about every two years I ditch my init.el and build back up from scratch, and every time it feels great. I'm finally breaking off from org-roam which, for all the good its done me, has become a thorn in my side due to the centralized store. about to give #hyperbole a try, which I've been interested in for a while now.
I remember hearing that #Racket made some progress towards migrating to upstream #Chez, does anybody know what the status of that is? Racket users and users of software written in Racket would benefit immensely from this.
Chez is a powerful infrastructure in a ~500kb statically linked binary that approaches C in performance. Due to its low-level and bare bones compilation and build tools, you can also compile in only what you need. racket-minimal on #guix is ~160mb, which is great for what you get, but still 320x the size of Chez. Having Racket's ecosystem while being able to ship only chez + the compiled chez code of the libraries you import could allow for shipping sophisticate programs in only a few mb. Just saying.
Malt: A Deep Learning Framework for Racket by Dan Friedman and Anurag Mendhekar
We discuss the design of a #DeepLearning toolkit, Malt, that has been built for Racket. Originally designed to support the pedagogy of The Little Learner—A Straight Line to Deep Learning, it is used to build deep neural networks with a minimum of fuss using tools like higher-order automatic differentiation and rank polymorphism. The natural, functional style of AI programming that Malt enables can be extended to much larger, practical applications. We present a roadmap for how we hope to achieve this so that it can become a stepping stone to allow #Lisp / #Scheme / #Racket to reclaim the crown of being the language for Artificial Intelligence (perhaps!).
I'm happy to say that ever since I started using Flatpak's desktop #Penpot package, its become one of my favorite programs and I use it every day day now. And it just got a slick new upgrade. So now I finally have a #clojure application in my daily repertoire.
Way better than #Figma, which has become total garbage ever since #Adobe bought it.
@anderseknert oh it didn't? well, regardless, during the time it was supposed to happen Figma made half of the features subscription only, limited a few project per team, and generally for all the tiny faults of Penpot, I feel like Figma has just as much jank at this point. The first time I ever used it I spent like 8 hours working on a presentation without looking up or even needing to look for documentation. The last times I've used it I felt like I couldn't even get things to snap into place where I want them -- and Figma's advanced guide snapping is part of what made it such a productive and pleasant tool in the first place; no need for setting up elaborate rules to keep work through tons of similar frames, like you need to do in Photoshop.
I've had some issues using their website ever since the (disbanded) Adobe partnership was announced, and they posted that the massive traffic influx was more than they were prepared for, but since just saying f*ck it and using flatpak instead of worry about packaging etc, its amazing. I'm not a UX designing but for doing bang up prototypes for presentations, its the jam and allows you to quickly illustrate how a proposed application works for a client in such a way that there is little ambiguity about what your delivering, which eliminates a lot of stress.
@grtcdr
So what do you think is wrong about PenPot's packaging processs? Do you think that software producers should be responsible for packaging their software?
@grtcdr I already said its a web app, its not a binary meant to run natively, but can be deployed to.
as someone who has packaged tons of software, expecting independent software producers to package their software for every use is silly. a "target ecosystem" -- what is that?? If you're writing free software, your initial users in particular are going to be spread across distros and systems. Hence why distros and package managers are typically responsible for packaging. What we need is better funding for package managers and distros.
Penpot is designed for commercial deployment, they've put plenty of work into designing a webapp that can leverage several servers across the globe to facilitate low latency "multiplayer" design workflows across several studios simultaneously.
If I wrote a guix package for penpot, it would take me several days, assuming no surprises. I would rather just use their software tbh.
A #twitter-like platform won't be how the #fediverse takes off. Fedi won't explode with popularity until there is a federated #expage-like platform, complete with #federated guestbooks
A little over a year ago, originally due to an interest in the deeper history of #compilers, I started diving deep into the #Talmud, studying #Aramaic, #gematria, and doing #DafYomi etc in what has become my deepest engagement with the rabbinic corpus yet -- the Talmud isn't a compiler but rather an extensible interpreter, compiled by compilers over the course of many centuries (build times have gotten significantly faster, my G-d), with novel extensions in the form of rabbinic commentary, glossia and the like being added nearly every century by publishers competing to compile the most elegant editions (Vilna Shaws being paradigmatic). And through studying Talmud and the greater body of rabbinic literature I've found myself encountering #magic/sorcery occasionally, and I just gotta say -- the #SICP metaphor of programming as pure magic, with the #hacker as a sorcerer, goes insanely deep when you start to dig into it.
A good deal of what constituted sorcery in ancient and medieval times consisted of elaborating early encryption methods. Even algorists, early computers who would sit at market stalls calculating costs, were portrayed as ascetics verging on the edge of the occult in most accounts and depictions of them. Gematria -- practices of elaborating symbolism through the association of values with numbers -- is fundamental to magic (as well as anti-magic mystical philosophical schools like rabbinic Judaism and most schools of Islam); it was one of the elementary tools of the trade.
#Programming is in many ways the representations of knowledge through the association of values with numbers, seeking to do so in as few steps as necessary, which as we know involves reducing some rich system of concepts to a more elementary representation, that is then translated into machine behaviour, replacing concepts with instructions, which may be thought of as a form of slight-of-hand.
For those with the same extremely niche set of interests (half my mutuals): gematria is like machine code, the Mishnah is like an assembler, the Gemara is like L1, the Rashi is like L2, and so on, with #kabbalah being for the talmud what graph theory is to register allocation. And Sefer Toyrah HaKoydesh is the reality their seeking to represent, an thus elaborating in the process.
The same could be elaborated together for Islam and the much larger variety of compilations of hadith literature, as well as most schools of Daoism -- all of which are ever-developing philosophical-spiritual traditions of compiled literature and subsequent non-apophantic doctrines of mystical jurisprudence. Vodun should also qualify although much of their ancient compilations of text were ravaged by colonialism and have yet to be re-assembled due to much of them being held in European archives far-and-away from the 40 million+ vodun practitioners around the world. I'm sure the same can be said for countless other elaborations of "the Way" that have been suppressed and destroyed.
@SmallOther fwiw, I think the easiest way to grasp the difference between the transcendental and transcendence is that a transcendental is an "object" that indexes a set of values in a hierarchy of concept-formation (similar to a subobject classifier in a topos), while transcendence is the movement from one level of a transcendental evaluation to another.
In a previous post I had mentioned that all of my students in #Gaza appear to have been killed since the massacre in #Rafah. Thank G-d, my student Mohammad has since appeared, having lost his remaining means to connect to the internet, but he and his family remain alive — struggling to survive, fleeing intermittent drone attacks on their tent city, and bombing campaigns by the occupation forces, but still thankful to have survived thus far. My friend Sith and I have set up a #GoFundMe for him, please consider giving whatever you can; he has lost everything, and is trying to care for a family of 6 from their tent in South Gaza.
Please boost to help spread the word and help Mohammad rebuild his life from the ruins of Gaza.
Wow, thanks to all of you Mohammad has raised nearly $900 in just the first day of fundraising! I just texted with him and he asked me to thank everyone for their #mutualaid support for his family. Individual family relief in #Palestine is every bit as important as donating to #UNRWA (which you absolutely should as well), as #Gaza has particularly collectivist culture and the more relief any one family has, the more they are able to help each other. Especially projects like this to rebuild their destroyed homes.
Thank you everyone so much for supporting Mohammad's family, I'm hopeful that by the end of the year we can reach his goal.
Hey #scheme, I made a new image that gets together many schemes (and schemers) in the same image, with a CLI that makes it easier, if not easy to build portable libraries, if not programs.
#;> united.scm available
loko
chicken
mit
sagittarius
gauche
racket
gambit
guile
chez-cisco
stklos
chibi
· Frontend for several Scheme implementations · One CLI, many schemes ·
· The best ui/ux to build portable Scheme libraries ·
· All around finest competition in software engineering ·
@dziban i use lispy and its pretty good, but if I arrive at someone's editor using paredit or another, I can note the different conventions and getting working in it pretty quickly too.