Alright after last week's dud stream, who knows if anything will work this week?
https://toobnix.org/w/gXLXQqxf5MYg1NDF2Ua6oA
tbh just looking @mdhughes 's latest full version #understone #itch_io #gamedev #release and the currently only successful jam application for the experimental first issue of the #lispyGopherClimate #lisp #zine in two weeks (I will take a week to 'put it together').
Update: Now I've solved sixteen Project Euler problems in muLISP on the #hp200lx. Sometimes I'm not mathy enough to find a closed-form solution to certain puzzles involving millions of terms. For those I am limited by only having 640KB RAM and 8 MHz processor, and it's necessary to send the program to a desktop (or even my phone!) running SBCL. Even though muLISP is not a Common Lisp, the porting is straightforward. One particular program that would have taken a full 24 hours on the palmtop completed in 0.3 seconds on my desktop.
As a #retrocomputing head I am attracted to the pre-Common Lisp dialects which were in use when Lisp actually experienced its heydey. CL was a good compromise, forward compatible with most of these dialects, but imo lacks the small size and conceptual unity that make a language easy to learn. I've found it easier to learn muLISP and gradually pick up CL features where appropriate when I'm on a modern PC. And I greatly appreciate CL for existing and being as stable as it is.
As much as I admire the purity of Scheme, the fragmentation hurts it a lot. Every implementation has some flaw or another that drives me away.
Show HN: Metaspec: The DpANS3R Common Lisp Spec in S-Expr and HTML Format
initially i thought that the many dialects of #Lisp (and the many implementations of #Scheme also) made it seem really daunting, but it really isn't.. at least not too much so. like, i thought that i had to figure out and choose The One variant/dialect/impl. and stick with it for doing everything that i want to do but that's just not the way it works with Lisps
you can (mostly) choose any one of the many variants and in lieu of your learning path becoming limited to just That One variant, it'd instead just improve your overall experience with using other Lisps in the long run too
and if you're worried that the variant you're using lacks something in features, You Don't, because if that ever happens, you can just go program your own implementation of a feature you need whenever that need arises .. it is very literally a Programmable Programming Language and it's so swag when you realise the extent of that paradigm
What do you know, my swiss-like table for dispatch is marginally faster without PIC/POLY quick lookup even for monomorphic call profile.
My hypothesis is that I've optimized it well enough against memory pressure, so I've hit a new frontier - instruction cache pressure.
Needless, to say, I was shocked and benchmarked it 2x on two different computers (4x), the results check out.
Как избавиться от множества раздражающих лишних скобок? Полагаться вместо скобок на порядок следования и чётность/нечётность! https://git.sr.ht/~technomancy/fennel/tree/main/item/reference.md#codecasecode-pattern-matching, https://git.sr.ht/~technomancy/fennel/tree/main/item/reference.md#codeifcode-conditional (Ебанутый синтаксис, но мне нравится.)
There's nothing to see yet in my GravityLoops simulator in Interlisp and LOOPS, but for good reason. I worked on the foundations that will make the program actually show something on the screen.
https://journal.paoloamoroso.com/preparing-gravityloops-for-simulation
screwlisp
psf
Curated Hacker News
Digital Mark λ ☕️ 8647
notptr
asnev 
Daniel Kochmański
mintbug
Geoff Wozniak
Paolo Amoroso