Notices by Andy Wingo (wingo@mastodon.social), page 4

full moon awoooooo

new bloggery: in which i see about adding freelists to the no-freelist space
https://wingolog.org/archives/2025/08/07/whippet-hacklog-adding-freelists-to-the-no-freelist-space

@whitequark lmao, that is amazing

i always thought faré was a bit of a turd and i am sorry to have my prejudices confirmed! https://chaos.social/@dpk/114977579680616235

ahahaha mccarthy was also salty about reviews in the 50s http://jmc.stanford.edu/articles/lisp/lisp.pdf

new gc just* dropped from tianle qiu and steve blackburn: "Iso, Request-Private Garbage Collection" https://www.steveblackburn.org/pubs/papers/iso-pldi-2025.pdf

* ok, june, but i hadn't seen it yet

as far as i am aware, *everyone* without institutional access who looks up information in research and libraries thinks open access is important.

the handwringing « nuance » only comes from those that already have access; but OA benefits don’t accrue to them anyway!

the 7yo is telling the 5yo that if she does not study math and stuff she will have a square brain (cerveau carré)

@whitequark @prophet @thephd @eniko @rotopenguin very slightly ;) since you can just expand the set of ordering terms to include operator and operands, unspecified order gains you only this dubious optimization at the cost of introducing ambiguous semantics for a class of programs.

racket went for specifying left-right eval order and i think it was the right thing

the way to remember e-cores vs p-cores is that the e cores are very Efficient with regard to time, calculating very quickly, whereas P cores are Parsimonious with Power, consuming very few electrons, of which there is a global shortage as we all know

"Like AWK, Perl, or any shell, Guile can interpret script files." wtf https://www.gnu.org/software/guile/manual/html_node/Guile-Scripting.html

guile's manual is funny in that much of what it has to say responds to concerns that people had in 2010

i guess people are already doing this but i suppose best practice is to submit your own papers to the AI gods and ask what is wrong with them, because that is what reviewer 2 will do anyway; "missing the point as a service"

"confident but wrong" is a very reviewer 2 vibe, so we should not be surprised in the rise of llm use in reviewing paper submissions

catching a poop from my youngest with my bare hand was funny last year, this year's poopsplosion was somewhat disconcerting, but hopefully it will be the last father's day in which i touch poop

it is friday the 13th, post horror

unicode is fun sometimes

interesting writeup on jujutsu from the POV of a magit user https://blog.alarsyo.net/posts/2025/02/on-jujutsu-and-magit/

why is lean implemented in c++

@whitequark well. i do not currently use them. but the agentic / mcp stuff seems to be getting more capable by the week for code things