"The 160 MFLOPS Cray-1 was succeeded in 1982 by the 800 MFLOPS Cray X-MP, the first Cray multi-processing computer. In 1985, the very advanced Cray-2, capable of 1.9 GFLOPS peak performance
...
By comparison, the processor in a typical 2013 smart device, such as a Google Nexus 10 or HTC One, performs at roughly 1 GFLOPS,[6] while the A13 processor in a 2019 iPhone 11 performs at 154.9 GFLOPS,[7] a mark supercomputers succeeding the Cray-1 would not reach until 1994."
These flops are not the same. The 2013 phone flops are fp32, the A13 flops look to be fp32 as well (not entirely sure), while the Cray numbers (like the rest of the HPC industry) are fp64 (Cray 1 predates what would become IEEE 754 binary64 though, so not same exact arithmetic but similar in dynamic range and precision).
A modern Nvidia GB200 only does about 40 tflop/s in fp64 for instance. You can emulate higher precision/dynamic range arithmetic with multiple passes and manipulations of lower precision/dynamic range arithmetic but without an insane number of instructions it won't meet all the IEEE 754 guarantees for instance.
Certainly if Nvidia wanted to dedicate much more chip area to fp64 they could get a lot higher, but fp64 FMA units alone would be likely >30 times larger than their fp16 cousins and probably 100s of times larger than fp4 versions.
Years ago, I installed the Facebook app on my phone. I immediately uninstalled it when I saw, horrified, that it had hoovered up all my photos and uploaded them to Facebook (there was no fine-grained storage permission at the time) "for my convenience". I never ran their app on my phone, again.
Location: Seattle, WA
Remote: Yes
Willing to relocate: Maybe to the SF Bay Area
Technologies: Java, Spring, TypeScript, React, Node, SQL, Rust, Python, Clojure, Solid-JS, Nix
Résumé/CV: https://drive.google.com/file/d/1NzYmjFItbxdlewYL_Lz--ZnE24_CqXvH/view?usp=drive_link
Email: mail@requirenathan.com
I'm a mathematician-turned-programmer looking for my next software engineering role after being laid off. I have professional experience as a full-stack developer working on a Java/Spring backend with a React/Next.js frontend. My background has given me lots of practice working away at difficult abstract problems. I'm an information sponge and a ferocious, self-directed learner. My portfolio can be found at requirenathan.com.
I briefly worked on a product related to this. It was a chatbot meant to replace the human phonecall in just this situation. The user would get a text from the bank with a link to the chatbot. They ended up not being able to sell; the common complaint from the banks was that they'd been training their users to never click links like that.
I've just started with jujutsu, as well. Jjui fills a little bit of the gap. Among other things, it allows for quick selecting and splitting of changes. But it's no Magit. I'm thinking of having a go at making an emacs interface for jj myself.
I haven't tried Cursor, but I've been messing around with Aider (aider.chat) with interesting results. With it, you add files to its context and then describe the edits you want. It then writes a commit with (hopefully) those changes. You can use it with pretty much any model, though only close-to-SOTA models work well. I've had pretty good results with Deepseek Coder, which is tens of times cheaper than Claude.
Location: Seattle, WA
Remote: Yes
Willing to relocate: Maybe to the SF Bay Area
Technologies: TypeScript, React, Node, SQL, Rust, Python, Clojure, Solid-JS, Nix
Résumé/CV: https://drive.google.com/file/d/1fxsejocFHalsDcW4VXSeBSSbMdstCs6P/view?usp=sharing
Email: mail@requirenathan.com
I'm a mathematician-turned-programmer looking for my first professional software engineering role. My background has given me lots of practice working away at difficult abstract problems. I'm an information sponge and a ferocious, self-directed learner. I'm returning to the workforce after time away dealing with (now under control) health issues. My portfolio can be found at requirenathan.com.
I'm looking for my first job in software development. My training is as a mathematician, so I have lots of practice working on tough, abstract problems. I'm a ferocious self-directed learner and have worked on a broad range of projects. Most of my recent work has been focused on frontend, with some smaller React/HTML/CSS projects, and a bigger full-stack one using SolidJS. I'm for you if you're willing to take on someone at the beginning of their career who's driven, knowledgable, and quick on the uptake.
A solution that hits the happy middle for me is to ban, or at least restrict, advertising of sports betting. The way they've devoured sports coverage is what I find the most annoying.
The problem with this is that it runs into free speech issues. The right to say most things, including advertising legal things, is more culturally significant than the right to make wagers in many societies.
Though oddly, this is a place where we can take some inspiration from tobacco. Restrictions on advertising work, but restrictions on sales mostly don't. The thing that works may be preferable even if it is culturally more difficult to accept.
> The problem with this is that it runs into free speech issues.
No it absolutely does not. Rules and regulations on advertising including what can be advertised, how it's advertised, who it can be targeted to, and the medium in which it can be advertised in are already strictly regulated in the US, Canada, and Europe.
Yes, it absolutely does. That competing interests sometimes prevail does not eliminate this particular interest as an issue. Also worth noting that I was appealing to cultural significance rather than legal significance for a reason.
"Commerical speech" in the US and is subject to much stricter limitations than things like political speech. The government is allowed to restrict advertising to fulfill it's governance aims, provided those aims are narrowly defined and specific. Incidentally, the specific case of barring Casinos from advertising has legal precedent. So no, gambling ads cannot be argued to be under the protection of the first amendment.
Edit: sorry, just saw your note about cultural significance. Not legal, so perhaps the legal argument will ring hollow.
But to address it from a cultural perspective, note that there is a long history of restricting advertising speech for public good. Doing so is not out of line culturally with what we do across thousands of different industries. It would be an extension of the status quo. Not an upheaval in our relationship to free speech
I'm looking for my first job in software development. My training is as a mathematician, so I have lots of practice working on tough, abstract problems. I'm a ferocious self-directed learner and have worked on a broad range of projects. Most of my recent work has been focused on frontend, with some smaller React/HTML/CSS projects, and a bigger full-stack one using SolidJS. I'm for you if you're willing to take on someone at the beginning of their career who's driven, knowledgable, and quick on the uptake.
"The 160 MFLOPS Cray-1 was succeeded in 1982 by the 800 MFLOPS Cray X-MP, the first Cray multi-processing computer. In 1985, the very advanced Cray-2, capable of 1.9 GFLOPS peak performance
...
By comparison, the processor in a typical 2013 smart device, such as a Google Nexus 10 or HTC One, performs at roughly 1 GFLOPS,[6] while the A13 processor in a 2019 iPhone 11 performs at 154.9 GFLOPS,[7] a mark supercomputers succeeding the Cray-1 would not reach until 1994."
reply