I asked Where do I find conceptual documentation for Windows Runtime? which was fairly quickly closed a "not constructive". Despite my pleas in comments as soon as the first close votes appeared, nobody explained to me what was nonconstructive about it. Now I'm confused.
I suppose those who voted to close did so because they thought my question was just a badly disguised rant. (If anybody can see a different motivation for closing, please do inform me). But since I do want an answer -- an actual answer going
you should RTFM [here](http://bla.bla/bla), not just validation of my frustrations -- can someone help me with how to ask the question in a way that won't get it closed.
The rant-sounding parts of the question was an honest attempt to avoid misunderstandings. There was an earlier question by someone else which got an easy to answer that is not actually helpful to me, and so I explained why it is not helpful, in the hope of preventing that misunderstanding and cluing the reader in to what it is I need.
I also put in a long series of deperrately-phrased example questions that I would like the documentation I'm trying to find to answer. That's because, since I'm looking for resources to help fill my ignorance, I need to describe how I'm ignorance to the reader so he can know what I want -- and simply providing a comprehensive example of my ignorance seemed like a good way to describe it.
It's true enough that what I ultimately want to know is something one could "write an entire book about" -- but I was closed as "unconstructive", not NARQ. And it's clear to me that somebody at Microsoft must have written that book already, somewhere, because there are actual people out there writing WinRT apps, and the documentation I've found so far would not in itself have enabled them to do so. Perhaps (even hopefully) it's a set of web pages rather than a book, but I need to know where it is. Something like this comprehensive introduction to COM, just with "WinRT" instead of "COM".
Or is everybody out there simply fumbling in the dark without documentation, doing what seems to work in practice, despite not having any way to know whether its working is part of the contract or just an incidental feature of the implementation they're testing with? Raymond Chen will have their heads.
What do I do now?