vanderZwan

While this is an extremely quick verification, the construction presented in this fashion appears like a massive miracle. The polynomial {F} has degree seven, so a priori the Jacobian {\mathrm{det} DF} ought to be a polynomial in three variables of degree as large as {3 \times 6 = 18}, so the fact that all non-constant coefficients of this polynomial vanish looks like a massive cancellation involving {\binom{18+3}{3}-1 = 1329} coefficients, which is much larger than the {\binom{7+3}{3} = 120} degrees of freedom for a generic degree seven polynomial of three variables. So finding such a polynomial looks highly unlikely to be located by brute force.

Sounds like the most interesting part would be learning what approaches the LLM did use to see if that's reusable elsewhere. I'm guessing that's what the rest of the article is about? Because I also couldn't follow the maths any more.

So I remember having physics homework a quarter century ago at university where we had to use Excel to determine a static electric potential field, by defining all cells in the grid as "the average value of the four surrounding cells", except the edge cells and source/sink cells which would get actual concrete values.

I distinctly remember being amazed that it would just iterate until it reached equilibrium (maybe we had to change some setting somewhere first though, I never really used Excel before or after that). I think you could change a value mid-iteration, and the grid would just continue on with the previously calculated cell values, instead of start over from scratch. That's basically "weakly connected", isn't it?

(anyway, the real challenge is to find a practical use for Excel's hidden fractal powers https://www.youtube.com/watch?v=b-Fa6HtvGtQ)

On 7 May 2025, in Rome, the Italian Chamber of Deputies approved a new law to help people with obesity. This law is the first of its kind in the world. It shows that obesity is a medical condition that needs proper care and support, and is not just a result of poor choices.

So I totally agree that obesity is better classified as having serious medical and societal consequenes, and something that needs proper care and support rather than shaming, and hope that this law represents a step towards that. The only problem I have with "medical condition" as a blanket description is that this appears to let the food industry off the hook way too easily, really.

Having said that I find the choice of "show" confusing: a law doesn't prove anything beyond some ruling class reached some form of consensus on any given topic at some point in time, does it?

Wait, is that the same guy who co-authorer the Floyd-Rivest Algorithm? Hmm, looks like it[0]. Huh, neat.

[0] https://en.wikipedia.org/wiki/Floyd%E2%80%93Rivest_algorithm

Floyd-Steinberg and Atkinson are the "play Freebird!" of dithering algorithms.

Airport controllers (and similar positions for trains and metros) probably also suffer from this then

The first Chrome and Firefox implementations did

I was about to say: I'm sure I've seen it work a t some point? I imagine it's a valuable thing to add for the web though. It would be really cool if you could use the same image source for thumbnail and full image, and the browser both just figures out how much to download based on pixel size and can resume previously partially downloaded images.

And yeah, the tiling isn't implemented anywhere yet, jxl doesn't really get enough funding for that. But it'll be really cool once it does since it also makes it really useful for giant images of geographic data. I don't know if it combines with streaming downloads as well, but it would be crazy cool if we effectively got OpenSeaDragon[0] support inside an image format

[0] https://openseadragon.github.io/

I wonder if you can do this in JPEG-XL. I know that that has actual animation support, but this would be a different thing.

Uh, there definitely is "hardware specific to processing language" for our species, they even have their own wiki pages[0][1].

And in general our brains differ significantly from that of a chimpansee. Human brains push neuron density to to absolute limit within the animal kingdom and have unique evolutionary adaptations that allow for it. As a result our species also has mental health disorders unique to us that seem to be cause by the neurons being pushed past that limit.

[0] https://en.wikipedia.org/wiki/Wernicke%27s_area

[1] https://en.wikipedia.org/wiki/Broca%27s_area

They technically didn't say that they don't use programming languages, they said they're not a programmer. There's plenty of other people who use programming languages regularly, but not necessarily because they're into them.

Come on guys, the contribution of the research is that it made measurements of how the processing of multiple inputs is represented in EEGs, not whether or not we can handle multiple inputs. Stop acting snarky, it just shows you didn't read beyond the headline.

Your example is needlessly bleak, but in general: yes, we're a social species and being able to process multiple speech streams seems obviously pretty important in many social contexts involving more than two people.

It's fascinating how similar-yet-subtly-different the cursive writing is across Europe. I wonder if you could map them into something similar to language families (or type traditions, I suppose). Would there be any "language gradient" equivalents for cursive writing across the continent? Well, before the industrial revolution I guess, I'd expect that after that you'd see more singular influences of designers backed by the state pushing for standardization of education.

After searching a bit more I discovered the situation is similar in the Netherlands too: schools can choose between the two styles (or more like three publishers each with their own styles, really).

Anthropic is not in the programming language market

Arguing that a product that sells itself as improving programmer productivity by writing the code for you has no stakes in "the programming language market" because it doesn't sell a programming language of its own is impressively shortsighted. Especially when the leader of a programming language has openly stated their dislike of vibecoding, critisized the industry, and the language project itself rejects PRs made with the product being sold.

Thanks for the in-depth reply, appreciate the elaboration on where you did and didn't use LLMs. Looking forward to reading the blog posts and other future updates!

Hey, that's the same one I was taught in the Netherlands in the late 80s! It seems to have been replaced with an English-style in recent decades though, is that the case in Germany as well?

Huh, I just indirectly learned from this article that the way I write a lower-case "t" in cursive is a Dutch way of doing so (edit: sollniss' comment implies it was a common style in Germany too). A quick search suggests it has been replaced with an English style of "t" in the last decades too.

I wonder if that makes my handwriting harder to read for anyone who isn't Dutch and over 40 years old.

Anyway, just bringing it up because you don't need to lift up your pen to write that kind of "t".

Search for "koordschrift" on https://primarium.info/countries/the-netherlands/ to find the illustration showing how I was taught to write it in the late 80s. It's the letter vaguely shaped like a pine tree.

So just my 2 cents: I also have an immediate "ick" reaction when I see the vibe-coded aesthetic of the website itself, expecting the worst, as well as serious ethical concerns about using LLMs at this point in history. Luckily in this case, it also was (somehow) immediately obvious to me that actual thought was put into the language, so I was willing to take a closer look to see if some genuine effort was put into it.

But be aware that a lot of people won't be willing to do that. Based on my experience in proglang chat servers and fora I'd say that most of the people with the niche interests most relevant to you (deep dives into non-standard ideas for proglang design & implementation) are in that subgroup. Especially the ones with inspiring ideas and knowledge to share.

Beyond expecting honesty about the usage of LLMs I'm not going to say anything about whether you should use them to develop your langoage or not, that's up to you. But I'd strongly advice for avoiding it in the presentation side of the language (website, whatever) at all costs, because it will get in the way of connecting with your potential audience more than it will help.

To give a point of comparison: you are undoubtedly aware of the issue that popular open source projects have with slop PRs? One problem is that it is so much harder to to see where the LLM output ends and where human effort starts. And the last part is where the actual conversation takes place! Having to filter for that is just exhausting and starts to feel disrespectful.

When I'm reading your language I want to read the part of it that is your language, not the LLM generated bits of it. The LLM aestetic and writing style of the website is an extra layer between me and your thoughts, and it's distracting the same way a chain smoker's body odor is distracting to non-smokers despite being oblivious to it themselves. If this is a language you're passionate about, and it looks like it, then I'd say it deserves better than that.

So I want to preface my next point by stating actually really like how syntax reads, hitting a sweet spot of leaning just enough into APL's symbol-focused terseness without losing the familiar structure of literally every other popular programming language ever that makes it easy to follow the code. I imagine it could be a great language to pick up for working out an algorithm with pen & paper (the best limus test for how well a language functions as pseudocode imo).

Having said that, I have my doubts about the reduced token claims because of a lack of "begin" and "end" or "{" and "}" delimiters. Surely, if your language is identation based, LLMs will treat significant whitespace as a token too, since they need to keep track of it? Which would result in an extra token per newline compared to using explicit begin/end token?

I suspect that to really optimize for tokens one would have to go the concatenative route, which significant whitespace very much isn't.

I mean the last bit is a pretty important skill regardless of safety features, especially in a country like the Netherlands.

those who don't want to (…) spend as much money

That would generally speaking be the average Dutch person, no?

You're making it sound like the motorcycle equivalent of the original Fiat Panda 4x4 (well, except that you didn't mention anything about repairability, but you get the vibe I'm going for)

Yeah, and keep in mind most of those houses where merchant houses where trade happened at the time so there probably was a lot of loading and unloading going on everywhere.

Although, given the current-day Dutch attitude to bike helmets I wouldn't be surprised if our ancestors were just being stubborn too, haha.

Funny, I'm Dutch and have an arts degree so obviously the name Van der Heyden is familiar to me, but I didn't know he was also an engineer who modernized our firefighting systems.

Having said that, "Dutch golden age painter invented better way to pump water out of the Amsterdam canals" sounds more like someone tried to cram as many stereotypes about the Dutch into a sentence than an accurate summary of something that actually happened, haha. I'm surprised it's real history.

I feel the same as you about programming games, but can sort of see it for truckers, because surely some truck drivers picked that job because getting in the "flow" of long-distance driving appealed to them. I presume the trucking games try to appeal to exactly that feeling without, as you say, many of the actual stressors of the job.

So I can imagine it might let them reconnect with that feeling and be a relaxing experience much more easily than a programming puzzle game would let us reconnect with what we love about programming. Being a puzzle game it inherently will involve some frustration, which is the thing I want to escape from after a day of programming.

Hey, as someone born in Ghana I just want you to know I think it's pretty cool you used a Mankala-based game as the minigame this time :)

(at least, that's what I'm prematurely concluding based on ten frames of trailer footage)

[0] https://en.wikipedia.org/wiki/Mancala?wprov=sfla1

The hive-mind reads another history article about how self destructive the various meta-hive-minds acted over the last centuries. While the critical self-reflection is justified, it thinks the non-hivemind versions weren't much better at first, stuck in a cycle of repeatedly nearly exterminating themselves with their own wasteful toxic oxygen for 800 million years, until some of them figured out how to use it for something.

[0] https://web.archive.org/web/20130124200735/https://www.pathe...

I think you're better off watching any of the various recordings "Performance Matters" conference talks[0] by Emery Berger to learn about all the ways that developers are benchmarking the wrong thing using the wrong techniques, and what you could do instead

[0] https://www.youtube.com/watch?v=7g1Acy5eGbE&t=5m3s

US drug policy is more "which population demographic uses it?" based than anything else.