It is easy for bad soft questions to get lots of votes and responses because MO has a large lurking population, and the more people understand and like a question, the more people will vote for and answer that question. These are precisely the kind of people who haven't read the FAQ very closely and who therefore won't necessarily act in agreement with it. On the other hand, it's harder for legitimate technical questions to get votes because many people feel less capable of evaluating whether those questions are actually good, even though they are precisely the questions we're trying to encourage! C'est la vie.

You also really need to mention how old these questions you're looking up are. MO policy has changed a lot since last year, and we're a lot stricter about soft questions than we used to be.

I was trying to explain why it is difficult to ask good questions about mathematics education.

When MO first started up, most of the rules and policies were not as strict as they are today, and a lot more bad questions fell through the cracks. Since then, the closure policy has evolved to take care of new problems as they arise (in the early days, Scott and Anton were singularly responsible for the vast majority of closures, for example). The newer policies here are often proposed by the moderators and put into effect after some debate. If you want to debate the merit of Scott's new closure policy on pedagogical questions, you should engage it on its own merits. We acknowledge that this is a change (justified by a reinterpretation of the goals and rules of MO), so arguing that the policy wasn't followed before is silly.

as has been said many times on meta (I understand you're not a regular, but still): the existence of a unclosed question that is strictly worse than some particular closed question is not regarded as evidence that the closed question was closed in error.

(For the record, I had nothing to do with closing or reopening the question that prompted this thread -- my involvement so far has been trying to explain why "subjective and argumentative" are bad qualities for a MO question. In the present, revised, form, I think it's an excellent question.)

Quoting from the FAQ

As a side-effect of being very good for to-the-point questions and answers,
the Stack Exchange software is bad for disscusions and designed to minimize
them. There's a place for discussion about mathematics, but it isn't MathOverflow. 

Hopefully by now we all agree with this --- MO questions shouldn't explicitly encourage discussion or argument. Rather, they should aim to provoke definitive answers, that can be identifiably "correct". Now, this is obviously difficult with many mathematics education questions. I think the best way to do this is to phrase questions as neutrally as possible, and to ask for answers that are not just opinion or anecdote, but can be backed up by evidence. Ideally, evidence provided by actual studies of mathematics education. (I think I heard this first from Ben Webster or Scott Carnahan: hopefully they can argue this more convincingly for us!)

Now, I appreciate that this ideal is a high one. I certainly agree that in many cases anecdotes from experienced teachers are very valuable, and we should respect and encourage these when appropriate. But I think we should actively discourage people like me, with limited teaching experience, from jumping in with answers that are merely opinions. A good mathematics education question will be asked in such a way that makes this clear!

Finally, given the extra difficulty in asking mathematics education questions that will produce impartial, evidence-based answers, I think it is particularly important that these questions are not themselves too opinionated! This can only hurt the chances of the resulting answers. I think in the case of this particular question, the initial phrasing was quite opinionated, and this may have provoked the initial closure (which, I reiterate, I had nothing to do with!).