In short, HtDP is a book about programming, whereas PLAI is a book about programming languages. You can think of PLAI as teaching you how to implement the tools you learn how to use in HtDP.
They both come from members of the PLT research group, so share a basic philosophy. In fact you use the very same tools (mostly structural recursion) in implementing languages that you learned in HtDP.
I also enjoy this "trend", but I think it's worth noticing that a small number of prolific posters can create a meaningful shift in the content that makes it to the front page.
E.g. rspivak who posted this submission, has also shared a number of other very interesting language related submissions over the past couple of weeks, many of which I remember seeing on the front page [1].
I take this as a reminder both that (a) individuals on HN can shift the conversation and (b) for that reason, if you think there's something missing in what's visible here, you can fix that.
HN topics tend to run in cycles. A couple months ago we were on an aviation kick, with a lot of stories about old high-technology bombers like the B-70 or A-12. Once, a few years ago, the entire front page was taken up by Erlang stories. Summer 2013 was all about Snowden & the NSA.
Don't extrapolate this into a trend. It's more that seeing an interesting story tends to prime people to notice other interesting stories on the same topic. This is also why you get press cycles - notice that the other topic that's been recurrently popping up on the HN front page is Theranos (and data breaches, and how we now have too many JS frameworks).
Eventually all these topics will pass, people will continue writing programming languages, and HN's collective attention will move to something else.
The reason I went into programming languages when I started my career was that everyone in CS uses them, needs them to be better, and has an opinion on them. You could even extend that to a lot of people in technology or science in general. A lot of physicist, biologists etc these days work primarily in programming languages, so they effect how they work as well.
Compare that to something like machine learning, which of course is extremely significant and important, but if I meet a random programmer they may know nothing about machine learning and may not be interested in any machine learning research. But they will know about a programming language and have opinions on it.
Doesn't explain why it's suddenly more popular now, though.
the main reason I'd look into programming languages is because I feel some problems can't/shouldn't be solved with the same tools that created the problems. Problems that are currently not solved: Dealing with low-level systems programming, building parsers for data
One of the benefits of learning how to create a language is you get better at analysing the various new languages that are popping up. Nothing like learning how the sausage is made. So now whenever I look at new language, I mentally take it apart to see what the underlying structure is, then I process the actual syntax (which ends up being the least important part) last.
I like this trend, but this change has made me curious. Why?
I'm guessing that it's a subject that most on HN can relate to, yet one where no-one who programs is completely happy with the languages and tools they use today. It's a rich area for potential new developments and ideas, and no doubt many of us have thought about what we'd like to do if we invented a new language ourselves. That makes it both an interesting topic for discussion and an interesting field to observe even for those of us who will never personally have the time and conditions to develop the next great language.
