For a different method of learning Arabic (or several other languages), check out https://www.languagetransfer.org. It teaches Egyptian Arabic, explaining that the entire region knows this dialect because Egypt is the TV and movie hub of the Arabic world.
It's all audio (MP3s or streaming) and completely free, and (IMHO) the best language learning system out there. I have used it to learn Spanish, my sixth language and can't recommend it highly enough.
I have mentioned this in a few comments: for my CS classes I have gone from a historical 60-80% projects / 40-20% quizzes grade split, to a 50/50 split, and have moved my quizzes from being online to being in-person, pen-on-paper with one sheet of hand-written notes
Rather than banning AI, I'm showing students how to use it effectively as a personalized TA. I'm giving them this AGENTS.md file:
And showing them how to use AI to summarize the slides into a quiz review sheet, generate example questions with answer walk throughs, etc.
Of course I can't ensure they aren't just having AI do the projects, but I tell them that if they do that they are cheating themselves: the projects are designed to draw them into the art of programming and give them decent, real-world coding experience that they will need, even if they end up working at a higher level in the future.
AI can be a very effective tool for education if used properly. I have used it to create a ton of extremely useful visualizations (e.g. how twos complement works) that I wouldn't have otherwise. But it is obviously extremely dangerous as well.
"It is impossible to design a system so perfect that no one needs to be good."
For anyone interested in this (and certainly for OP) I highly highly recommend the book Working in Public: The Making and Maintenance of Open Source Software by Nadia Eghbal. When I was raising my profile on my open source farming robot, this book really helped me understand the types of projects one might want to foster, how to think about users, and generally gave me very helpful guidance on becoming an open source maintainer!
Neat, but I think it's deceptive for the website to claim this is a "new type of graphing" [1]. The fuzzy graph of F(x, y) = 0 is simply a 3D plot of z = |F(x, y)|, where z is displayed using color. In other words, F(x, y) is a constraint and z shows us how strongly the constraint is violated. Then the graph given by F(x, y) = 0 is a slice of the 3D graph. If you're claiming that you've discovered visualizing 3D graphs using color, you're about 50 years too late.
People are paying off these devices and then once they have paid them off, they break and people in these areas don’t have the skills or resources to fix them.
This has led to over 250 million of the units lying around broken in peoples homes, leading to solar being one of the fastest growing e-waste streams in the world.
It’s hardly solar punk to sell people cheap crap at a 10x mark up that pretty much immediately breaks once the warranty period is over.
> Hoberman makes clear one crucial factor in the city’s creative energy: “cheap rents.”
I keep seeing this in various places. The rise of the "College Music" scene in Athens, Georgia during the 80's has also been in part attributed to the cheap rent in the student ghettos (typical of many college towns).
Growing up in Kansas City, the neighborhoods around the Kansas City Art Institute were also low-rent. Child (impressionable) me remembers walking through the neighborhood at night, let by my mom, for the free Friday night film ("Journey to the Far Side of the Sun", "Fantastic Planet" to name a few I recall). There was a large chicken leg sculpture, perhaps 8' tall in one yard that always spooked me to walk past. Some kind of sculpture of broken bits of mirror and glass made another small lot look like an alien set from "Star Trek"....
I can also highly recommend Enzio Mari's Autoprogettazione furniture. Although slightly more involved in construction, all you need is standard planks, hand saw, a hammer and some nails. The instruction PDF can be found online (chairs in the latter half):
For Japanese specifically I made an iOS/macOS native app for learning by reading. It curates a library of RSS feeds and some books rather than generating them with AI because there are still issues with the naturalness of LLM-made Japanese writing. I've found many learners are still apprehensive of using generated content.
To calibrate the content to your reading level, rather than generating the content, it tracks your comprehension and shows you how much of a given webpage or book you already understand.
It has optional Anki integration if you don't want to use the built-in ones. I work on this full-time now and am about to launch a manga reading mode, plus Netflix caption lookups.
For those interested, see Daniel Shiffman's Nature of Code[1], a book in which you go from simple "ant" simulations to "boid" flocking behaviors, and from physics simulation to machine learning, neuroevolution, and NEAT using p5.js for graphics.
Not my project, but another approach recently published used Depth Anywhere to create a virtual depthmap for a given 360º equirectangular image and then apply to point cloud and render using three.js / A-Frame.
Appears to be similar capability as OP for creating scene depth from 2D, but using point cloud instead of gaussian splatting for rendering so looks more pixelated:
https://github.com/akbartus/360-Depth-in-WebXR
Also unlike the World Lab example you have the ability to go further outside the bounds of the point cloud to inspect the deficiencies of the approach. It's getting there but still needs work.
I quit Google Search (for DDG), Chrome (for Iridium on OSX and Safari on iOS), Android (for iPhone), Dropbox (for Internxt). I am still using Gmail but only for dumb emails. For the important ones, I moved to Tuta.
I also quit Facebook, but firstly because it’s a shithole, it’s the place where you ‘reconnect’ with ‘friends’ from Junior High… Good Lord!
I don’t have accounts on Instagram, Snap and, God forbid, TikTok. I only access Twitter via XDeck, and for no more than 5 minutes a day.
As a local, pretty impressed with the listed items in Turkey!
Still, the rabbithole goes much deeper. I highly recommend The Turkish Cookbook by Musa Dağdeviren, a food anthropologist of sorts who spent his formative years traveling the country town by town and collecting recipes that were often condemned to be forgotten due to the modernization and globalization of food and food culture: often they'd be passed down generationally and not written down, rely on manual and work intensive or just plain unusual methods, have ingredients that are easy to acquire in a rural household but hard to acquire in a grocery store, have flavors that would be considered unusual by the global standards of the 70s-90s, etc. It's a truly massive and amazing collection.
A little surprised that neither the paper nor the discussion here cite Francois Chollet's 2019 paper on defining intelligence for machines [1], which talks explicitly about visual reasoning and proposed an RPM-like visual task as a benchmark. I found it very useful for thinking about what it means for an AI to be generally intelligent.
From what I remember (having not read it in a while), his idea is basically that general intelligence is the efficiency with which you can learn, after appropriately "penalizing" any baked-in priors or knowledge beyond what a typical animal is born with.
As an example, Magnus Carlsen's chess skill is a testament to his general intelligence because he became skilled at chess in one human lifetime, starting out with only basic priors about the world like objectness, agentness, and some others I can't remember. In contrast, DeepBlue is not as intelligent, because it started with many more priors specific to chess, and AlphaZero is also not as intelligent, because it acquired its chess skill much less efficiently, requiring the equivalent of thousands of human lifetimes.
I don't remember if the paper considers whether 3.5 billion years of "pre-training" counts in this regard, though.
Matrix theory by Franklin is a great, affordable book containing many interesting results such as this — can highly recommend for those interested in linear algebra.
I was gonna say something along that this would be cool to use to imagine streets built differently. Like with less traffic, cycle lanes, street vendors etc., and then they link to a Dutch website already doing something similar. Very cool.
Things like these could be useful in helping to push decision makers and the public to see new opportunities. Right now when something is being built, it's always a optimistic 3d render on a sunny day with people laughing being shown to sway the public in favor of the project. Letting us "normal people" fight back against certain projects or suggest our own without needing to have professional architects draw a concept could be nice.
One fun thing to do with a swarm of bug near a light is to take out your keys and shake them.
Keys will often ring in the ultrasonic and poorly mimic the frequencies that a lot of bats use.
If you have a population of bats that hunt near you, the bugs will typically just drop. Like, just drop to the ground as fast as they can.
They know the sound of hunting bats and your keys may be just close enough to that sound that they think they're being hunted. So they do the best that they can to get out of the way and go with gravity. At least, that's my theory.
Fun little thing to do as a bet or with the kiddos.
This is really interesting research. It’s fascinating to hear about how these experiments are designed in the first place.
If anyone is interested in this topic, I highly, highly recommend the book “Ways of Being” by James Bridle. [0]
It covers such a diverse range of ideas about intelligence, including neural network–style intelligences, that I think pretty much everyone would find something of interest.
One of the takeaways is that these intelligence-seeking experiments have been poorly designed for so long, that they tend to reveal way more about the humans running the experiment than the animals being researched. And part of that is due to how uncomfortable we are with admitting that animals have completely different (not better or worse) ways of being and thinking than us.
It makes you aware of the limitations of our hierarchy-and-comparison-based concepts. For example, even a simple phrase from this article:
> Although spiders can’t literally count one-two-three, the research suggests some jumping spiders have a sense of numbers roughly equivalent to that of 1-year-old humans.
The experiment doesn’t really prove that spiders can’t count, and surely if they did count it wouldn’t be by speaking out loud like we do. Saying that their faculties are equivalent to 1-year olds only really serves to diminish them. Not that I’m blaming the author, it’s just so easy to accidentally slip into these very limited ways of thinking.
Immich (self hosted Google photos alternative) has been using CLIP models for smart search for a while and anecdotally seems to work really well - it indexes fast and results are of similar quality to the giant SaaS providers.
Shameless plug - we recently released our $80 headset with support for spatial videos. It's a plastic headset, with two bluetooth controllers, that you put your phone into 'Google Cardboard style'. No need to wait for Vision Pro :-)
https://www.zappar.com/zapbox/
Not exactly in a DAW, but I found Syntorial to be incredibly helpful. Teaches you a lot of jargon around sound manipulation and how to reproduce a desired sound. It'll make you way more comfortable navigating a DAW later.
It's all audio (MP3s or streaming) and completely free, and (IMHO) the best language learning system out there. I have used it to learn Spanish, my sixth language and can't recommend it highly enough.