It feels like this list, if it focuses on individual authors is - if successful - overwhelming in scope. There are, what, tens of thousands, hundreds of thousands of authors ? If 1%, for argument's sake, this list would have thousands of links.
Nonetheless, I'll promote an author of both non- and fiction ebooks I've bought from, Michael W Lucas [0].
I think that this is a plugin library for teams that want to offer a platform for the public (or an LLM-AI) to submit code to. If your team writes some code, you don't generally sandbox it from yourself, you just amend your program: you don't need a sandbox. But, if you want to run code that you don't trust, you should run it in a way that prevents it from causing problems if it is actually dangerous (like a virus or accidentally overwrites your files with blank files). That's what a sandbox like kyushu promises to do.
So, with a sandbox library like this, you could - say - write a website that hosts games (like itch.io or newgrounds) that hosts games on the world wide web. The sandbox part can give you confidence that, if a villain's programmer henchmen uploads a virus instead of a game, it can't infect your platform or other games on the website. Or, if a LLM-AI written game is accidentally tries to take up all the memory of the computer, it can't ask the operating system for more than is in the sandbox.
tl;dr Peter Steinberger shared a product demo for CodexBar [0] with a graph of OpenAI token usage. This graph shows one million spent, prefers gpt-5.5 and spent twenty thousand today.
However, I do not see a strong reason to believe that this is his actual, personal usage. It could be all openclaw usage or some subset of openai usage, given that he is inside them. I suspect it is far more likely to be fake data [1] that exercises the graph library in a visually satisfying way. Notice that it has no usage for a 'week' after April 15 (a Wednesday), but picks up a bunch later. As marketing copy it needn't have any basis in reality [2]. I should hope openai would put a procedure in front of their entrepreneur acquisition that prevents accidentally exposing trade secrets [3].
tl;dr Quasicrystals are aperiodic structures. The author notes that the conditions for creating them are rare, given the need for instantaneous high temperature. They recount that these can happen during space debris impact and when lightning hits sand. They close out by describing some of the chemical 'formulas' for these materials, given that characterizing a prototypical section is difficult without repeating elements.
I don't have anything to say about quasicrystals, other than it seems right up this blog's alley, as the other most recent articles are about math and materials (like feldspars [0]).
tl;dr The author [0] describes a simple ticketing system that his very busy coworker, Peter Högfeldt, created in 1986. Basically, it checked in numbered files to their version control (CVS, at the time) that could produce reports with grep. This system's simplicity lead to its longevity, as people could learn it easily and trust that its bugs had been ironed out long ago. This is in contrast to some OS software that tries to be everything to everyone and changes all the time [1].
I find this a helpful example. When I've heard the unix philosophy in the past, I didn't feel super convinced. Like, sure tar can do one thing, because it is a library (ignore that it can use gzip). But, where do you draw the line with a program like gnucash (financial tracking software) ? The core of the domain will involve keeping a ledger of transactions and converting them to relevant units. But, typing credit card charges in by hand is tedious [2], the kind of tedious that a computer should be good at. I would much rather that the program connect to my bank [3], to get the transactions directly.
tl;dr Eels have a long lifecycle with several stages. They do not develop sexual organs until late in their life, when they migrate back to the Saragossa Sea. This meant earlier autopsies of eels revealed no sexual organs, even though scientists could provoke them with hormone therapy. So, a team lead by Jose Azevedo tagged female eels in the Azores in 2018, and tracked them via satellite [0].
I sometimes think about the selection pressures that lead to complex life cycles, like fig wasps. I find myself thinking about it naively, like one existed and the other grew into the niche. But, realistically, everything is changing (slowly) all the time. I just notice it for, say, influenza because their cycle time is so short.
You could add an appendix with printed, scannable, binary data. You could create a page with a bunch of QR codes. Martin Monperrus vouched [0] for Twibright Optar [1].
tl;dr This is a project page describing a small hand woven loom. Small, means a little wider than a palm and maybe half the length of a forearm (depending on which you buy). Basically, you will run a string between two circular combs along the length of the stand and weave a separate thread or yarn horizontally many times to make a piece of fabric. These seem to cost at least 100$.
I recommend listening to DJ Earworm's mashup of the top song across the world [0] to hear a greater variety of countries, albeit for this or last year.
In fairness to this post, they probably posted as much as they could get easily and consistently. I feel like the same situation reigns for country history in book shops. When I go to a used book store in Queens, New York, I'm not apt to find any books about Jaipur, India. But, if I were to go to Mumbai, India, I could probably find something. So, maybe we should prevail upon or sponsor people in Africa to volunteer their lists.
It looks like Mark Thomas maintained a phone number database up until 2007 or 2023 for many areas in the USA. I guess that could be a basis for starting 'my own' instance of payphone-go, maybe with twilio (or equivalent) to receive the calls.
This sounds a lot like Jason Rohrer's project with an LLM, Sammy Jankis [0]. There, the posts are generally about the continual deaths when the LLM's context window fills up. The posts do not disclose an overt goal, as they do here, so the LLM has supposedly created a bunch of games and simulations.
In this case, I think that 100$ as a starting budget (while probably appropriate for the initiating person) is not enough for a substantive business. I guess the server that runs the LLM is a subsidized asset, so this situation doesn't require renting one. I guess, as a hypothetical LLM, I could try to sell customized fan fiction to fetish communities. Most of the mainstream LLM hosts reputedly have guardrails to discourage such output. On the other hand, selling in those communities largely requires trust [1], but that might take time that this scenario doesn't permit.
For targeting advertising expendatures at the site level. If most of my traffic, as revealed by referrer links, comes from social-media-platform-foo and only a little from social-media-platform-bar, then I am likely to spend more on ads from foo than from bar. I'll grant that it is a noisy measure, but doesn't need to be about tracking a particular individual.
Did you expect some answer that decried world peace as impossible ? It's just repeating what people say [0] when asked the same question. That's all that a large language model can do (other than putting it to rhyme or 'in the style of Charles Dickens').
Could y'all point at instructions for how to imitate this limited internet situation ?
I ask because, two years ago, I was able to circumvent the Windows-11-requires-internet-and-a-microsoft-account part of the set up for a new laptop computer by doing this on a flight. Apparently, connecting to the airplane wifi (without yet logging in) was enough to satisfy the OS set-up, but limited enough that my laptop didn't require a microsoft account. With windows 10 now end of life, I will probably get a new desktop computer and would like to repeat the feat at home. Thanks
Likewise. In fact, I was under the opposite impression because of the benefit that sulfur enriched shipping exhaust had for our climate [0]. It looks like these clouds are thinner and don't have the same impact as that, though. While I felt that the featured article linked to their favorite site aggressively (four links to contrails.org), it looks like the google site is legitimate [1]. I couldn't find a recent [2] paper on NoAA about contrails, but presumably others have studied it.
My understanding is largely ten years old and high level and only for one kind of fully homomorphic encryption. Things have changed and there is more than one kind.
I heard it described as a system that encrypts each bit and then evaluates the "encrypted bit" in a virtual gate-based circuit that implements the desired operations that one wants applied to the plaintext. The key to (de|en)crypt plaintext will be at least one gigabyte. Processing this exponentially larger data is why FHE based on the system I've described is so slow.
So, if you wanted to, say, add numbers, that would involve implementing a full adder [0] circuit in the FHE system.
It looks like [0] access to gemini will only be for subscribers, given that it costs them money. This is, "of course", distinct from ai mode [1] in google search that happens from the address bar in chrome. The first video implies that the difference involves throwing the current web page into the query as context so someone can ask "is this recipe gluten-free?" on a recipe webpage.
#/media/File:Full-adder_logic_diagram.svg){kind=link}
Nonetheless, I'll promote an author of both non- and fiction ebooks I've bought from, Michael W Lucas [0].
[0] https://www.tiltedwindmillpress.com/