In mathematics, as many of us already know, generality is a good thing. We aim to prove a result as general as possible.

However sometimes generality is ambiguous and gives us no help in trying to solve a particular case - whereas there are particular approaches which are applicable and useful. One good example which comes to mind is Ramsey's theorem. In its finite case the proof is combinatorial and a bit difficult to obtain, whereas in the infinite case the proof is quite simple and one can deduce the finite case by a compactness argument. But the infinite proof gives us no intuition on how to solve the finite case on its own.

Similarly on meta we try to be broad and general, we try to ask questions which apply to a particular case at hand - but that will be useful later on in an obvious way. What is not obvious sometimes is how the particular case is different than the general case, and how to apply it.

As a result there has been several cases when the community spirit pointed that a certain action is appropriate, whereas many have also agreed that a particular case is inappropriate.

This leads to a [reasonable] question, how to find the proper amount of generality for meta posts? And when should we draw the line and post particular problems, instead of general ones?

share
3  
If people would have it, I do have some thoughts on this on a general sense and not as from some manner of official statement. –  Grace Note Dec 21 '12 at 15:20
    
@Grace: I have to admit that I don't fully understand what you are saying. But if you're saying that you have something to say on this, but as a person and not as an official StackOverlord, then I think you should say it and put a disclaimer on the top. Surely you have enough experience over all the meta sites than most of the users here combined, and we could all benefit from this experience. –  Asaf Karagila Dec 21 '12 at 15:28
    
That's pretty much exactly what I was saying. –  Grace Note Dec 21 '12 at 15:29
    
I just wanted to be sure. :-) –  Asaf Karagila Dec 21 '12 at 15:39
4  
Shouldn't this question be posted to meta.meta.math.stackexchange.com? –  Gerry Myerson Dec 21 '12 at 16:44
6  
@Gerry: meta. is idemptotent. –  Jonas Meyer Dec 21 '12 at 16:51
3  
@Gerry: Is this the site for the meta-math SE site, which is about metamathematical theorems? –  Asaf Karagila Dec 21 '12 at 16:52
    
I wonder what the downvote signifies. Perhaps someone that doesn't want to ever deal with the particular case? –  Asaf Karagila Dec 21 '12 at 19:58
    
We could keep volumes upon volumes of case law and precedents. –  Baby Dragon Dec 28 '12 at 12:37
    
@BabyDragon: Say what? –  Asaf Karagila Dec 28 '12 at 12:39
    
@AsafKaragila I'm only kidding. –  Baby Dragon Feb 11 '13 at 18:35
add comment

1 Answer 1

I'm Grace Note, a Community Coordinator at Stack Exchange. I'm dropping by here right now not in an official capacity, but just to give a viewpoint as a simple user of the network. Basically, I'm just giving another viewpoint on the subject.


I like to look at two basic things when thinking about how to scope a meta discussion - how it helps the community and how it helps resolve the issue. I actually apply these concepts to a lot of different spectra, so let's see how it relates to the general cause versus specific issue gradient.

  • Is discussing the problem on a general scope beneficial to the community? Identifying everything as part of some general scope certainly makes judgments easier, but sometimes, it really is more helpful to look at the problem on a case-by-case basis. This may be because a draconian guideline must be developed just to address the specific instance, or it could just be taking extra time to solve a problem that won't resurface. Basically, how does working on the general scope differ from working on the specific, and is it a good idea.

    On our fresh new Anime & Manga site, we recently had a question from a user asking "Are we making too many rules?" It came up because three or four questions in a row all spurred discussion about major guidelines, all of which were pretty isolated and users were unsure that they would ever run into it again. In their case, they were still in private beta, so it helps to discuss nearly everything that comes up. A point was made nonetheless, though, that it's important to take a step back and confirm whether or not every point and essence needs to be regulated when it comes up.

    End line of this is, if discussing the general scope would be detrimental to the community, such as by being a waste of time over a very minor occurrence or by being too complex of a problem for the application of a general rule to be healthy, the specific issue is a lot more helpful to look at. Sometimes, the general scope at all is distracting from productive discussion - this is what can often happen when buzzwords like "list question" or "meta tag" come into play.

  • Is discussing on a general scope beneficial to resolving the problem at hand? A broadly applicable rule helps create structure for users to be able to enforce. Without hard rules, it makes the task a burden. But it is important for users to be open to the fact that there can always be exceptions. Coming to a general conclusion is not helpful if there are unique identifiers or exceptional circumstances to the issue that render the ruling ineffective.

    Look at what prompted you to bring up the discussion, then look at what you chose to discuss. Would a solution to the general problem resolve this situation properly? Is the general problem even relevant to what inspired you to come to meta? If either of these is a no, then it means that asking about the general issue is avoiding the real problem. It's always important when analyzing anything, to look at the core of the issue. In the same vein that one should look at what someone is truly looking to solve versus what they wrote in the question body, one should look at what truly is the cause for concern versus what it can be grouped into as a set unit.

    Sometimes this kind of analysis, by the way, helps identify means to divorce the specific instance from the general problem. That is, the question that is being talked about can be revised in ways that the general isn't even applicable at all.

share
    
Thank you very much for your answer. It is very helpful! –  Asaf Karagila Dec 21 '12 at 19:18
add comment

You must log in to answer this question.

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