- Location
- USA
- Pronouns
- He/Him
Oh, that reminds me:
@Jello_Raptor , remember back in the long-ago when you told me that Haskell is better than Racket at everything as long as you'resmart enough to understand it willing to do the hard work of wrapping your head around something that most non-math people find unintuitive? Well, it may be time for me to look into Haskell more seriously.
There has recently been a conversation on the Racket-users list that went like this:
Racket core developers: Hey, everybody! We've decided we don't like Lisp syntax anymore, so we're going to rewrite Racket to be more C-like. Infix operators, operator precedence, no parentheses, significant whitespace... The whole nine.
Literally everyone else: Haha, that's hilarious. You guys are the best.<3
RCD: We're not kidding.
LEE: ...
RCD: Yeah, we want Racket to be more widely used and we've decided that the main reason it isn't widely used is because it uses too many parentheses, so we're getting rid of them.
LEE: Haha guys. It's funny. You can stop now, okay? You got us. You win.
RCD: Really not joking. Don't worry, though -- the old Racket code will still be supported.
My cofounder: You know, if you want to make Racket more widely used, the lowest-hanging fruit would be a richer set of libraries.
RCD: Nah. We think we'll get more use out of completely rebuilding the syntax from the ground up in a way that adds no useful semantics but does introduce all the normal bugs that you get with operator precedence and associativity. Oh, and it will be unfamiliar to all the current Racket programmers too, so that's a win.
Me: Who hurt you guys?
RCD: Hey, we're all in academia and we know that many of our first-year programming students find parentheses confusing. Therefore, clearly, getting rid of parentheses will result in increased adoption by experienced professional developers.
...I feel like there really needs to be a punchline here, something that wraps up the whole story in a witty and satisfying way. I don't have one.
Oh, want, yes I do!
Guess how they're going to implement all this? It's a #lang written on top of a modified version of the current Racket interpreter. A middle layer will consume the input source, convert it into parenthesized Lisp code, and pass it to the bytecode compiler.
Seriously, I couldn't write this if I tried. And if I did, all y'all would tell me to stop pulling such BS out of my butt because it was breaking your SOD.
@Jello_Raptor , remember back in the long-ago when you told me that Haskell is better than Racket at everything as long as you're
There has recently been a conversation on the Racket-users list that went like this:
Racket core developers: Hey, everybody! We've decided we don't like Lisp syntax anymore, so we're going to rewrite Racket to be more C-like. Infix operators, operator precedence, no parentheses, significant whitespace... The whole nine.
Literally everyone else: Haha, that's hilarious. You guys are the best.<3
RCD: We're not kidding.
LEE: ...
RCD: Yeah, we want Racket to be more widely used and we've decided that the main reason it isn't widely used is because it uses too many parentheses, so we're getting rid of them.
LEE: Haha guys. It's funny. You can stop now, okay? You got us. You win.
RCD: Really not joking. Don't worry, though -- the old Racket code will still be supported.
My cofounder: You know, if you want to make Racket more widely used, the lowest-hanging fruit would be a richer set of libraries.
RCD: Nah. We think we'll get more use out of completely rebuilding the syntax from the ground up in a way that adds no useful semantics but does introduce all the normal bugs that you get with operator precedence and associativity. Oh, and it will be unfamiliar to all the current Racket programmers too, so that's a win.
Me: Who hurt you guys?
RCD: Hey, we're all in academia and we know that many of our first-year programming students find parentheses confusing. Therefore, clearly, getting rid of parentheses will result in increased adoption by experienced professional developers.
...I feel like there really needs to be a punchline here, something that wraps up the whole story in a witty and satisfying way. I don't have one.
Oh, want, yes I do!
Guess how they're going to implement all this? It's a #lang written on top of a modified version of the current Racket interpreter. A middle layer will consume the input source, convert it into parenthesized Lisp code, and pass it to the bytecode compiler.
Seriously, I couldn't write this if I tried. And if I did, all y'all would tell me to stop pulling such BS out of my butt because it was breaking your SOD.
