Goal: flags situations that might be hazardous when they occur to trigger some fallback mechanism.

Three steps:

1. check if you are even in the design: operational design domain monitoring
2. even if you are, check how certain you are: uncertainty quantification
3. once you know what your model is going to do: failure monitoring

operational design domain monitoring

“Out of distribution detection.” Insight: if we operate outside of the ODD, our validation results may no longer be valid, so we should monitor whether or not we are in the operational design monitor whether or not we are in the operational design domain.

Rural Electrification Administration create electrification throughout cities. Most of American infrastructure still 1930s.

Observations from studying the comedian Russel Howard.

• Stretching analogies
• Using language/motion/figure do describe something on the opposite end of the spectrum
  • Take, for instance, age: 5Y/O: “cheers mum, wasen’t on my to-do list”
  • A surprisingly sentimental dog: “because when I wake up tomorrow I want to see you, and I want to go for a lovely walk”
• Large motions + deadpan after
• Endless extrapolations of a normal setup: setup: Russian hackers were controlling people’s toys; punchline: “5 men were dildo’d to death, we don’t have a recording but here are their final words — ‘oh yeaaah’, ‘oh fuck yeaaah’”

Setup: gweneth paltro Punchline: “put an egg up there, you will feel more femenine. no! you will feel like a chicken”

1. if paper has software isolation using Rust, sort into software isolation categories
2. if the paper does not, scan for incorrect claims and then sort into either irrelevant or wrong mutability (keywords: mutable, borrow, reference, invariant)

1. IM/PUI does not cause an issue because of narrow extension interface (NetBricks)
2. IM/PUI does cause an issue, and is handled
3. IM/PUI does cause an issue, and is deferred to future work
4. IM/PUI does cause an issue, and is ignored (RedLeaf, Splinter)
5. No User Language Extension/No Isolation Needs

• How many papers use Rust, how many use Rust for language extensions? (a count on these would be nice if possible)