asyncio is the best thing about Python.

Hey John.

That run_subprocess_with_callback depends only on python standard lib so you can copy the file into your project instead of depending on pyedifice. See comments in the source file.

For years I've been trying to convince people to use monadic parsers instead of regex for everything.

hackage.haskell.org/package/repl...

Your essay about bicameral syntax is the first time I've read a principled reason (other than speed) about why a regular parsing pass is a good idea.

So, thank you Microsoft, I guess.

Thank your for posting that.

In 1997 Wadler and Jones published the Haskell language, so now in Haskell we can program do-this-thing-then-do-that-thing by using only IO Monad expressions, no statements. Happily, forever.

Wadler solved the problem of saying do-this-thing-then-do-that-thing using only expressions by adopting the Monad described by logician Eugenio Moggi.

homepages.inf.ed.ac.uk/wadler/paper...

Act III Climax and Resolution

In 1987 the great computer scientists Simon Peyton Jones and Philip Wadler decided that the good part of computer programming was the expressions, and the bad part was the statements.

Later, at IBM, Backus tried to create a language named Function Level which could say do-this-thing-then-do-that-thing using only expressions. IBM never published the Function Level language, and the source code was lost.

worrydream.com/refs/Backus_...

In his Turing Award lecture entitled Can Programming Be Liberated from the Von Neumann Style, he condemned his language, Fortran, and all other languages, as “fat and weak.” Backus warned the that languages with statements would “make their expressive weakness and their cancerous growth inevitable.”

Act II Rising Action

In 1977 the great computer scientist John Backus decided that the good part of computer programming was the expressions, and the bad part was the statements.

Landin abandoned his effort to rid programming of statements and joined the Gay Liberation Front.

www.cs.cmu.edu/~crary/819-f...

In Landin’s paper The Next 700 Programming Languages, he invented the ISWIM language “biased towards expressions rather than statements,” but he failed to excise the statements. He failed because he couldn’t figure out how to say do-this-thing-then-do-that-thing in ISWIM by using only expressions.

banger