I know I’ve chatted with a few people about probabilistic programming. I wanted to provide a detailed reference for folk who are more accustomed to PL, here is an in-progress grad textbook by Jan-Willem van de Meent at Northeastern: https://arxiv.org/abs/1809.10756

https://dustycloud.org/blog/racket-is-an-acceptable-python/ Just noticed this on HN

was also posted in #random ealier today :wink:

I’m really going to miss this view from Racket School when I leave!!

OK, I’m just perplexed at this: in this code snippet, DrR was correctly id’ing e1- and e3- as bound on lines 5, 7, 12, and 14, but complaining that e2- (lines 6 & 13) was unbound. Deleting & retyping the e2- seems to have fixed the problem. Wonder what that was all about.

e2– and e2- are different. The left one is unicode en-dash. The right one is ASCII hyphen.

Gah. Now that you mention it I see it. Wonder how that got in there.

ahhh unicode

constant source of joy

ẏ̸͔̤̥̱̱̘̬̰̼̦͍͚̽̑͌̅͂̔̇̀̆͊͂̎̽͋̚͜ȩ̸̡̢͖̥̠̱̺̟͈̞̫̘͕̦̞͂̌a̸̢̡͍̯̘̟̱͙̠̳̫̺̫̗͊̓ḧ̴̛͎͕̞̣͖̔͂̏͝ ̴̛͎̓͗̍̓̑̾̈́̐͛̄̈́̆̽̇͌̕ǔ̶̧̬̼̦̜̗͓͖̪̗̦̻̭̬̓͜͜ǹ̸̨̗̞̳͙̗̬͚̳̼̭̗̱̞̗͎̿͊̀͊̄̏́͝ͅi̶͍̗̪̗͙̹̣̩͎̍ͅc̷̳̒̈̿͜o̵̺͙͠d̶͔̲̟͈͔̜̭̄̈́̏̎̎̈́̎͐͌̈́̓͘͘͠e̵̱̙͊̈́̕ ̵̢̧̲̼̺͔̹͉̘̩̳̖̻͙̜̎͑̎͑̊̃͊̒̄͂̚̚͜i̶̧͇͔̲͎̰̭͙̅͘s̴̢̮͖͖͉̠̔̽̓̒̎̾̾̈ ̴̧̢̛̟̬̣̟̤̺̮̺͍̮͚͙̖̈́̽̿͗̏ġ̶̭̜̱̜͕̫̼͚̙̮̙̩͋̊͊͛͆̄̅̔͗̌́̇̚͠ͅṙ̴̡̢̡͔͕̘͈̫͚̹̳̯̰̮̹͆͛̽̒ę̷̤̤͇̞̝̫̟̻͔̮̣͈̱̦̤̖̂͛͗̉̎̔̕͝͠͝ạ̵̤̱̝͙͕̝̍̂̈͌̑̎͋̍̀̇̂̅̓ͅt̵̢̨̘͍̟̖̱̺̯̙͔̰͋̚͜͝

eris those error messages

at least now my factorial works as well, but I have the same porblem that @me1531 had today: I am not sure my type inference actyally does what it should :confused:

lets try coming up with some tests

wish I could participate. too busy deleting slides. you can tell by how on slack i’m not

out of curiosity, does anyone know how this support in DrRacket is implemented for figuring out this binding:

it splits the single identifier body:expr that contains a colon and figures out that body is bound below.

^ this also seems pretty magical. How does this work? special-case support in DrRacket? Or some more general functionality that we can hook into as well?

@me1150 that uses the 'sub-range-binders property, see here: https://docs.racket-lang.org/tools/Check_Syntax.html#%28part._.Syntax_.Properties_that_.Check_.Syntax_.Looks_.For%29

Oh neat! Thank you for the link

is there a (built in) way to have racket infer the type of an expression in our STLC example?

(and print the resutl)