19

A few recent questions seem to exhibit issues explained in The Trouble With Popularity article:

we discovered that these posts... truly start to drown out everything else on the site... it's too addictive and too easy, and in the absence of any moderation, the community would do nothing but add and upvote the easy, fun stuff. This is why community moderators have real power; they need that power to intervene, educate, and refocus the community's exuberance on more substantive content...

While these questions were open, they were widely advertised on the Hot Network List and gained relatively high scores despite piling on of close votes. All of them are now closed, but due to their high scores they present appealing examples for newer site visitors to try their luck asking similar questions.

High scores send a fairly strong signal that those sorts of questions are welcome here. I think this signal is much stronger than the subtle hint of a close banner, especially for inexperienced users ("it's okay to ask about roses here...").

The crux of my question is: what can we do to make these closed questions less like broken windows?


Recent examples for review:

I think the issues with this question are best covered in a custom close reason borrowed from the Workplace:

Real questions have answers. Rather than explaining why your situation is terrible, or why your boss/coworker makes you unhappy, explain what you want to do to make it better. For more information, click here...

Topicality wise, it appears to be in overlap between our site and Workplace, probably closer to Programmers. Our close reason (opinion based) looks like a fit. As for better matching Workplace close reason, it is likely because they (Workplace) are getting more troublesome questions of that kind and have acquired more experience at efficiently handling these.


"...How would you explain a statement... to a beginner?" - We have a dedicated meta post explaining exactly what the issues are with these kinds of questions: How do I explain ${something} to ${someone}?

Of course, hot list visitors who voted it up likely have no idea about our meta guidance. And even if they knew, they likely don't care. (And even if they would care, system simply wouldn't let them tell.)


"I personally prefer 4 spaces for each indentation level, but there are colleagues that prefer 2 and even 5 spaces... I decided to post a question see if there are tools/method to settle the problem without calling a meeting." - If that doesn't make it a shopping list question then I don't know what it makes.

The revision history suggests that even some moderators have tried to figure out a way to protect and salvage the question. But, honestly, I am not surprised to see that this failed. Asker clearly indicated that they were just shopping and there's nothing else on their mind.

If someone tried to "derive" a better, more substantial question from this one, they would end up with their own, and different question.


Yet another case when Workplace folks seem to be better prepared to recognize and handle troublesome questions of that kind with their custom close reason:

Questions asking for advice on what to do are not practical answerable questions (e.g. "what job should I take?", or "what skills should I learn?"). Questions should get answers explaining why and how to make a decision, not advice on what to do. For more information, click here...

Granted, I think that our own education advice close reason also makes a good fit, but the majority of close voters picked another reason (too broad) which suggests that the match with the education close reason is not as clear.


And yet again, issues with this question seem to be best covered in already quoted custom close reason from the Workplace:

Real questions have answers. Rather than explaining why your situation is terrible, or why your boss/coworker makes you unhappy, explain what you want to do to make it better. For more information, click here...

Although still somewhat workplace-y, this question looks much closer to Programmers than previous example covered by the this close reason ("mentor's concern"). Still, their close reason looks like a really good fit.


I'm often conflicted when creating a project that requires manipulation of data in files... <example how JSON looks like> <example how CSV looks like> ...is it better to use larger, more readable formats, like JSON, or smaller, harder to read formats, like CSV, for data storage?

If memory serves, the close reason I picked was too broad. The majority of voters have chosen opinion based which also looks like a reasonable match (Gorilla vs. Shark).


If these questions were old, or if their score and views were acquired organically due to great content, I'd probably ask for a historical lock. But that doesn't seem applicable due to the recency of the questions.

For what it's worth, I considered an option to edit them into better shape. Unfortunately, it looks like the answers that piled in prior to closing "lock the question" from any substantial changes I could think of.

One possible exception (which would involve removing most of the answers though) is the question about indentation width. It looks possible to convert it to a collaborative effort question. If we pick this option though, we had better do some substantial edits to the question text, to make it look like less of a wide open broken window for inexperienced site visitors. "Hey, folks do shopping over here, lemme try my own."

6

It was quite interesting to observe impact of this meta post on the listed questions ("meta effect"). Or, more precisely, lack thereof.

First thing I noticed is there were no attempts to improve / salvage these questions (except for an old one about indentation width, there was only one, more on that below). Attention brought to these by this meta post did not really help: "active" notices on listed questions (except for last one) say 2 months ago, 21, 17, 14, 16 days ago.

This may ring a bell to readers familiar with famous post about Programmers / NPR history:

Turns out that while everyone loves those questions, very few are actually willing to spend any time to... maintain and moderate them.

The only recent edit was done to question about JSON vs CSV, which recently changed wording to "what are some advantages/disadvantages..." Not that it made it worse (as far as I can tell, edit didn't even invalidate existing answers) but it didn't really help. Edited question now prominently exposes issues described in another meta post, What is the problem with “Pros and Cons”?


Another thing worth paying attention to is how little of negative feedback was given to these questions.

It gets especially prominent if you account for attention given to this meta post, which is mostly about explaining what's wrong with them. It stuck in hot meta posts for about a week (quite the irony), has got almost 400 views and solid positive score. Now, let's take a look at the questions it lists:

Not much really, huh? Considering that some of these votes were cast long before the meta post, it becomes even weaker. Closer check of The Trouble with Popularity suggests that this may actually be a common case.

Note how the article says nothing about negative feedback to such questions. This makes a fairly strong indication that there is no such feedback, at least not at a scale worth mentioning. (Atwood would probably be happy to point that at least part of community actively opposes such questions but it looks like this just doesn't happen.)

Reflecting on my own behavior in these questions makes me feel that there could be a good reason for this. I tend to feel reluctant giving negative rating (down, close, delete) when post has high score. This may seem strange, given that I am well aware of upvotes to popular posts frequently degrading into useless "likes" of Facebook kind.

But the thing is, I am much more used to looking on regular, on-topic, good quality questions. And in normal questions, high score fairly reliably indicates quality content worth voting up, not down.

If many users are like me, it should indeed be that high score posts rarely get negative feedback, no matter how bad these are. People simply build an opposite habit by reading on-topic, good quality questions, where high score means that post deserves upvotes. Not surprising that they would feel discomfort trying to do opposite to what they are used to - at least that's the way I feel myself.


This probably won't make me many friends over here but summing up above, it looks like the most realistic way to address the matters of broken windows would be a moderator deletion of these questions.

Which, interestingly, brings us back to the quote that opened this question:

in the absence of any moderation, the community would do nothing but add and upvote the easy, fun stuff. This is why community moderators have real power; they need that power to intervene, educate, and refocus the community's exuberance on more substantive content...

  • 1
    ...after all, it looks like this guy Atwood knows a bit or two about trouble with popularity :) – gnat Aug 16 '15 at 18:29
  • 1
    Your delete vote counts are off. Just sayin' – user53019 Aug 16 '15 at 18:42
  • @GlenH7 down vote counts are off too. Looks like someone's trying to break Atwood's law, "in the absence of any moderation, the community would do nothing..." (not that I complain:) – gnat Aug 16 '15 at 21:02
  • 2
    (follow-up) these questions were deleted by moderators about a week ago. That probably qualifies this answer as accepted – gnat Aug 23 '15 at 12:46

You must log in to answer this question.

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