It seems to me the trend in the on/off topic question really comes down to "What's subjective but not subjective to the point of pointless and/or uninteresting."

So far it seems that we're making our way trying to find good vs. bad questions and I think that's a pretty good approach. The downside is that without a definition, I think we leave ourselves open to "if asking about lighting for programmers was good, then why was asking about shoes for programmers bad?" type questions.

So, should/how do we capture this in the FAQ or elsewhere?

  • 1
    This is already a very subjective question you know...
    – Graviton
    Sep 30, 2010 at 12:04

5 Answers 5


Certainly we need. This is a work in progress and will be a hard work. It seems pretty subjective say what is subjective :-)

In Beta phase some undefined is normal. We're discussing here and on meta.stackoverflow.com. Your contributions are welcome.


Looks like the almighty ubermods have answered this...

The Six Subjective Question Guidelines -- Enforcement Notice


As such questions get closed, we see discussions like these arise on the meta, reopen votes.
And exceptionally this could start over again if the question does get opened again...

So yes, we certainly need a simple definition to understand which at the same time limits clarity discussions.

As bigown stated in his own answer:

It seems pretty subjective say what is subjective :-)

So it will take some time for a perfect simple understandable definition to form.

Another example of this subjectivity can be seen in my comment on ChrisF his answer:

Sorry, but lightning isn't unique to them either...
A lawyer typing a text document at day or at night will have the exact same light issues.

I guess a large share of people not agreeing over things is what slows this discussion down.

I'm not into statistics, but it might even be that there are enough people that do want these questions
thus try their best to beautify them with subjective comments. This made me think...

If no good definition can be derived in a reasonable time. why not allow/ban them all at once?


Well, to take your example of lighting vs shoes, here's my take:

Lighting is OK because it affects the environment we program in, and while there will be some discussion over the details in general there is a broadly "right" set of answers.

Shoes is not OK because while bad shoes can cause problems, they don't cause problems that are specific to programmers developers. Also the answer is "buy a new pair of shoes".

So does the question raise an issue that's specific to, or affects in a specific way, software development? If so then it's OK to be asked here.

  • @ChrisF: He post it because this question: programmers.stackexchange.com/questions/7315. As I've not experience yet, I don't know what the best approach in this case as I stated on my comment in question. If I did something wrong you can contact me on chat or my mail.
    – Maniero
    Sep 25, 2010 at 22:47
  • @bigown - I think that question fits with "shoes". It doesn't really uniquely affect (or affect in a unique way) software developers.
    – ChrisF Mod
    Sep 25, 2010 at 22:50
  • @ChrisF: It's my opinion too, my doubt is about my comment on that question.
    – Maniero
    Sep 25, 2010 at 23:04
  • 2
    -1 Sorry, but lightning isn't unique to them either... A lawyer typing a text document at day or at night will have the exact same light issues. Also, the only possible answers is broken up in two sections: Setting the brightness of your monitor which is quite objective, improving the lightning of your room which is more a kind of a home improvement thing. Related jokes: "Q: How many programmers does it take to change a lightbulb? A: That's a hardware problem." or "Q: How many developers does it take to change a lightbulb? A: The lightbulb works fine on the system in my office. NOT REPRO." Sep 26, 2010 at 1:05
  • @TomWij - I think @ChrisF was using "unique" in a relatively loose sense. Certainly many of the work environment type questions are going to be applicable to anyone who drives a keyboard most of their workday. It's more like "well targeted" or something like that.
    – MIA
    Sep 26, 2010 at 15:38
  • @Jim - I think you're right - "unique" is too strong, but I'm still struggling to think of something better.
    – ChrisF Mod
    Sep 26, 2010 at 15:40
  • @bigown - I think I posted this before I voted to close that one... or was at least thinking about posting this question. But that q? became a case in point. ;)
    – MIA
    Sep 26, 2010 at 15:44

No, you can't have a definition for subjective. Because it's subjective. That's what troubled SO for so long, and eventually lead to the banishment of the subjective tag - no one could agree on what it meant.

Time isn't going to fix this. Discussion isn't going to fix this. Not in any reasonable time-frame at least - or it would already have been settled on SO.

If "subjectiveness" is the criteria by which questions are judged, then everyone's just going to argue that their favorites are on the low end of the scale...

...But the current arguments aren't really about "question subjectiveness" anyway. Some of the commonly-cited "good" questions are extremely subjective. The problem is that the obvious rules for separating on- and off-topic questions (strictly programming related, no GTKY/write whatever you want) exclude some rather popular questions - so instead, you get these arguments in favor of more (ahem) subjective criteria:

  • "answerability"
  • "utility"
  • "seriousness/professionalism"
  • "potential for long-term benefit to humanity"

...and those are at least as difficult to nail down as "subjective".

So to answer your second question: forget it. You're not gonna capture this in the FAQ in any useful fashion. I say we just admit this right at the start, and codify it thusly:

What kind of questions can I ask here?

Questions about things programmers care about. Good ones.

How do you know if your question is good enough for Programmers.SE? You don't. Question "goodness" is hard to define, and so we're not gonna try. But we know it when we see it, and if we don't see it, you're outa here!
Have fun...

  • 1
    You might've been facetious in suggesting it, but I think the Miller test is the right approach. Before that, Programmers.SE needs a house-cleaning to set precedent of what counts as good and what doesn't, and thankfully, that's going to happen.
    – user8
    Sep 26, 2010 at 3:23
  • @Mark: well, I don't think anyone will actually dare writing that into the FAQ, but near as I can tell it's the only approach that's practical short of just killing 90% of the questions and moving the rest back to SO. Boundaries are made to be pushed - if you can't build 'em strong, don't build 'em at all.
    – Shog9
    Sep 26, 2010 at 4:40

You must log in to answer this question.

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