13

Anyone else finding that their instincts on what is a good question are wrong? Since questions that are off topic, on any other SE site are on topic here. Ie any software-rec. and questions that are there on topic cousin: What are criteria for X? are off topic (probably).

I am finding it hard to pose my questions in a good way. Much harder than answering questions. (Normally I'm the reverse, I think I've become quiet good at asking a good SE question, but I am rarely fast enough/good enough to write a top quality answer, before someone else does.)

How can I break out of my mold?

  • 2
    This question is hard to answer if you don't articulate how you find that your instincts fail. At least give examples where your insticts are wrong, and explain why you think they're wrong (second thoughts? a majority of votes disagreeing with you? getting crap answers on what you thought was a good question?). I find asking good questions a lot harder than writing good answers (provided I have the requisite knowledge of course). – Gilles 'SO- stop being evil' Feb 6 '14 at 10:44
23

It's a pretty simple checklist to ask a good, narrowly-scoped recommendation question. I'll make a rough outline here, I'm currently revising this (which is based on the original ground rules that I posted pre-launch).

1. Straight to the point, succinct title

Don't use words like 'best' or 'good' - just tell us what you want. We're not going to recommend the worst, or bad software. What [editor/utility/program/plugin] does [task] in [manner]?

2. Describe your task

Tell us what you're doing, or intend to do with what we recommend. If you're bulk re-sizing a bunch of pictures while converting them to another format, or backing up lots of files and trying to avoid data duplication - let us know with as many specifics as you think might be relevant.

3. Describe what you have, if anything, and what you don't like about it

This can sometimes be optional, but let us know what you've got or what you've tried and didn't like, and why. Let us know if you looked at something and decided it wasn't for you. Note - answers may recommend something you note, but might have overlooked.

4. Give us an enumerated list of constraints, in order of importance

Every recommendation question probably needs this list. Tell us the features or operating constraints a good fit would meet, ordered from must-have to nice-to-have. An example:

  1. Must run on OS/2, with 128 MB of RAM
  2. Must not be pink
  3. Ideally takes less than 2MB of disk
  4. Big plus if it plays music

5. Wrap up your question, if it needs wrapping up.

You can probably skip this most of the time if you want, but this is a place to put anything supplemental. It's hard to define that, beyond you'll know it when you encounter it.

Your goal when writing is to narrow the scope enough so that realistically, 10 answers might directly answer it. It's really not too hard and the simple exercise of just writing it all down can sometimes make you think of something you might have overlooked.

That's everything someone needs in order to give you a great recommendation. We may not be able to meet all of your requirements, but since you've listed them in the order of importance, we'll be able to recommend something that gets the most important job done.

  • 4
    "and the simple exercise of just writing it all down can sometimes make you think of something you might have overlooked." This is one thing at least that carries over from all successful sites on our model. Crafting a good question often helps you figure out the answer yourself. – hairboat Feb 6 '14 at 14:47
  • I almost wish the first 4 points were an auto-loaded template... Not as a requirement, but an aid. – kayleeFrye_onDeck Dec 17 '15 at 5:30
5

It's going to take practice. For all of us. Each member of the community will need to revisit their habits, their kneejerk reactions to questions, their judgments, etc. if they have prior experience on SE.

All I can say is practice, pretty much, and see what feedback you get from the community. If you get negative feedback, see if they will tell you why. If you get positive feedback, try to bottle that lightning.

  • Exactly. Many of the users I see around are active mSO people, which is a blessing in that they know a ton about what makes a good site and a curse in that by definition they've been doing things 'the normal way' for a long time. – Undo Feb 6 '14 at 4:48
  • 3
    The silver lining is that "MSO people" usually understand the reasoning behind the rules, so they're likely to be able to adapt better than people who just apply rules blindly and inflexibly. But yeah... it'll take some active work. I know it has for me. – Pops Feb 6 '14 at 8:20
  • @Undo Don't treat “MSO people” as a single unit. We all have our own opinion — and often duke it out on MSO. For example, I think I count as an “MSO people” by now, and I tried to make Science Fiction & Fantasy recommendation questions on-topic (they weren't because the first such questions were crap, and the experiment on Literature.SE showed that they could work but only with heavy-handed moderation). – Gilles 'SO- stop being evil' Feb 6 '14 at 10:42
  • 1
    @gilles I'm not saying that we're all the same, but by and large most of us agree with the "old rules" about opinion-based stuff and closing and stuff. Right? – Undo Feb 6 '14 at 15:02
  • Undo isn't saying we're friends.. He's just saying we all know the old school rules and most of the reasoning behind them. No one said Gilles had to agree with either Seth or Undo. – Seth Feb 6 '14 at 16:23
  • @Undo I guess we all agree that we need to close some stuff. But not on what needs closing. I don't see why this site would change any of the fundamentals: we're trying to build a questions and answers site, not Usenet or Slant or Reddit or whatever-SE-is-not. – Gilles 'SO- stop being evil' Feb 6 '14 at 17:09
  • @Gilles I don't think this site changes any of the fundamentals, but it depends on how deep you consider fundamentals to be. I.e. some might consider closing opinion based things a fundamental. I don't disagree with you at all - we don't want a Reddit, and we're all learning through this. – Undo Feb 6 '14 at 17:18
0

Hm. No idea where to put this exactly, so putting it here:

Thanks to everyone on this site here for improving my general instincts about what makes a good answer (and question) on SO/SE in general. While I still am of the opinion that to make the user happy is the end goal (my posting in the “You're all doing it wrong” thread), I've learned – just by being a part of the private beta in here.

Thanks!

.oO(How do I leave feedback like this on a site's meta? It's also structured Q⇒A-like…)

  • 1
    It is structured like QA, but you can get away with using Meta much more like a forum, esp in the discussion metaflag, (idk if this is a rule or just soemthing that is true). You can also go on chat. – Lyndon White Feb 28 '14 at 0:49

You must log in to answer this question.

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