1

I need some clarification on reviewing First/Low Quality Posts.

I encountered (and failed) this audit which is basically the answer "Use a physical keyboard" to the question "How do I boot a laptop with a non-working keyboard?" The answer is short, but in my opinion otherwise factually correct/helpful.

Is this answer's length the issue? Do we not want one-line answers that would otherwise solve the OP's problem/question? To that point I could see commenting to encourage the poster to add an explanation why this solution would work, but that wouldn't have materially changed the answer's general approach to the problem.

I frequently refer to this Meta answer when reviewing First Posts. I had used its checklist for common reasons to Flag/Delete when considering this audit's answer:

  • Is the post a link only answer? No
  • Check for the instance of code if they provide a link. n/a
  • Is the person asking a new question? No
  • Is the poster answering the question? Yes
  • Not relating to the question It's related
  • Someone sending a "thanks" to another user No
  • The original user posting the answer as the exact copy of someone else's answer (similar to a thanks) No

Meta also has a Low Quality posts guideline answer that doesn't provide clear guidance on dealing with short, non link-only answers. It does however suggest that for

Wrong and unhelpful answers: If you can fix it without making an intrusive edit, do so. Otherwise, leave a comment explaning [sic] what’s wrong and possibly downvote. If there is no other problem, choose Looks OK.

I'm not complaining about failing the audit, but without understanding why I failed it I will likely do so again and fail to support the community's wishes for handling short but otherwise helpful answers.

2 Answers 2

2

The question doesn't sound to me like it's an issue with the user's keyboard.

The answer is just a short suggestion, which, as the second comment suggests, should be a comment on the question.

The answer is not really a solution to the user's problem. It might help him, but it lacks the focus on the issue at hand. It might be a valid suggestion, so it should be a comment. But there is not enough information to properly see how this answer would a solution to the issue.

Beyond the length of the answer, using terms like "pre-boot stuff" isn't helping. The user should give clear instructions on how he thinks an external keyboard will solve the problem here. Why is an external keyboard going to solve what the user describes? It isn't clear to me.

That being said, a failed review audit every once in a while is no big deal. Don't try to over-analyze it. Sometimes you might fail an audit because the the deletion of the post was questionable in the first place. It's not a perfect process, but it's good enough.

8
  • So we are to gauge the helpfulness/accuracy of an answer to the point of flagging/deleting ones that only "go halfway"? I'd edited this out of my question but it now seems more relevant to point out the "First Posts" reviewing guideline I linked states Don't focus on the actual answer itself. Focus on the formatting and the etiquette of the author. I don't know that I agree and would I correctly observe you don't fully concur either? May 27, 2015 at 19:00
  • @Twisty: It is simply not a good answer for many reasons. Guidelines are just that; guidelines. They aren't rules. I honestly don't see how the post attempts to answer the question. Cases where a single sentence is a valid answer are rare. May 27, 2015 at 19:05
  • I understand. For one still learning the process, gaps in the guidelines constitute gaps in a clear understanding of the community's wishes, so I appreciate the clarification. Based on your feedback I'll begin scrutinizing beyond answers that are blatantly wrong or in need of grammar/formatting corrections. Thank you. May 27, 2015 at 19:17
  • Not to be a pest, but should I have recommend deletion for superuser.com/review/low-quality-posts/382809? May 27, 2015 at 19:43
  • @Twisty: It's not a good answer, as it doesn't describe how to do what the user suggests doing. So, yes, I would probably suggest deleting. But, don't take my word for it. You'll probably get a much better sense of what is a good post the longer you take part in this community. Posts from the starting days of Super User are often not up to the higher standards we have today. May 27, 2015 at 19:45
  • Ok, that's very helpful to know. I'm glad I asked as this has (obviously) been a point of confusion for me. May 27, 2015 at 19:50
  • @Twisty: Just always remember, when you're reviewing, you're battling a machine. Failing reviews is no big deal, it doesn't mean you've done something wrong, maybe the machine just made a bad decision. If you get review banned for failing too many, you should definitely bring it up. But don't get discouraged and don't let a failed audit have too much of an impact on your judgement. Review your decision, but it's fine to come to the conclusion that the machine was at fault. Btw, thanks for taking the time to review posts to begin with :) May 27, 2015 at 20:04
  • Thanks for the encouragement! Failing the audit didn't bug me. It did (helpfully) highlight the fact I didn't know how to handle answers of this type. Thanks for gracefully helping me improve in this area. May 27, 2015 at 20:58
1

An answer should do the following in my opinion. If it fails to do these three things, there is likely something wrong with the answer, and much of the time something can be done to improve the answer.

  1. The question should clearly demonstrate an understanding of the author's question. For example a question on how to copy and paste in a Windows 7 command prompt shouldn't talk about that capability being different in Windows 10 and being able to do it with a keyboard shortcut.
  2. This leads to the second point, the answer should in detail answer the question proposed, a statement the author of an answer, does not explain how to SOLVE the problem then it can't possibly be considered an answer. Stackexchange websites are about answers not questions honestly, all it takes to make a good question, is one person to understand what the question actually is and explain how to solve that problem.
  3. The third, and perhaps the most important reason, but only listed last for no particular reason. An answer shouldn't produce additional questions. If an author of an answer is asking questions, like asking for system information, that is more of a comment, and best done outside of an answer.

Since I almost forgot this, I won't even give it a ranking, an answer should be self-contained. It after all is an answer to a question. You don't answer a question, when you are talking, by telling somebody to click on a link to more about a subject when asked "what would you link to order?" You just answer the question, and order, what you would like.

You should also speak from a stance of authority. If you use the words "I believe ..." the you better be able support that stance.

For example: https://superuser.com/review/low-quality-posts/382809

Mount your external drives to a server with NFS and Samba.

This statement is an action statement, but there is no explanation, of how to perform that action. This type of an answer is what Q&A websites were like before Stackexchange, in other words, its complete rubbish. Anyone could have told the user to mount their external drives to a server. It isn't helpful to anyone honestly.

Of course if one takes a literal view at the question. One can see this statement does not answer the question. So you are left with an irrelevant statement about NFS and Samba.

I would like my external drives to be readable and writable from Linux, Mac OS X and Windows. FAT32 works, but ....Are there any alternatives?

So the author is asking for ways to share external drives with OS X, Linux, and Windows. Multiple answers explained what file systems could be used across those platforms.

We can also looked at your failed audit: https://superuser.com/review/first-posts/382819

Borrow/buy a USB keyboard to do the pre-boot stuff.

So the user's question indicates that he has known the keyboard won't work before Windows boots. He also explains he normally just let it crash or automatically pick the option, and he was willing to accept this. He wants to know the following.

Is there a way to change a file with something such as a Linux live CD to force the computer to skip the start-up repair menu?

So the answer to "is there a way to change a file with something" is to burrow a USB keyboard? Lets just put aside the fact, if the built-in keyboard isn't working before Windows boots its very unlikely a USB keyboard is going to work either. Even if this solution did work, it does not answer the author's actual question, how to bypass the start-up repair menu.

So you are left with a below average comment, broken by the technical facts of the problem, which should have been researched before even submitting the answer. If the author of the answer had done this, they wouldn't have bothered submitting the answer.

You must log in to answer this question.

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