Note: Here, when I say question, I typically mean support and discussion questions. Though this marginally can apply to feature request and bug reports as well.
So, I recently spent a good chunk of time writing this question: Pros/Cons lists comparing two methods/technologies/etc. Constructive or not?
I tried to make it the ideal "is this appropriate" question. I put forth existing questions and blog posts about what I thought was a similar, but not quite the same topic. I then put my arguments in for why these existing stances maybe don't make sense for this behavior. I was expecting for it to be a rather well accepted question(ie, get upvoted), but most likely with an answer saying "no it isn't appropriate for these reasons ...".
I got good answers for why it's inappropriate, but for even mentioning the subject, instant 10 downvotes and 4 (probably) sympathy upvotes. I don't care about reputation, and I know votes on meta have traditionally been different, as mentioned in the FAQ as well. However, I've stopped coming to meta because no matter how much you say "votes don't matter", it does bug me when I spend a lot of time building a good quality question only for it to get downvoted into oblivion.
I can really see why people rage-quit from meta now. No matter in how bold of font you say votes don't matter on meta, people still feel bad when their perfect quality question gets downvoted.
Why can't we (somehow) change it so that downvoted questions/requests mean "bad quality" and not "unpopular"? And then to show disagreement, upvote(or downvote) an answer that says "no this won't work because of X".
This isn't a technical problem, it's the mindset of the community. However, I think we scare off a great number of people because it's traditional for a users first post to be "why was my question closed" only to be followed by 20 downvotes of "disagreement that it should be reopened".
I have no idea how to solve this, but I do think it's a problem. And admitting that it's a problem is a big step toward a solution :)
References for "downvote for disagreement" just from my recent questions:
Why was this question on unit test generation closed as not constructive? -4(read the comments for some interesting discussion about this issue.. and also, this one before Shog9's edit, probably deserved some of the downvotes)
(also, if you disagree it's a problem make sure to follow convention and downvote into oblivion)