Have LLMs produced something novel like Cantor's diagonalization procedure once was? Come up with new concepts like set theory once was and axiomatized it? Is there any indication or evidence LLMs can do this? If not, I would wager mathematicians are still going to be a profession in the future. Maybe not as theorem provers in a particular axiomatized system where LLMs can be effective tools but the wider project of synthesis and unification, and the discovery of new modes of mathematical thought will still require mathematicians.
I read through most of the article but didn't see motion sensors mentioned. A lot of suburban streetlights are lighting up empty streets at night. With a combination of motion sensing and LEDs, lighting could follow pedestrians. Cars really don't need street lights in most cases but even then, in higher risk areas, motion sensors could turn on streetlights just as long as needed.
35 years ago I had to program a Z80 by "assembling" programs by hand and punching hexcodes into a board. This was made easier by writing an assembler for it. This was my way into tools which took me eventually to working on a major C++ compiler.
Not really as large capacity capacitors (this one seems to be 4V) tend to be low voltage as well. You can touch the terminals without harm but attempting to short them is going to be, er, entertaining. I have accidentally touched a 200V capacitor and it was not fun.
I call all these new fonts monofonts, mono in the sense of monoculture. Aesthetics practically indistinguishable from each other. Give me one of the IBM Selectric fonts in a modern form and I'll be happy as a clam.
They can ban immigration from the third world all they want but at some point, they'll have to confront the economic reality of a 1.42/couple fertility rate.
You missed the point - understanding comes from working through the code, not just reading it. This is nothing new: nobody has learned or done new mathematics or physics or whatever by just reading a textbook.
During the last few months when AI usage was mandated in our team and usage exploded, our team's throughput has barely changed. Now, if this was due to people working 2 hours a day and painting, cooking and playing golf the rest of the day, this would be a great result, but I see many people work past 6pm, and yet the output is mostly the same. We are not tackling harder problems or fixing more bugs despite authoring numerous skills for AI. Eventually the reckoning is sure to come, and I think it will not be pretty.
As I am continually amazed at how well Claude 4.7 deals with highly complicated C++ code, I am also becoming painfully aware of the developing situation mentioned in this article: I no longer completely understand the code it is editing, not because I'm incapable of doing it, but because I have not authored the changes. I am trading throughput for understanding, and, eventually, judgment.