Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
davidmurdoch
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
91.
▲
by
davidmurdoch
9mo ago
We do care. We REALLY don't want AI on by default on our PCs.
92.
▲
by
davidmurdoch
9mo ago
Are you looking at the wrong website?
93.
▲
by
davidmurdoch
9mo ago
David Bars, while not even close to anything resembling a whole food, have made hitting macros so much easier. End up being cheaper than chicken, per gram of protein per calorie, sometimes too!
94.
▲
by
davidmurdoch
9mo ago
The administration putting out the "eat more meat" guidance is simultaneously telling everyone they need to work out. The recommendation seems consistent when their started goal is to change current "American" habits.
95.
▲
by
davidmurdoch
9mo ago
And none of that, if true, matters in the slightest to most of America because they otherwise wouldn't be able to afford these foods.
96.
▲
by
davidmurdoch
9mo ago
And costs significantly more money.
97.
▲
by
davidmurdoch
9mo ago
The "... just let me know if there is anything else you'd like to know." after every long-winded explanation is so infuriating.
98.
▲
by
davidmurdoch
9mo ago
Are you using the same models and thinking levels for each? I too have found Codex better than Copilot, even for simple tasks. But I don't have the same models available since my work limits the models in copilot to the stupid ones.
99.
▲
by
davidmurdoch
9mo ago
Naming things is hard. So hard every AI company isn't even trying to come up with good names.
100.
▲
by
davidmurdoch
9mo ago
Wait. Why? What prompted this? The phrasing?
101.
▲
by
davidmurdoch
9mo ago
In the US, yes.
102.
▲
by
davidmurdoch
9mo ago
Only like $400 million per fab.
103.
▲
by
davidmurdoch
9mo ago
State of the art lithography currently uses extreme ultraviolet, which is 13.5nm. So maybe they are EUV laser-focused, just with many mirrors pointing it in 5 different directions?
104.
▲
by
davidmurdoch
9mo ago
I don't mind the icons, but the snow effect on the website makes it much harder to read.
105.
▲
by
davidmurdoch
10mo ago
Great point!
106.
▲
by
davidmurdoch
10mo ago
Yeah, I'm not seeing fast uncached times either. I usually hit Cloudflare's Miami datacenter, which is only about 200 miles and very low latency. But I'm seeing 200+ms on this site right now.
107.
▲
by
davidmurdoch
10mo ago
I only meant in this example. I agree, sites in general are fat.
108.
▲
by
davidmurdoch
10mo ago
But now it's that fast from almost everywhere on the planet, with nearly zero effort from the developer. We've been limited by light speed here, not compute.
109.
▲
by
davidmurdoch
10mo ago
Most front end engineers could do it in JS without ever having to look something up. But the CSS to do it is still obscure to most.
110.
▲
by
davidmurdoch
10mo ago
Huh. I know hundreds that use LLMs in a VSCode based IDE, and 3 that use the CLI.
111.
▲
by
davidmurdoch
10mo ago
Someone else said it's not actually interactive. So which is it?
112.
▲
by
davidmurdoch
10mo ago
My shield turns my receiver on, sets it to the right input (then the wrong input, then back to the right input), then... disables the decoder so there is no sound. Then sometimes enables the decoder about 20 seconds later.
113.
▲
by
davidmurdoch
10mo ago
14. Ask why it was banned and they respond with something like "oh you know what you did".
114.
▲
by
davidmurdoch
10mo ago
Is this not a bipartisan issue?
115.
▲
by
davidmurdoch
10mo ago
I can't figure out an answer to why they'd do this. They're still supporting their custom Chromecast "receiver" app (for now?), which is a standard HTML app. But now they're detecting the Chromecast-capable dev
116.
▲
by
davidmurdoch
10mo ago
It'd be a good option for failover Internet.
117.
▲
by
davidmurdoch
11mo ago
Your can do that now in Typescript. But it will take several minutes to resolve your types. I've done it and it's a horrible dev experience, sadly.
118.
▲
by
davidmurdoch
11mo ago
It's not though. Chromecast, g suite legacy, podcast, music, url shortener,... These weren't small products.
119.
▲
by
davidmurdoch
11mo ago
This just happened to me this week. I work on the platform everyone builds on top of. A change here can subtlety break any feature, no matter how distant. AI just can't cope with this yet. So my team has been told that we are too slow.
120.
▲
by
davidmurdoch
11mo ago
I've mostly stopped using Claude because of it, it will still try use Python for the most random tasks. It recently wrote an HTML file with some inline js in it, then started a local python server to open the HTML file, and check the l
More ›