Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
zulban
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
121.
▲
by
zulban
1y ago
> Group education isn't my job It doesn't sound like I'd like your dev culture. This is also explicitly part of the work objectives of me and my team. > Sometimes it doesn't need to be smart I prefer code that is d
122.
▲
by
zulban
1y ago
Rough. I hope to get automated CR bots in my org working soon. But with 2025 capabilities it should definitely only be brief feedback that people can choose to ignore.
123.
▲
by
zulban
1y ago
Maybe the decent comments hadn't had time to bubble upwards yet. HN is a relatively higher quality online forum but... I still only expect half the commenters to have read the full post. I'm curious to hear what rationale you part
124.
▲
by
zulban
1y ago
Indeed. OSS maintainers may need some kind of response like the one I've written here that can be strategically dropped on the worst "bad AI" contributions. I certainly wrote this for myself to make my job easier, anyway.
125.
▲
Why I'm declining your AI generated MR
(blog.stuartspence.ca)
108 points
by
zulban
1y ago
|
67 comments
126.
▲
by
zulban
1y ago
Because for most people hearing the term is the only education they get about the concept.
127.
▲
by
zulban
1y ago
He can be mostly right but also terrible for his own cause at the same time. Anyone that doesn't see that must not know even the tiniest fraction of the stories, or like him also has a cognitive disability.
128.
▲
by
zulban
1y ago
"These things should literally never be able to happen" If we consider "humans using a bank website" and apply the same standard, then we'd never have online banking at all. People have brain farts. You should ask y
129.
▲
by
zulban
1y ago
There's thousands of enterprise SaaS products. I seriously doubt the OP is referencing any grand audit of them or could even list 20. Yet one they dislike is "easily" the worst. They write like a child.
130.
▲
by
zulban
1y ago
Most people will never learn. It's an endless cycle.
131.
▲
by
zulban
1y ago
The entire point of AGI maybe. But until we get that, if it's useful then people can do it.
132.
▲
by
zulban
1y ago
> that's just called documentation. Ironically you sound very optimistic about AI capabilities here. This implies that an AI is just as capable as a human because all they need to contribute is what a human needs to contribute. Alte
133.
▲
by
zulban
1y ago
Absent owners of malls don't want to think or work with the community to earn money. They just want big anchor store contracts like it's the 90s.
134.
▲
by
zulban
1y ago
> GitHub is _easily_ the most unreliable SaaS product. I'm no fan of Microsoft either but when you say ridiculous things it's hard to take you seriously.
135.
▲
by
zulban
1y ago
Depends on whether it's a fast or slow takeoff and the control problem. Both are open questions. Personally I think we're in the midst of a slow take-off because AI researchers and IT support for them are all using AI. That gives
136.
▲
by
zulban
1y ago
> Even if they do make that breakthrough I don't see how that will equate to that kind of valuation The superintelligence breakthrough..? I don't think you realize what that word means. Every single white collar job could be au
137.
▲
by
zulban
1y ago
You must be American if you think very violent games are not extremely popular.
138.
▲
by
zulban
1y ago
Training data like audio, pictures, and video perhaps.
139.
▲
by
zulban
1y ago
A real person showing up is a huge cost and risk. No threat actor will continue an attack on just a hospital like that. The economics make no sense and any money is already extracted. Ransomware shops are very happy to just shotgun the inte
140.
▲
by
zulban
1y ago
Very nice. Years ago I applied to a weather centre for their data and supercomputer team. I color-printed screenshots of a game I made: a 3D Mars sim with real Mars terrain data where you built and collected data from Mars weather stations.
141.
▲
by
zulban
1y ago
It absolutely doesn't have to be good to be real. Fear of building something crappy is a great way of never building anything real though.
142.
▲
by
zulban
1y ago
There's no way that you have a computer with internet access but cannot afford a few dollars a month on a micro instance backend. My database for a big project is on the same machine as my web server which works fine. If you don't
143.
▲
by
zulban
1y ago
Yes, people say all kinds of things.
144.
▲
by
zulban
1y ago
> Remember ... No, I don't remember it like that. Do you have any serious sources from history showing that Segway hype is even remotely comparable to today's AI hype and the half a trillion a year the world is spending on it?
145.
▲
Finding a Job: Build Something Real
(blog.stuartspence.ca)
2 points
by
zulban
1y ago
|
7 comments
146.
▲
by
zulban
1y ago
Sometimes I find myself rewriting the same comments online over and over. That's when I write a blog post to just link to that. Let me know what HN folks think!
147.
▲
by
zulban
1y ago
I wasn't writing about consumers though. I was writing about "hackers" who might read this article and try this hack.
148.
▲
by
zulban
1y ago
You can always tell how much someone has tried Linux based on how they talk about it.
149.
▲
by
zulban
1y ago
"Because that's what it means to be a hacker." Sure. But to me "hacking" this cat and mouse game is not very compelling. I feel like I've seen a thousand articles exactly like this over the years. This won'
150.
▲
by
zulban
1y ago
Right back at you. If you think my comment is a non-sequitur, maybe you didn't read the article?
More ›