stared

37 ms·

571.
by stared 1y ago
People fiddle with SEO with a lot of effort and some mixed success. While it takes a single solid hit at the HN (or Reddit) to get on the top. Of course, it means that a post needs to have more than suitable keywords. So, I never sacrifice
572.
by stared 1y ago
I have a slightly different reading on Ted Chiang's approach to free will. As I see it, free will is a perspective - not something true or false, just looking at things from a certain angle. Knowledge of future is another one - and it
573.
by stared 1y ago
I it the same story as I keep sending as a taster for Ted Chiang.
574.
The Pragmatic Engineer 2025 Survey: What's in your tech stack?(newsletter.pragmaticengineer.com)
2 points by stared 1y ago|0 comments
575.
by stared 1y ago
A lot of debugging, code and mind alike, benefits from rubber ducking. LLMs do it on steroids. At the same time, if you take their output as some objective truth (rather than stimulus), it can be dangerous. People were already doing that wi
576.
by stared 1y ago
It was not my intention to bring the HN hug of death. (For a reference, I shared a link, I am not the author.)
577.
by stared 1y ago
My take is that it’s easier to train a model to ace short, low-context tasks like IQ tests. That doesn’t necessarily transfer to more complex reasoning. While on the Mensa Norway test GPT-5 gets over 140, on an offline test it goes down to
578.
IQ tests results for AI(trackingai.org)
220 points by stared 1y ago|343 comments
579.
by stared 1y ago
Not sure if "Python 4.0" is the best comparison. Go is compiled, Python - interpreted. Go focuses on concurrency, Python on making it easy to do things with code. Many things that are a one-liner in Python (but itself or with a li
580.
Sandboxing AI-Generated Code: Why We Moved from WebR to AWS Lambda(quesma.com)
4 points by stared 1y ago|0 comments
581.
WeirdML Benchmark v2(htihle.github.io)
1 points by stared 1y ago|0 comments
582.
Canada airdrops aid into Gaza, says Israel violating international law(reuters.com)
16 points by stared 1y ago|0 comments
583.
A quantum leap for antimatter measurements – the first antimatter qubit(home.cern)
3 points by stared 1y ago|0 comments
584.
by stared 1y ago
People (even from the same culture, with the same native tongue) use words differently. It is one of the reason for which I am mildly skeptical about all such things as "How happy you are on the scale 1-10?" - no matter what is th
585.
Claude Code + OpenTelemetry + Grafana: A guide to tracking usage and limits(quesma.com)
3 points by stared 1y ago|0 comments
586.
by stared 1y ago
I have the same impressions. Typing helps a lot, and (I think) in a few ways - one is being a safe guard, second a constraint (so say, AI is less likely to create a clunky variable which can be a string, list, or a few other things), third
587.
First-Ever Antimatter Qubit Could Help Crack Cosmic Mysteries(scientificamerican.com)
3 points by stared 1y ago|0 comments
588.
by stared 1y ago
> I love fireflies. But in recent years, they stopped coming for reasons I don’t know. No tiny, glowing dots in the dark like they used to. I miss them more than I expected. Light pollution, and even more: pesticides. Population of all i
589.
by stared 1y ago
It is interesting to see different formats for notebooks - Jupyter uses JSON, RMarkdown (as you may have guessed) Markdown, and Observable - HTML.
590.
by stared 1y ago
Also it took surprisingly many clicks to see any code. After 3 clicks I had impression that it is not yet a language, just and idea for one.
591.
by stared 1y ago
When I saw "Your device" I assumed it is about device for watching, not about specs of the currently used device. It should be communicated in a clearer way.
592.
by stared 1y ago
Roko's Basilisk is a modern version of Pascal's Wager. The same counterarguments works. Just search-replace God to AI, are you are done.
593.
by stared 1y ago
I agree that after a challenge is something can be done at all (heavier-than-air flight, Moon landing, Gold medal at the IMO) then next question is makes sense economically. I like ARC-AGI approach for the reason that it shows both axes - s
594.
by stared 1y ago
Sure, some people use classes when all they need is a namespace - and for that having a file is enough. If there is a collection of related data, it is (often) good to have it as a class. In the config example, a class may work better - as
595.
Which chart would you swipe right?(quesma.com)
3 points by stared 1y ago|0 comments
596.
by stared 1y ago
I guess the name comes from "find", spelled alternatively. My initial sense comes from "PhD" (PhD-index?).
597.
by stared 1y ago
I disagree. In each practical project, there is an order of magnitude more things than you could do. One of the crucial job is to prioritize things well, knowing the context. Regular tasks trackers put things outside of context. Low-priorit
598.
OpenAI Places Second Behind Human Coder at AtCoder Progmming Event(officechai.com)
7 points by stared 1y ago|1 comments
599.
by stared 1y ago
Coauthor here. We got frustrated that creating custom dashboards is more work than it should be - especially the boring and tedious parts, like sifting through 150 columns to check which ones contain more than just NULLs, let alone figuring
600.
Building Grafana dashboards with AI, CLI and a bit of pragmatism(quesma.com)
5 points by stared 1y ago|1 comments

More ›