# LQP review & voting

I often come across answers flagged for LQ in the corresponding review queue which don't really fit into the categories of a LQ post - not spam, not better as a comment, not better as a question, etc. Using a broader definition, the post is of low quality; perhaps it's not very clear, or the physics is incorrect, but it remains an apparently honest attempt to answer the question. The correct course of action would seem to be to downvote rather than to vote to delete, but voting isn't offered via the LQP review interface. I often just open the question in a new tab via the title-link and vote there, while completing the review as 'Looks OK' in the other tab. This seems annoying; why shouldn't I be able to vote on content I happened to come across in this queue? Or to take it further, why shouldn't a downvote be an acceptable resolution to a LQP review? Is there a stated rationale for this, or should this become a ?

• This might be more of a question for the mother meta; but I agree with the frustration. Somewhat relevant mother meta question. – JMac Jan 25 '18 at 14:13
• Is e.g. this answer (screenshot) an example of what you have in mind? – Emilio Pisanty Jan 25 '18 at 14:48
• @EmilioPisanty that is exactly how I treated that answer. – Kyle Oman Jan 25 '18 at 14:50
• @KyleOman Indeed, that's what it says on the review ticket ;-) - and it also shows that there is a nontrivial support from the 2k<rep<20k population for removing that answer. – Emilio Pisanty Jan 25 '18 at 14:54
• @EmilioPisanty Well, you can see the review, but only speculate about the downvote, I suppose. The circumstantial evidence is strong, though ;) – Kyle Oman Jan 25 '18 at 14:59
• @Kyle No, my real question is whether this is a good example of the type of answer you were talking about, or whether you were trying to explore our reaction to some other bit of questionspace. – Emilio Pisanty Jan 25 '18 at 15:07
• @EmilioPisanty that's an example of the general idea. There are many flavours of such answers, but the point is that they don't fit into the canned LQP categories. – Kyle Oman Jan 25 '18 at 15:31
• As one of the more prominent LQ reviewers, my opinion is that answers such as the one Emilio gave ought to be deleted (in addition to downvoted). Basically, "not even wrong" answers should be deleted without hesitation & I'm not exactly sure why that would be a question on what to do. – Kyle Kanos Jan 26 '18 at 1:01

This issue has been discussed repeatedly on mother meta. Some examples:

There is a (debatable) rationale behind the decision of removing the voting option from the LQP queue. I'll copy some relevant paragraphs here, mainly for completeness, but I really encourage you to go there and read the posts for yourself.

It's always seemed to me that the purpose of the low-quality-posts queue was to automatically detect and remove low quality posts. [...] Voting on the post is unnecessary in order to accomplish this.

${}$

The core philosophy here is that each queue focuses on a specific task or question, and provides the tools most important for resolving it. If you want to do something else, there will always be a prominent link to the full question page. A concern with allowing voting in the Low Quality queue specifically is that you're not viewing answers in the context of other answers, or a question in the context of its answers. You're never able to vote in these circumstances normally - it's worth remembering that voting directly from the list of questions was considered early on, and discarded as too likely to produce bad results.

You may or may not agree with these statements. If you don't, the best you can do is to go there and post an answer (or create a new post, if necessary). Unfortunately, I don't think you will be able to change people's mind, as this has already been thought through and a sort of consensus has been reached. The LQP queue is not, in general, for voting. If you do want to vote, you'll have to open the post in a separate tab.

There's a few things going on here that I can see, and I previously ranted about that subject here, about a year and a half ago. At the time I proposed trying out a change in the mods' response to VLQ flags, which I think is still in place, but we didn't really have the time or energy to analyze any changes that happened at the time.

The first thing to keep in mind is that much of the existing guidance, both on this meta and more generally in SE, was written with the expectation that VLQ flags would be handled by moderators. Within that paradigm, deletion is not the right response: moderators are not and should not be arbiters of the technical correctness of posts.

However, just because moderators shouldn't be doing that job, that doesn't mean that the job shouldn't be done at all: the whole point of this site is that we have mechanisms (however imperfect) for recognising people with expertise in the field and knowledge of this site, and which give those people moderation abilities that can affect other users' content, from how prominently displayed they are (via the voting privilege at 10 and 125 rep) through editing (at 2k) all the way up to voting to delete answers (at 20k). The LQP queue is the point of contact that puts experienced users in front of content that might need to be deleted.

I think the answer that probably inspired you to write this (screenshot for stability) is an excellent example of not-even-wrong post that, while it is ostensibly an honest attempt to answer the question (so Not An Answer doesn't apply), is so deeply and completely wrong that this site is not obliged to give that type of content a platform. Those types of answers harm the site: the signal-to-noise ratio matters, very much, in attracting the types of expert answerers that make this site great.

So: independently of whether the system allows you to downvote, I would encourage you to vote to delete (or Recommend Deletion for <20k users) those kinds of posts. We have some ~30 users with the 20k rep needed to delete answers, and over the past year or so we have started to get more and more community-driven deletions of posts that really shouldn't be given a platform here. As a particularly positive development, a nontrivial fraction of them include Recommend Deletion votes from the much wider and active >2k rep section of our userbase (example, screenshot).

However, while I've yet to come across a deletion that I've disagreed with (or that anyone with this site's best interests in mind could reasonably disagree with), the three-20k deletions that currently occur tend to feature, much more often than I'd like, a set of three or four regular faces instead of a wider set of characters, which would be a much healthier situation. So: please review the LQP queue, and please don't hesitate to hit the Delete / Recommend Deletion button, particularly if you're 20k+, if you think the answer really doesn't add anything at all to the site.

Now, having said all of that, I should add a huge caveat, which is the reason I haven't pushed further in this direction since my previous rant. The wisdom-of-the-crowd method, as implemented broadly by SE, does have its pitfalls, and the claims that it can unduly lead to censorship are real and cannot be ignored. In order for this moderation mechanism to work correctly, that danger needs to be offset by a suitable set of checks and balances, which can detect and correct deletions where the crowd gets carried away.

To my mind, what we really need here is (i) a list, accessible to 10k+ users, of all the posts that have been deleted (maybe only including 60 or 90 days' worth of data), with the ability to filter according to whether they were deleted by moderators / the roomba / the community, and (ii) a healthy population of users which regularly checks that list and then corrects any undue deletions by flagging or taking them to chat or meta as required.

However, the software tools aren't really up to scratch yet (the closest we have are the 10k review tools, which haven't gotten any love from the dev team in a long time), and I'm not really sure that we're quite there yet in terms of a critical mass of actively-reviewing 10k+ users. It's a problem that can be solved with some additional software features (← nudge nudge vote vote) and some more time, so I think the best thing to do now is wait some more until that critical mass is there before doing a solid push for the software.