Could we be a little less eager with the duplicate votes to close?

Question

_{It was pointed out to me it was unfair to single out folks with dupe-hammers for this. It was a poor choice of words, on my part. The core of the post is still something I feel needs addressing, so I have edited appropriately}

I've noticed that quite a few folks tend to be really enthusiastic with their dupe-votes. I'm not the only one, and it's come up in the comments of a few posts. A few folks put a lot of effort into updating the community FAQs (which is excellent), and trying to get stuff closed against those (which is a little less so).

While both voting to close as a duplicate and community FAQs are super useful, as a space acting as a community helpdesk of sorts, It seems useful to have answers specific to a problem,even if it references a community FAQ entirely (that's to say "Oh, it's in Section 3C which says...") as opposed to closing as a duplicate directly.

Practically, duplicate closures ought to be for exact duplicates, and in some cases, it feels like getting an answer in that helps the user might be preferable to the quick closure.

So, should we be closing stuff less quickly as dupes? And is there a workflow that might be a reasonable compromise between handling stuff quickly and being helpful to folks who don't know MSE as well as others?

_{Yes, it's a well-known pet peeve of mine. Also, not something I really noticed before being a mod.}

Answer 1

One of the problems here is that some of the FAQs have gotten... Really long. Like this one: The Complete Rate-Limiting Guide

...or, hell, this monstrosity: What are the badges I can earn on each site, and what are the exact criteria for earning each badge?

When you need nine answers to cover a topic, it's not really an FAQ anymore, eh? The original Stack Overflow FAQ started out as a single question with loads of answers, and got split up before the site even made it out of beta because it got too unwieldy. I sympathize with the desire to be thorough, but both of those questions are now terrible duplicate targets because you practically have to use in-page search to find what you're looking for once you get there!

Anyway... Here's my rule of thumb: if the question you're closing isn't represented in the question you're picking as a dup-target - yes, the question itself not a subsection of an answer - then it's probably a bad dup-target. Pick a different one, or write an answer & use your massive reference book of a FAQ as a reference.

Answer 2

Since I'm using my dupehammers quite often (and TL;DR, I'm open to using them less often, if the community desires so), I feel compelled to add my 2 cents. First of all, I like to have some data available, so I spun up a monstrous SEDE query to set the record for most often applying CROSS APPLY and OPENJSON to see in which cases dupehammers have been used this year. Feel free to use to analyze specific cases. There have been 259 closures so far, and only 63 (a quarter of them) towards faq questions. (I expected a higher proportion, what about you?)

Whenever I close something as a duplicate of a long FAQ, I tend (or at least plan to do this more often) leave a comment quoting the relevant part of the FAQ (often just a single sentence). I know that comments are not meant for giving answers and that I'm still duplicating content (the main problem of answering duplicate questions), but at least it helps the author of the question. The disadvantage of posting an answer (and not using the dupehammer) is that I know that other users will complain "hey, this question has already been asked before, you're only posting an answer because you want to earn more reputation". <-- that's the gist of such comments, it's often worded in a way that seems more neutral but is still open to interpretation. It's not an MSE specific problem, it happens on Stack Overflow as well. Is it OK if I'm going to flag those as unfriendly (even if they're directed at another answerer)?

Also, I'm not responsible if in a few years we discover that we have a massive chunk of outdated content on the site. I'm one of the senior developers in our company and I'm often amazed of how much duplicate or near-duplicate code we have in our systems, or to put it differently: how little people can achieve with so much code. Decisions like this (allowing everybody to write their own solutions (read: answers) to common problems (read: questions)) might seem doable at first, but in the long run they only hurt the product (read: Meta Stack Exchange).

Final thought: it seems that some people feel offended if their own question is closed as a (near-)duplicate (or even think it is 'unwelcoming' when another question, not their own, is dupehammered). I ... just don't get that. When one of my first question was closed as duplicates, I felt more like "great, I'm not the only one with this problem". I think the unwelcoming part is more about the downvotes, which may be warranted if the question is poorly researched, but there's no need to go all -7 on it (in my opinion, YMMV).

Answer 3

So, should we be closing stuff less quickly as dupes?

No, I'm trusted with that dupe hammer privilege (and regular close voters with their non-binding vote) and I still go by the vote early and often guidance. It is also unclear what less quickly means: do we wait for 6 to 8 hours?

I'm still having to dedicate my time to find the frigging dupe first and then judge if that question and answer fit my own quality rules. More often then not the dupe I find is closed against an FAQ post. Based on my judgement I pick the best target. If I find that subpar I try to find extra dupes and edit those into the duplicate list.

And it is easy to keep an eye on and/or revisit a couple of times to fix any mishaps, if comments or edits give reason to do so. I also feel that the dupehammers on MSE are receptive to feedback and are happy to respond if they got anything wrong. How that can be perceived as not helpful is hard to grasp for me.

And is there a workflow that might be a reasonable compromise between handling stuff quickly and being helpful to folks who don't know MSE as well as others?

I think the current workflow works for me and without any more concrete issues or examples to work from I see no reason to change my workflow. I've already changed my down voting workflow to help in keeping a post longer on the front-page so other users can get their chance to see a post instead of the usual bunch.

I'm also not sure what other options we have, instead of repeating an answer from a dupe that is, to be more helpful to a specific OP or future visitors that have the same question and came looking here for a canonical answer.

tl;dr
On all SE sites closing a question as a duplicate is a core feature and in my opinion is a good way to bundle knowledge. If anything, new users should be informed upfront that when a duplicate is offered they are probably helped much better then if they waited for someone to write a custom answer to their common issue.