The first question is debatable. In my opinion (and only my opinion), it falls into the realm of tools that programmers commonly use, even though lots of other people use the same tool. However, I can't reasonably articulate what makes it not a shopping question, even though it does seem rather narrow.
If it was 'Can I make links clickable in [terminal program]', I think it would be a fine question.
The second question is, and always has been objective. However, I can see why people would find it broad. As asked it's basically Concurrency is hard, how do I get it right?, where you probably wanted "I have these specific problems with Global mutexes and the mutex class, how can I avoid these problems with decoupled reusable code?" - which is the same question, just alleviating any ambiguity as to its objectiveness and scope.
As you pointed out, these are old, and asked at a time where tool recommendations were more or less tolerated by the vast majority of the community, and a time where specific quality standards were a bit less defined because we hadn't come up with them yet. Had you asked these today, you would have received instant feedback that might have avoided closure on both. I suspect you would have asked both a bit differently as well.
I think the real issue here isn't a lack of notification, it's the sting of questions that seemed perfectly acceptable several years ago being closed all of a sudden. While I see the utility value of a notice when this happens, I really feel like it would just exacerbate the sting in cases where there's just not much that could be done to make the question suitable again - which is the majority of these cases.
I don't think that's a problem we're going to have much longer going forward, as our question and quality guidelines have become much more refined and accepted. Believe me, it's no fun being on the closing end of them either, I was one of the mods who had to enforce the new standards just after being elected, it often felt horrible. I'll be very happy when 'not constructive' is redefined.
Still, the kind of notifications you're talking about would not really apply to questions asked in the recent past unless they just outright fell through the cracks, a problem that /review is doing a good job of addressing. People know quickly if their question is closed, or likely to be closed because the feedback is fast. I think the focus here should be on getting the system to the point where a question that remains open after being asked simply stays that way, changes to how the close workflow works will probably bring us much closer to this.
