RE: https://mastodon.social/@stroughtonsmith/116030136026775832

This is one of the worst takes from LLM enthusiasts.

Compilers are deterministic, extremely well tested, made out of incredibly detailed specifications debated for months and properly formalized.

LLMs are random content generators with a whole lot of automatically trained heuristics. They can produce literally anything. Not a single person who built them can predict what the output will be for a given input.

Comparing both is a display of ignorance and dishonesty.

OK, if you're a huge nerd who spends Saturdays online too much I'm sure you've heard about the NIST NTP servers failing and being taken offline¹ ² ³.

If you're in Canada, please know that our own NRC offers stratum-2 time sources as a government-funded public service⁴ & monitored/auth'd stratum-1 services on a cost recovery basis⁵.

We can be proud of this.

¹: https://mathstodon.xyz/@johncarlosbaez/115752907255467092

²: https://hachyderm.io/@jbcrawford/115753138339762616

³: https://lists.nanog.org/archives/list/[email protected]/message/ACADD3NKOG2QRWZ56OSNNG7UIEKKTZXL/

⁴: https://nrc.canada.ca/en/certifications-evaluations-standards/canadas-official-time/network-time-protocol-ntp

⁵: https://nrc.canada.ca/en/certifications-evaluations-standards/instrument-calibration-services/frequency-time-calibration-services

I think many people misunderstand the purpose of code review. The purpose of code review is not for the reviewer to find bugs, and certainly not for them to ensure that the code is bug-free. Anyone who depends on code review to find bugs is living in a fool's paradise. As everyone should know by now, it is not in general possible to find bugs by examining the code.

The primary purpose of code review is to find code that will be _hard to maintain_. The reviewer looks at the code and tries to understand what it is doing and how. If they can't, that means it will be hard to maintain in the future, and should be fixed now, while the original author is still familiar with it.