This definition only covers memory modification but read-access must also be so restricted, which in practice is as well.
@adamshostack Was your preference of the first linked definition that "certain types of bugs"? If so I think that can be tightened up along the lines of, "prevents programmers from accessing outsides the bounds of allocated (and not yet freed) memory".
If not what?

Great point. I'd be astonished if any of the crawls consider that - it's a pure externality.

I think these qualify: Chapter 3 (mitigations) and 4 (secure patterns) from my book.
https://designingsecuresoftware.com/text/ch4-patterns/

I was thinking about an zero cost and easy way to prove primacy creating a document. While there's no serious security here, at all, thanks to federation a post gets quickly propagated publicly so it would be very hard to hack all the host servers later. Like this, but etiquette requires only occasional use, this isn't a blockchain. Hashtag #dibs
12036e414db4c1ec821f7b1f21f72afced91562248d1c8c1690d4a99f091a06b00f3e9fbfa25469d17102a2911283da79e7b57eecb682e10b441e4f6f7673db1 (sha512) or 6bfc959c1abeed8c9b1d832824cf0083302b6f7cc1951178cae06217cfe785ed (sha256)

The LLM grammar fixer I use acts silly in that doesn't seem to know about threat models. It may correct "threat model" to "threaten the model" or "threaten to model", neither of which I would ever do but it's fun trying to imagine.

Embarrassing: "After installing the August 2025 non-security preview update (KB5064081) or later updates, you might notice that the password icon is not visible in the sign-in options on the lock screen."
Rather than ridicule, I have one question: How does this kind of bug get released with professional testing?
https://support.microsoft.com/en-us/topic/august-29-2025-kb5064081-os-build-26100-5074-preview-3f9eb9e1-72ca-4b42-af97-39aace788d93