The HN title here is currently “Performance Hints (2023)”, but this was only published externally recently (2025). (See e.g. https://x.com/JeffDean/status/2002089534188892256 announcing it.) And of course 2023 is when the document was first created, but much of the content is more recent than that. So IMO it's a bit misleading to put "(2023)" in the title.
The page is about tips for writing fast code. Much of it applied 20 years ago, and will apply 20 years from now. If by "the numbers" you mean specifically just the table ("rough costs for some basic low-level operations") in the "Estimation" section (which accounts for less than 0.5% of the words on the page), then that table was initially created in 2007, and is up-to-date as of 2025. Other numbers on the page are given with their dates, like 2001 and so on. So 2023 does not seem relevant in any way.
If you buy something from Tiktok, you presumably visit the merchant's website, which almost surely will have chosen to have a tracking pixel that sends data to FB (Instagram). You can read a bit about how tracking pixels work here: https://jvns.ca/blog/how-tracking-pixels-work/
In this case it's not Tiktok and Instagram that are sharing data with each other, but the product website that is choosing to share data with both of them.
What if the AI kept bringing up unspecified cases and all you (the human) had to do all day was respond to it on what the behavior should be in each case? In this model the AI would not specify the outcomes; the specification is whatever you initially specified, and your responses to the AI's questions about the outcomes. At some point you'd decide that you'd answered enough questions (or the AI could not come up with any more unspecified cases), and bugs would be in what remained, but it would still mean substantially more thinking about cases than now.
There are some basic invariants like "this program should not crash on any input" or "this service should be able to handle requests that look like X up to N per second" — though I expect those will be the last to be amenable to formal verification, they are also very simple ones that (when they become possible) will be easy to write down.
> "this program should not crash on any input" [...] though I expect those will be the last to be amenable to formal verification,
In the world of Rust, this is actually the easiest to achieve level of formal proofs.
Simple lints can eliminate panics and potentially-panicking operations (forcing you/LLM to use variants with runtime error handling, e.g. `s[i]` can become `s.get(i).unwrap_or(MyError::RuhRoh)?`, or more purpose-specific handling; same thing for e.g. enforcing that arithmetic never underflows/overflows).
Kani symbolically evaluates simple Rust functions and ensures that the function does not panic on any possible value on it's input, and on top of that you can add invariants to be enforced (e.g. search for an item in an array always returns either None or a valid index, and the value at that index fulfills the search criteria).
(The real challenge with e.g. Kani is structuring a codebase such that it has those simple-enough subparts where formal methods are feasible.)
This was a nice profile of (one side of) Sacks and his life, and as usual some mischievous or click-seeking online editor has given it a headline (and sub-heading) that are almost completely unrelated to what the article is about. In fact, at the bottom it says:
> Published in the print edition of the December 15, 2025, issue, with the headline “Mind Over Matter.”
and a headline like that (saying nothing) would be more appropriate to this.
The very fact that Sacks wrote about his patients has always had its detractors—based on his book The Man Who Mistook His Wife for a Hat, someone called him “the man who mistook his patients for a literary career”—but what was surprising (to me) from this article is that it seems that after that early book, he actually became careful not to exaggerate or make up stories, to the extent that someone closely following him looking for discrepancies was not able to find any. I would have expected the stories to be mostly fictional, but it appears that this is so only of his early books.
I assumed the books were somewhat fictional (i.e. they were Gladwell-style) because if he meant to make a claim seriously he'd have published in a medical journal instead of a popular/literary book. But since writing the comment above, I've learned that over the years many people actually believed that all details in the books were literally true (you can search for e.g. [Sacks prime] to see many people who took the story seriously and analyzed them), which does put things in a different light.
To me, it's when narrative has priority over accuracy. There are a number of popular edutainment figures who fit this mold, but Gladwell is probably the most prominent example.
Interesting, there seems to be a different Sacks (software engineer Robert Sacks) who devised his own Ulam-like spiral.
Anyway sorry I didn't keep track of the pages I visited, but here are some of the search results I see now, indicating at least some time wasted exploring something that did not need to be explored if it had been clear that the story was not fully real:
BTW the same account of the twins (https://archive.is/MmogP) also has several paragraphs revealing major misunderstandings on the part of Sacks, e.g.:
> And yet they are called “calendar calculators”—and it has been inferred and accepted, on next to no grounds, that what is involved is not memory at all, but the use of an unconscious algorithm for calendar calculations. When one recollects how even Carl Friedrich Gauss, at once one of the greatest of mathematicians, and of calculators too, had the utmost difficulty in working out an algorithm for the date of Easter, it is scarcely credible that these twins, incapable of even the simplest arithmetical methods, could have inferred, worked out, and be using, such an algorithm.
[Needless to say, calculating the day of the week for a fixed date is a much easier and completely unrelated problem to that of Easter, “the first Sunday after the Paschal full moon (a mathematical approximation of the first astronomical full moon, on or after 21 March” https://en.wikipedia.org/wiki/Date_of_Easter ]
Rmapi calls to sync. My use case is updating an annual calendar pdf which is inked on tablet but shows calendar updates day to day, so I run it on a cron
This couldn't be further from the truth. He has given several talks where he's projecting his computer, you can see him comfortably switching between all the programs he uses (Emacs, Mathematica, etc); in fact he is very efficient and has them customized just the way he wants it. (I even recall some blog post where the author watched one of these talks and was amazed by just wizardly he was navigating between programs or Emacs buffers or whatever.)
If you scroll down to the bottom of https://cs.stanford.edu/~knuth/programs.html you can see his configurations for Emacs and fvwm and even macOS keyboard layouts; some of them were updated as recently as this year.
> I designed my own bitmap font for use with Emacs, because I hate the way the ASCII apostrophe and the left open quote […] I prefer rxvt to xterm for terminal input. Since last December, I’ve been using a file backup system called backupfs, which meets my need beautifully […] Incidentally, with Linux I much prefer the keyboard focus that I can get with classic FVWM to the GNOME and KDE environments that other people seem to like better. To each their own.
This was a good concise review! The book itself is surreal.
One thing worth pointing out that is Knuth wrote it just as the characters described: he had only a vague memory of the initial axioms that Conway had told him, so the characters' discovery matches his own process. (He wanted to illustrate how research is done, how new things in mathematics are discovered.) He gave himself a week during his sabbatical year, just locked himself in a hotel in Norway, wrote it over 6 days, “and on the seventh day he rested”. (Though he later took another week to rewrite the whole book as he had made a mistake in the axioms.) He tells the story here (see 40–42): https://www.youtube.com/watch?v=lICebKDaZRU&list=PLVV0r6CmEs...
And this book was published before Conway published anything about surreal numbers, so surreal numbers are probably the only mathematical concept whose first publication was as a book of fiction. The name “surreal numbers” also comes from Knuth (while Conway simply called them “numbers” — he later came around to agreeing that this was a good name).
reply