2024 year-end link clearance
The cycle completes. The post 2024 year-end link clearance appeared first on The Old New Thing.
We made it to the end. It seems to get harder every year.
- An Abundance of Katherines: The Game Theory of Baby Naming (html version) develops a simple model for baby naming that helps explain why baby names follow trends. And since it is a SIGBOVIK paper, it containing a bunch of jokes. Appropriately, the paper is authored by Katy Blumer, Kate Donahue, Katie Fritz, Kate Ivanovich, Katherine Lee, Katie Luo, Cathy Meng, and Katie Van Koevering.
- It’s not a coincidence that π² ≈ g. It’s due to Christiaan Huygens, though he didn’t realize it.
- Every UUID. A list of all the v4 UUIDs. The real magic is the search function (Ctrl+F). Once you have reveled in the page, read the secret to the magic.
- Another entry in the long list of “bugs that are very difficult to reproduce” is the game that crashed but only in Japan and only at 4am.
- Arthur O’Dwyer explains why you shouldn’t blindly prefer
emplace_backoverpush_back. Basically, this is “Remember what ’emplace’ means (or learn it now if you were never taught what it means).” - Matt Keeter learns about the CPU return address predictor.
- C++ Seasoning, a 2013 talk by Sean Parent which opens with a principle that has stuck with me: No raw loops.
- If you’ve been wondering where the old MSDN blogs went, most of them have been archived on learn.microsoft.com/archive/blogs.
- My colleague Joe Bialek gave a fascinating and eye-opening talk on pointer validation errors in the kernel, and how the C and C++ languages give the compiler leeway to perform optimizations which invalidate all your hard work.
- I mentioned some time ago that the Dazzle hardware was the original target for Windows NT, and while it was able to run Reversi, the pieces were square because GDI didn’t know how to draw circles yet. Well, here’s some visual evidence.
