suboptimization; learned about this in engineering school 40 years ago: don't optimize the most visible parameter by itself, or you'll likely end up crashing. why do we have to freaking rename everything?
6:50......TODO wake up
7:00......TODO brb
7:02......TODO catfood
7:04......TODO fasting glucose
7:05......TODO ketones
7:06......TODO weight
7:07......TODO hydrate
7:08......TODO am meds
7:10......TODO shower
7:20......TODO hair
7:22......TODO shave
7:25......TODO teeth
7:27......TODO deo
7:28......TODO morning netty
7:28......TODO wash netty
7:33......TODO chg jeans (+3d)
7:33......TODO dress
7:38......TODO pockets
7:42......TODO litter
7:45......TODO bkfst/coffee
8:00......TODO journal
NB: this is a good time to
apply the quiet art of attn
(see my other HN post) and do
your best to like or dislike
current thought patterns
8:15......TODO email
8:30......TODO chats
8:40......TODO calendar
8:45......TODO planning
9:00......TODO login/execute
12:00......TODO lunch/nap
18:00......TODO logoff
18:10......TODO text friends
it's not just morning that matters. and don't let the
jerks and trolls get to you:
it's your time and your life.
At 60, I applied to Canonical and got hired. The hiring process is thorough but very fair, all WFH, and they generally don't care about personal attributes. Gave me 20% above my asking price and a higher starting rank than expected.
I thought Canonical was the place where every candidate had to write a bunch of essays and talk about how they did in math in high school (regardless of how long ago that was).
Glad you had a good interview experience there, but it's genuinely the first time I've heard of it happening.
I interviewed with Canonical in 2023. It was an enormous waste of time. I would never suggest anyone interview with them while they still have the same interview process.
I would struggle to have a polite response to that question. Come to think of it, the last company that even asked me about high school was a horribly toxic place, and it'd be a red flag that this wasn't a place for me.
forget benchmarks. use emacs lisp as a test case. i have yet to find an LLM that can consistently generate working functions in elisp (possibly even lisp itself).
attempt to interpret the parent comment more charitably: when different people who've shared a common experience try to put it into words, there is some consistency between what they write. the harder the experience is to verbalize, the harder it is to come up with meaningfully different precise descriptions of it, and the more aware you are of the limitations of language
reply