fourthark

7 ms·

61.
by fourthark 1y ago
> A study asked different age groups to mentally count 120 seconds. People under 30 averaged 115 seconds; those over 50 just 87. That's a 24% reduction in perceived time. This seems bogus to me. I’m 51; I set a timer on my phone for
62.
by fourthark 1y ago
Who builds phones?
63.
by fourthark 1y ago
Upgrades and “collateral evolution” are very strong use cases for Claude. I think the training data is especially good, and ideally no logic needs to change.
64.
by fourthark 1y ago
Originally it was based on HyperTalk, then it switched to ECMA later on.
65.
by fourthark 1y ago
How long would it have taken if you had written “the detailed specs, detailed commit log, and thousands of tests”?
66.
by fourthark 1y ago
I asked Grok and Gemini and they both said there have been reports that Google search has dropped below 90% for the first time, so it’s significant but it’s like a 1-2% drop.
67.
by fourthark 1y ago
I think it’s humanities slop and not AI slop but I’m not sure…
68.
by fourthark 1y ago
So describe your experience without being a booster
69.
by fourthark 1y ago
I ran into this too. In my case it was exactly the kind of situation where I would also run into trouble on my own - trying to change too many things at once. It was doing superbly for smaller, more contained tasks. I may have to revert and
70.
by fourthark 1y ago
AGPL suggests you are correct
71.
by fourthark 1y ago
I'd add, ask it to ask you questions! That way you can make sure it understands the problem.
72.
by fourthark 1y ago
Oh I've definitely seen that too, even with common front-end stuff. I think people might be exaggerating how much they are out of the loop. Claude often needs to be guided away from stupid decisions, and it will absolutely delete code
73.
by fourthark 1y ago
You're probably in an obscure niche domain, or asking it to do something creative. Try like upgrading JS package dependencies, or translating between languages, limited tedious things, and you will be surprised how much better it does.
74.
by fourthark 1y ago
This is all stuff for private use. It's credible that the author built these things with LLMs, not that they are secure or robust.
75.
by fourthark 1y ago
Like anyone, it can always keep adding new code that “works”. The problem is that it’s causing other code not to work at the same time, and it doesn’t see that.
76.
by fourthark 1y ago
A thermostat is a cybernetic device.
77.
by fourthark 1y ago
Maybe there's some copying and pasting involved. ;-)
78.
by fourthark 1y ago
I’ve done this without AI. The thing was not as hard as I thought it would be.
79.
by fourthark 1y ago
I’ve used this productivity hack without AI!
80.
by fourthark 1y ago
Pretty common word online, especially on the site once called Twitter.
81.
by fourthark 1y ago
Don't worry, this script dangerously skips permissions (by default?) You don't need to press 'y'.
82.
by fourthark 1y ago
Harmonic analysis?
83.
by fourthark 1y ago
Says nothing about economic feasibility, but I guess they weren't asked.
84.
by fourthark 1y ago
Oracle purchased Sun, which purchased Netscape. I had to look it up. Edit: more complete history https://news.ycombinator.com/item?id=44408760
85.
by fourthark 1y ago
> "The ontology of all of this is nebulous"
86.
by fourthark 1y ago
A lot of companies think they do
87.
by fourthark 1y ago
He covers error correction and information densities on the linked page: https://www.monperrus.net/martin/perfect-ocr-digital-data (Last section before conclusion.) IIUC this provided the best overall reliable informat
88.
by fourthark 1y ago
LLM, please find all occurrences of "shut the fuck up" and suggest actual arguments for the author to use. Otherwise, well written and persuasive.
89.
by fourthark 1y ago
Well that's a frightening scenario... but how does it relate?
90.
by fourthark 1y ago
Who is "we"?

More ›