1

During my participation of SO I often recognized a pattern regarding supposedly subjective questions.

The questions looked like this:

  1. Why is feature X in product Y implemented this way?
  2. Why is there no support of X in product Y?
  3. Why would one choose practice A over practice B?

The expert answers looked like this:

  1. Because with this kind of implementation you gain the advantage of ...
  2. Because the developers wanted to prevent misuse of ...
  3. Because practice A provides the advantage of ...

The non-expert answers and comments looked like this:

  1. No reason, it's a matter of opinion.
  2. If you don't like it, don't use it.
  3. -1 This is a subjective question and should closed (or at least be CW)

To summarize: For some "why" questions the (rare) expert actually knows the reason, but the (common) user does not and thinks "this can only be subjective". Especially in cases where the expert answers don't come soon enough, the down and close votes start to accumulate.

How can one handle these questions which at first glance smell like subjective questions but actually can be answered with precise facts? (Both from the perspective of the questioner and the perspective of the answering SO user)

Improve this question
6
  • 10
    The problem with your examples 1 and 2 is that only the people who developed the product would know and the chances of them seeing the question are slim.
    – ChrisF Mod
    Commented Sep 8, 2011 at 20:17
  • 1
    Duplicate of a question by Eric Lippert: meta.stackexchange.com/questions/96310/…
    – Justin Morgan
    Commented Sep 8, 2011 at 20:43
  • 2
    Isn't Eric an example of a person who could actually answer questions 1 and 2?
    – Bo Persson
    Commented Sep 8, 2011 at 20:56
  • @Bo Persson - yes, but what are the chances of him seeing any particular question amongst the 1,000's that are posted daily?
    – ChrisF Mod
    Commented Sep 8, 2011 at 21:53
  • 1
    @Chris: he's pretty good in the C#/.Net tags when it comes to answering useful "why" questions. Otherwise, I'm not aware of other tags where the community has the same level of designer involvement.
    – user7116
    Commented Sep 8, 2011 at 22:33
  • @sixletter Alex Martelli used to be kinda like that on the Python tag. Wonder where he's gone.
    – NullUserException อ_อ
    Commented Sep 9, 2011 at 1:50

2 Answers 2

Reset to default
2

I look at some factors: How coherent is the question? Is it a thinly-disguised flame (these often are)? How likely, really, is it that someone like Eric Lippert will show up and answer the question? How useful is the answer to anyone -- will it actually help anyone accomplish anything to find out?

Often, these are questions about design tradeoffs in Java or C++, where the history is lost in the mists of time. The only answers that show up are bloviation or confabulation. (Windy opinions or 'just-so' stories.)

Perhaps 'subjective' is the wrong reason, and 'not a real' would be more on point.

Improve this answer
13
  • 1
    I usually do Not Constructive. In the C#/.Net tag I carefully weigh the usefulness of the question, and if not useful -> Not Constructive. I started treating those tags differently after seeing the rabid participation of Eric Lippert in those tags, which has benefited us all.
    – user7116
    Commented Sep 8, 2011 at 22:32
  • @six Why should other tags be treated any differently?
    – NullUserException อ_อ
    Commented Sep 9, 2011 at 1:58
  • 1
    I would trade 100 trivial, boring questions that anyone can answer for one question that actually makes you think about a language - and hopefully, better understand it.
    – NullUserException อ_อ
    Commented Sep 9, 2011 at 2:02
  • Go right ahead, but not on stackoverflow.com.
    – Rosinante
    Commented Sep 9, 2011 at 2:08
  • So you'd rather have the same kind of boring questions asked over and over again ad nauseum – and learn nothing from them – than one question that could actually teach you something ? Interesting attitude.
    – NullUserException อ_อ
    Commented Sep 9, 2011 at 2:25
  • 1
    @Null: as they have language designers answering language design questions, rather than folks pontificating. Not saying other tags don't have the same thing, just I've not been around to see it.
    – user7116
    Commented Sep 9, 2011 at 2:39
  • @six So the problem here is not the question itself, but the people answering it without actually knowing the answer.
    – NullUserException อ_อ
    Commented Sep 9, 2011 at 2:44
  • 2
    @six perhaps, but the odds of getting 1 Eric Lippert are astronomically low, while the odds of getting 1,000+ "gee, I dunno, but here's my opinion and guess on the matter" users is ASTRONOMICALLY HIGH. So I think closing is actually correct, and if someone like Eric Lippert wanders by, let's just handle that as the super rare exception it is...
    – Jeff Atwood
    Commented Sep 9, 2011 at 4:02
  • @Jeff I believe you were addressing me instead. FWIW I think we actually have a couple other "Eric Lipperts," like John Resig, the creator of jQuery, Tom Christiansen aka "one of the guys who wrote Programmming Perl" and the aforementioned Alex Martelli - who unfortunately has disappeared.
    – NullUserException อ_อ
    Commented Sep 9, 2011 at 5:45
  • @six what Jeff said, plus: so is not a discussion forum, and anyway, 99.99% of the time, nothing remotely interesting appears.
    – Rosinante
    Commented Sep 9, 2011 at 14:43
  • @Ros: I'm with you and Jeff, I vote to close! Just when someone like Eric shows up and answers the question I let it ride.
    – user7116
    Commented Sep 9, 2011 at 20:09
  • @sixlettervariables: "Rabid"? Really?
    – Eric Lippert
    Commented Sep 12, 2011 at 18:02
  • @Eric: haha, perhaps I'll use a different adjective. But with 1300+ answers in the C# tag, it's a fair lot more than the rest of your tags.
    – user7116
    Commented Sep 12, 2011 at 18:46
2

I think cases 1 and 2 should be left alone, until someone who is qualified can answer them. The non-expert answers/comments you point out are definitely all the wrong way to handle it.

Improve this answer
4
  • 1
    Oh? How will the system clean them out? The system is full of useless, pointless, unanswered questions from 2008.
    – Rosinante
    Commented Sep 8, 2011 at 22:03
  • There is some mechanism for that, though I have no reference handy.
    – Lance Roberts
    Commented Sep 8, 2011 at 22:07
  • I edited out the system reference, since I can't find a reference to validate it.
    – Lance Roberts
    Commented Sep 8, 2011 at 22:33
  • 1
    @Rosinate It won't. Just because no one can answer it yet it doesn't mean it's an invalid question. People are too overzealous when it comes to closing questions on SO.
    – NullUserException อ_อ
    Commented Sep 9, 2011 at 2:00

You must log in to answer this question.

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