ivanjermakov

Embarrassing...

I also had a great experience consulting LLM regarding registering a business as a foreigner. Except I'm still very sceptical since it's not my area of expertise and consequences for a mistake are very high.

Dang, need to add two more +s

It's certainly a chore, especially for a project with multiple dependencies (i.e. a game). I wrote about it: https://news.ycombinator.com/item?id=49851097

> The pitch was that AI would give us our time back

The pitch was that same amount of work would be delivered 10 times faster. Meaning you would do 10 times more work in the same time span.

No employer will ever pitch you 4 hour work week, only Tim Ferriss.

Software bottleneck has moved from hardware performance to management efficiency.

Oh you're right, except since LSP 3.17 position encoding can be negotiated with clients/servers that support it. I initially got confused because my work with LSP started with 3.17 and I always used UTF-8 encoding.

https://microsoft.github.io/language-server-protocol/specifi...

No? In LSP source locations are line+character: https://microsoft.github.io/language-server-protocol/specifi...

https://microsoft.github.io/language-server-protocol/specifi...

What's wrong with UTF-16 for transport?

It will be even more wild when investors will notice growth slowing down.

Delivering iPad mini but no iPhone mini is double disappointing.

Do you know where <||> syntax is coming from? I find it very odd that existing markup wasn't used and there is very little info online about it.

There is more than ever unresolved usability holes in programmer's workflow. Coordinating work across agents and other team members, making specs and change them together with implementation, code review, support workflow of different level of "vibiness", give agents more IDE features to play with, and much more. JetBrains has both capacity and expertise to deliver a product in this space.

Because now it's much harder to gauge how much human effort went into building a project.

Also basic brain chemistry: seeing a pack of impressive (by 2020 standards) solo dev projects a day makes it the new norm, raising the bar for the future. But now I feel like this bar is not reachable because of no way to accurately measure "impressiveness".

LLMs might help you find harder challenges.

I'm no longer excited about projects other people share online.

Do I understand it right that it can be set up in a way that if exploit is used it can be "recorded" and revealed?

I like it this way, it gives correct expectations regarding effort put into the rest of the project.

> fits so much better in "regular" company/organization projects

How so? I used both at work and they're functionally the same.

UTM was almost good enough but the procedure to enable JIT every time you run it was extremely clunky.

Maybe tomorrow we get Android running on the iPhone (and Linux on the iPad please).

Reminds me of NFT trolls "sorry I just screenshot it for free".

Prompts as a service is something new.

Agent runtime - actually a good name for what is called today coding agent or agent harness.

Dense UIs are work of art when done right, especially in combination with keyboard centered workflow.

Maybe https://news.ycombinator.com/item?id=48201484

The first rule of the fight club only applies if you know about the fight club.

`npx dogsay`?

Because "10 out of 10" is a better headline for the average reader than "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H".