3
\$\begingroup\$

When I review posts in the low quality posts queue, sometimes I stumble upon answers which provide alternative solutions instead of reviewing the OP's code.

It would be nice to add this entry (alternative solution) to this list:

enter image description here

\$\endgroup\$
1
  • 3
    \$\begingroup\$ I've checked with the devs and this is (as Toby Speight has already mentioned) a network-wide list with no way to customize it per site. Apologies. \$\endgroup\$
    – Catija StaffMod
    Apr 13 at 19:45
4
\$\begingroup\$

I'm reasonably sure that the set of review comments is common to the whole Stack Exchange network, so we can't adapt these to Code Review.

My usual approach is just to add a canned comment (using the AutoReviewComments userscript), then select Delete with "no comment needed".

\$\endgroup\$
4
  • \$\begingroup\$ I am not sure that No Comment Needed is a good option for this type of answer. The reason that I say that is because it is a low quality answer. The way that I see these types of answers is that they are answers, just not very good ones, and I often times down vote them when they already have comments on them like the ones that Rolf provides in this case. they could be very good reviews or new questions all on their own. If we use the No Comment Needed selection it could give the wrong impression in the metrics and then we would stop seeing these in the queues. \$\endgroup\$
    – Malachi Mod
    Mar 26 at 14:52
  • \$\begingroup\$ I don't think we can change these options, I haven't been able to find a place where we can edit them. --Edit I added the [Status-Review] but this is likely to be declined... \$\endgroup\$
    – Malachi Mod
    Mar 26 at 14:54
  • 1
    \$\begingroup\$ Ah, the point being that having added a comment manually, no further comment should be added from the review. \$\endgroup\$ Mar 26 at 16:13
  • \$\begingroup\$ I think I disagree on your comment that code-dumps are (low-quality) answers. The code could become part of an answer (specifically, an illustration of the suggested improvements), but where I come from (in my day job, for example), replacement code with no other comment doesn't constitute a review. \$\endgroup\$ Mar 26 at 16:18
6
\$\begingroup\$

When I come across these answers I mark them as This is a different question posted as an answer.

To me that's the best option of the current set - it is not a review - and would make a decent question if introduced right.

The real question here is whether there should be a dedicated alternative for this scenario. I agree that it is relatively common (I'd guess 1-in-5 reviews I do in the queue are like this).

For me I am content with the this is a different question posted as an answer option - it encourages the person to think about reformulating it, and adding more questions to the site, inviting additional answers too. What's missing is the wording to encourage the user to alternatively reword the post to make it a review rather than an alternate solution.

So, my suggestion is to add wording to the this is a different question posted as an answer option to make it clear that editing it to be a review instead of a code-dump is an alternative to posting it as a new question. Something like:

Answers on Code Review should provide insights on how the question's code could be improved. Alternative solutions should include descriptions of why the alternative is an improvement. Alternatively, if you have a new question, please ask it by clicking the "Ask Question" button. Include a link to this question if it helps provide context.

For what it's worth, the other somewhat related option 'This is an "I'm having this problem too" comment' is something that should be removed ;-(

\$\endgroup\$
1
  • \$\begingroup\$ That would be ideal, but I think it would need some Stack Exchange support to be able to customise that list on a per-site basis. \$\endgroup\$ Mar 22 at 16:56
-3
\$\begingroup\$

When I see an answer that is code only, I believe that the answerer created the code because they felt it was better in some way. If the answer is better in some way, but the Answerer hasn't explained why, I feel that it is still a review, but one that I would downvote until the answerer explains why/how their code is better than the OP's.

In this case, I look at the How do I use the Low-quality posts queue? and in the section marked Basic Workflow, it says:

Choose Looks OK for questions that are on-topic, sufficiently detailed and not too broad. Use this option for answers that answer the question as asked, even if they may be incorrect.

So, this is the option that I would choose. I would also comment on the answer, asking the answerer to explain the code, and downvote the answer. When the answerer edits in the explanation, I would then change my vote.

\$\endgroup\$
4
  • 4
    \$\begingroup\$ In the past the recommendation has been to delete such posts 1 2. Additionally the common solution right now is to comment, downvote and delete such answers. Sometimes with the help of a moderator; as getting enough 20k+s to delete such posts can be challenging at times. I don't think the help center's guidance is accurate here. \$\endgroup\$
    – Peilonrayz Mod
    Mar 29 at 17:30
  • \$\begingroup\$ Downvoting ultimately leads to deletion if enough people downvote, it just takes longer, giving the answerer/reviewer a chance to make their answer/review better. \$\endgroup\$
    – Malachi Mod
    Mar 29 at 20:24
  • 2
    \$\begingroup\$ Apart from downvoting the post to have a negative score; downvotes play no role in deleting answers. We 20k+ users can only delete answers with a negative score. As such I downvote unless the requirement is met and then vote to delete. Yes, not voting to delete / recommending deletion does give the user more time to improve the answer, because, well, then the answer won't ever be deleted. No roomba exists for deleting negatively scored answers. \$\endgroup\$
    – Peilonrayz Mod
    Mar 29 at 20:54
  • \$\begingroup\$ I was thinking of Questions, they are auto deleted in some cases once they have enough negativity (-3 IIRC and some other criteria also). you are correct. I don't VTD answers anymore for various reasons. \$\endgroup\$
    – Malachi Mod
    Mar 29 at 21:26

You must log in to answer this question.

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