Hacker Newsnew | past | comments | ask | show | jobs | submit | redtaperat's commentslogin

For 100 million people and a confidence interval of 99% you only need 3000-4000 samples, depending on your moe.


At that point just elect them! https://en.wikipedia.org/wiki/Sortition


“Solve the straight male loneliness epidemic.”

FTFY. Gay men have many more community options. Not sure why, but we just do togetherness inherently.

Glad to see this though. It can only be solved by the people in the epidemic but it only takes the effort of a handful of organizers in the community/town/city to jumpstart events and hit critical mass. I’ve organized things on meetup and it takes persistence. Good luck!


> Gay men have many more community options. Not sure why, but we just do togetherness inherently.

How much of that is motivated by dating and hooking up?


I would guess its more like if you want to meet any other gay men you have to work at it because you are a minority. Being in a minority does engender some comradery all by itself.


All caffeine consumption stopped at noon. It’s literally in the paper.


That's the time that the subjects stopped consuming caffeine as part of their usual diet, not the time that the pill (either 200 mg caffeine or a placebo) was administered. The subjects were required to maintain their usual caffeine intake (to prevent withdrawal), but had to stop by noon to prevent that caffeine from interfering with the experiment.


Thanks for clarifying.


From bill cosby’s “Himself” live standup recording from the early 1980s: “Cocaine intensifies your personality … but what if you’re an asshole?”


This is one of several shortcomings I’ve encountered in all major LLMs. The llm has consumed multiple versions of SDKs from the same manufacturer and cannot tell them apart. Mixing up apis, methods, macros, etc. Worse is that for more esoteric code with fewer samples, or more wrong than right answers in the corpus means always getting broken code. I had this issue working on some NXP embedded code.


Human sites are also really bad about mixing content from old and new versions. SO to this day still does not have a version field you can use to filter results or more precisely target questions.


We have MCP servers to solve that (and Context7)


I see those as carefully applied bandaids. But maybe that’s how we need to use AI for now. I mean we’re burning a lot of tokens to undo mistakes in the weights. That can’t be the right solution because it doesn’t scale. IMO.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: