Even if backticks have the advantage of using less storage space, making edits just to change them is pointless - because then there are two versions of the post to be stored, rather than once slightly unoptimal version.

Algorithmically distinguishing incomprehensible wall of text from a valid question won't be trivial, I'm afraid. Questions that are obviously (for a human) not worth clicking would be uncloseable forever.

Two weeks ago: Question closed as unclear but an answer was already posted

I don't think we should classify it as off-topic just because it's not the intended way of using this hardware. It's hacky and everyone trying it should be aware of it, but we've probably been answering many questions with hacky answers. I've quickly looked through the Hackintosh meta questions and I think being hacky wasn't even mentioned as an argument for banning them, quite the opposite actually. This may apply to converted Chromebooks too.

Thanks for the stats link. Unfortunately it's hard to say how many of these questions actually should stay closed. And I can't think of a practical and objective way of checking this.

@JourneymanGeek It takes 1 diamond vote to close a question immediately, but 5 regular votes to reopen it - it may never happen, and if it does, it takes time and leaves bad aftertaste. If a question is closed with 5 regular votes and I disagree, it's less likely that it was a mistake. Either I'm missing something or there's a disagreement in the community that has to be addressed. Either way a quick reopen isn't the way to go. But with mod♦ voting powers it takes 1 click to close a question by mistake and in such cases I see flagging as the fastest way to get this straight.

PS. While including a base64-encoded DEB file in the answer would be possible and would remove dependency on the link, I think it's not a sensible thing to do ;)

"Imagine what would happen if the link is no longer available" - I agree that link-only answers are bad and we should encourage including all necessary information in the answer. But in this case it's impossible/impractical. This answer is as good as it gets: it's useful and includes as much as can be included realistically. What about other questions where the solution is to install a particular driver? Here's an example, answer score +27. Another one but for a problem-focused question, answer score +16.

@Ramhound This joke. Sure, we can frame closing a question as an opportunity and that's a valid way to look at it. But simultaneously it's an ultimatum: either you fix your question or you don't get any answer. And I'm afraid due to human nature rather than "I can't get an answer because my question is bad", users will think "I can't get an answer because they've closed my question". It's a systemic problem and I'm kind of questioning one of SE's fundamental principles here, but does the community actually benefit from closing as unclear?

Regarding closed questions with accepted answers. There are just two (maybe three?) in December 2020 that shouldn't have been closed IMO, so it's not much of a problem. I agree with your point though, too many otherwise answerable questions are getting insta-closed.