Nathan built a few small projects. Have fun with them!
EEG analysis presented robust evidence that LLM, Search Engine and Brain-only groups had significantly different neural connectivity patterns, reflecting divergent cognitive strategies. Brain connectivity systematically scaled down with the amount of external support: the Brain‑only group exhibited the strongest, widest‑ranging networks, Search Engine group showed intermediate engagement, and LLM assistance elicited the weakest overall coupling.
I entered the workforce about 9 months ago and my optimism for building new tools and services that help the small starting up business I work for has all vanished. I work in an environment that changes every 2 months or so, as my boss finds a new business venture she wants to enter. This has me starting and stopping quite a few projects that could have been solved in an afternoon with a quick Google Sheet.
Principles behind the Agile Manifesto
Simplicity--the art of maximizing the amount of work not done--is essential.
Functors and Monads For People Who Have Read Too Many “Tutorials”
If you find yourself confused, I recommend literally saying “Functable” to yourself. Eventually you will naturally start to hear Functor as just a synonym for how you say it to yourself. You may also consider “Functionable”, which is perilously close to “functional”, but carries the idea of “we can apply a function to it”, albeit a very particular function with a certain signature.
One Token to rule them all - obtaining Global Admin in every Entra ID tenant via Actor tokens
Somehow the API seemed to accept my token even with the mismatching tenant. I quickly looked up the netId of a user that did exist in the target tenant, crafted a token and the Azure AD Graph happily returned the data I requested. I tested this in a few more test tenants I had access to, to make sure I was not crazy, but I could indeed access data in other tenants, as long as I knew their tenant ID (which is public information) and the netId of a user in that tenant.
APT Down - The North Korea Files
It appears that KIM maintains access to internal South Korean Government network systems. There is a project named onnara_auto, which contains several interesting files.
The project appears to be tools to query internal government servers.
I now have much more appreciation for OpenAI’s insistence on doing it all, for two reasons. First, this is a company in pure growth mode, not in decline. Tradeoffs are in the long run inevitable, but why make them before you need to? It would have been a mistake for Microsoft to restrict Windows to only the enterprise in the 1980s, even if the company had to low-key retreat from the consumer market over the last fifteen years; there was a lot of money to make before that retreat needed to happen! OpenAI, meanwhile, is the hottest brand in AI, so why not make a play to own it all, from consumer touchpoint to API to everything in-between?
How Rockefeller and His Partners Built Standard Oil
Scale is helpful for most businesses, but refining might be one of the most extreme examples. A typical rule of thumb in chemical engineering is that capital costs increase sublinearly with capacity, usually by (capacity ratio)^0.6. A plant with double the output is only 50% more expensive to build, and operating costs tend to follow similar trends. The reason behind this is that chemical plants and refineries are agglomerations of steel vessels and pipes. Vessel and pipe volume increases faster than surface area as size increases, decreasing steel and fabrication costs per unit of volume
What is the Funniest Number? An investigation of numerical humor
According to the survey, the funniness of a number is associated with the degree of specificity: χ2zc(2, N = 424) = 15.30, p < .001 and χ2mc(2, N = 310) = 6.15, p = .046. This association is less significant in the Minimal Context data, although the significance substantially increases for both sets of data upon removing responses of 0: χ2mc,no 0(2, N = 288) = 16.39, p < .001
Qualcomm to Acquire Arduino—Accelerating Developers’ Access to its Leading Edge Computing and AI
“Arduino has built a vibrant global community of developers and creators. By combining their open-source ethos with Qualcomm Technologies’ portfolio of leading edge products and technologies, we’re helping enable millions of developers to create intelligent solutions faster and more efficiently—including a path towards global commercialization by leveraging the scale of our ecosystem.”
How I, a non-developer, read the tutorial you, a developer, wrote for me, a beginner
So here’s how to set it up:
In the terminal, ajkl;gawgor;iqeg;iJLkqen. wl;R aw;oeiga 4648664 arjarwgj;llj;ja fadgfgajkljl; wlj;sdjk;lfas
Next go to folder/hidden/deep/in/the/file/system/surprise!.file and copy the contents of the file. If it’s not there, it might be in library/library/library/llibrary/liiiiiibrarrrary/llllliiiiibrary/hidden/hidden/hiding/you can’t find me/hidden/nope/never/hahahahereiam.file.
It’s clear that if the box is checked, the contact list will be sorted by last name. How will it be sorted if it’s not checked?
grug say: “if design need manual, design too complex.”
Circular deals among AI companies
<infographic>
In CI for Wagtail projects, uv now sees more usage than any other installer. It’s at 66% of CI downloads of Wagtail, compared to 34% for pip. Other installers don’t report their CI usage specifically, but their overall usage is very distant.
One process programming notes (with Go and SQLite)
I start with: Do you really need N computers?
Some problems really do. For example, you cannot build a low-latency index of the public internet with only 4TB of RAM. You need a lot more. These problems are great fun, and we like to talk a lot about them, but they are a relatively small amount of all the code written. So far all the projects I have been developing post-Google fit on 1 computer.
