Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
nojs
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
121.
▲
by
nojs
11mo ago
It’s ironic because you have to go to additional effort to turn autocorrect off, which contradicts the “i dont even care” effect they’re going for
122.
▲
Coffee Sniffers
(en.wikipedia.org)
3 points
by
nojs
11mo ago
|
0 comments
123.
▲
by
nojs
11mo ago
If you don’t collaborate before it’s shipped and don’t retroactively review after it’s shipped, how do you provide input?
124.
▲
by
nojs
11mo ago
This nag often comes up at the very beginning of a session, before you could realistically evaluate whatever A/B test they’re running with any accuracy. I hope they’re not taking the data too seriously.
125.
▲
by
nojs
11mo ago
Same. It used to go down all the time but the past couple of months it’s been very stable.
126.
▲
by
nojs
1y ago
The font size is perfect for me, and I hope it doesn’t get a “usability update”.
127.
▲
by
nojs
1y ago
It’s definitely an imperfect signal, but we are capable of making decisions under conditions of noise and uncertainty. In other fields we would quantify the uncertainty, the consequences of making the error in either direction, and then act
128.
▲
by
nojs
1y ago
I do not buy this argument. The error would be in misinterpreting the image and taking the unnecessary treatment, not in doing the test in the first place. How is there any benefit in having less information?
129.
▲
by
nojs
1y ago
Apparently it’s fine according to this xkcd: https://youtu.be/EFRUL7vKdU8
130.
▲
by
nojs
1y ago
The reason is cost and availability, not risk.
131.
▲
by
nojs
1y ago
The “X% AI coded” metric needs to die. It is completely meaningless. The nature of development work is changing. A project can be 100% written by AI but guided so closely by humans that the process wasn’t much faster. Alternatively, AI can
132.
▲
by
nojs
1y ago
FYI you can avoid the manual sprinkling of select_related somewhat with https://pypi.org/project/django-auto-prefetch/ , which avoids the lowest hanging N+1 loops automatically. Definitely not a cure all but it hel
133.
▲
by
nojs
1y ago
This is going to be extremely useful. A lot of people have hacked together similar things to get around waiting for CC to finish without mangling worktrees and branches manually. I was curious how the 'Open in CLI' works - it copi
134.
▲
AI Darwin Awards – 2025 Nominees
(aidarwinawards.org)
3 points
by
nojs
1y ago
|
0 comments
135.
▲
by
nojs
1y ago
https://archive.is/slfnP
136.
▲
by
nojs
1y ago
I think this is hard with only a photo because you can’t always see what’s inside. But I’ve always dreamed of something like this paired with some kind of affordable hardware scanner that can get just enough data to fill in the blanks from
137.
▲
Rue Merdiere
(languagehat.com)
2 points
by
nojs
1y ago
|
0 comments
138.
▲
by
nojs
1y ago
Incidentally this is one reason why people in the past seemed braver than now and did crazier things. When your life expectancy is 25, you take a lot more risks.
139.
▲
by
nojs
1y ago
Where did you get that figure from?
140.
▲
by
nojs
1y ago
It feels similar to Google to me - what is (was) their moat? Basically slightly better results and strong brand recognition. In the later days maybe privileged data access. But why does nobody use Bing?
141.
▲
by
nojs
1y ago
I think people are massively underestimating the money they will come from ads in the future. They generated $4.3B in revenue without any advertising program to monetise their 700 million weekly active users, most of whom use the free pro
142.
▲
by
nojs
1y ago
> The most useful thing about Cursor was always state management of agent edits Note that CC introduced this yesterday, it’s very fast and good.
143.
▲
by
nojs
1y ago
I mean a naive implementation of this would just make regular git commits to a special hidden repo and revert them (ignoring changes outside project root). I always assumed that’s how cursor did it. Presumably they have good reasons not to
144.
▲
by
nojs
1y ago
Great writeup, thanks.
145.
▲
by
nojs
1y ago
From the docs it looks like this feature only reverts the edit tool calls, and not e.g. bash commands that have been executed: > Checkpoints apply to Claude’s edits and not user edits or bash commands, and we recommend using them in comb
146.
▲
by
nojs
1y ago
36M subs in a year? That account is super fishy https://www.reddit.com/r/youtubers/comments/1lfbz86/who_is_m...
147.
▲
by
nojs
1y ago
This is very relevant to this release. It’s way faster, but also seems lazier and more likely to say something’s done when it isn’t (at least in CC). On net it feels more productive because all the small “more padding” prompts are lightni
148.
▲
by
nojs
1y ago
The “big money” is having the attention and trust of high purchase intent users, which OAI does
149.
▲
by
nojs
1y ago
How are you extracting this - aren’t the main labs obfuscating these (meaning it’s likely to be a decoy or incomplete version)?
150.
▲
by
nojs
1y ago
A syntax highlighter for Chinese: https://dragonmandarin.com
More ›