Worse yet, more drivel, and then that drivel being popped to the top by edits, will chase away the experts and those that write good answers and care about quality. You can't have it both ways. By making it more inviting for the dweebs you do the opposite for the experts, even if you're not doing so conciously and intentionally.

Good quality begets more good quality. It works the same for poor quality too.