Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Vibing a non-trivial Ghostty feature

mitchellh.com

Post reply on HN
Print view View original

101–110 of 160 posts

Re: Vibing a non-trivial Ghostty feature

#101

Post by stavros »

tptacek wrote: post #100
stavros wrote: post #99

Earlier quoted context omitted.

Yeah, definitely. I do agree with the skeptics to a point, as I don't let the LLM write code without reviewing (it makes many mistakes that compound), but I'd still rather have it write a function, review and steer, have it write another, and so on, than write database models myself for the millionth time. It's not that I find it hard, I've just done it so many times that it's boring. Maybe I should be solving differ…

I was just talking about this in a chat today, because 'simonw had at some point talked about getting to the point where he was letting go of reviewing every line of LLM code, and I am nowhere close to that point --- I'll take Claude's word on Tailwind classes as long as the HTML looks right, but actual code, I review line-by-line, token-by-token, and usually rewrite things, even if just for cosmetic reasons.

Yeah, there's a definite continuum for where LLMs are most to least expert. They seem to be fairly OK with JS, less so with Python, and C is just a crapshoot.

It depends on the project as well, for throwaway things I'm fine to just let it do whatever it wants, but for projects that I need to last more than a few days, I review everything.

Re: Vibing a non-trivial Ghostty feature

#102

Post by jebarker »

mitchellh wrote: post #83
jebarker wrote: post #80

Earlier quoted context omitted.

It was this one: https://sourcegraph.com/blog/dev-tool-time-mitchell-hashimot... They even used the quote as the title of the accompanying blog post. As I say, I didn’t mean this as a gotcha or anything- I totally agree with the change and I have done similarly. I’ve always disabled autocomplete, tool tips, suggestions etc but now I am actively using Cursor daily.

Yeah understood, I'm not taking it negatively, I just genuinely wanted to understand where it came from. Yeah this is from 2021 (!!!) and is directly related to LSPs. ChatGPT didn't even get launched until Nov 2022. So I think the quote doesn't really work in the context of today, it's literally from an era where I was looking at faster horses when cars were right around the corner and I had not a damn clue. Hah. Off…

Are you still using the NixOS setup you showed in that interview or do you now use the MacOS native ghostty?

Re: Vibing a non-trivial Ghostty feature

#103

Post by tptacek »

stavros wrote: post #101
tptacek wrote: post #100

Earlier quoted context omitted.

I was just talking about this in a chat today, because 'simonw had at some point talked about getting to the point where he was letting go of reviewing every line of LLM code, and I am nowhere close to that point --- I'll take Claude's word on Tailwind classes as long as the HTML looks right, but actual code, I review line-by-line, token-by-token, and usually rewrite things, even if just for cosmetic reasons.

Yeah, there's a definite continuum for where LLMs are most to least expert. They seem to be fairly OK with JS, less so with Python, and C is just a crapshoot. It depends on the project as well, for throwaway things I'm fine to just let it do whatever it wants, but for projects that I need to last more than a few days, I review everything.

A friend of mine has been doing embedded C stuff (making some kind of LCD wall) and has been blown away by how well it's been doing with C --- he went in an LLM skeptic (I've been trying to sell him for months, it finally clicked).

Re: Vibing a non-trivial Ghostty feature

#104

Post by stavros »

tptacek wrote: post #103
stavros wrote: post #101

Earlier quoted context omitted.

Yeah, there's a definite continuum for where LLMs are most to least expert. They seem to be fairly OK with JS, less so with Python, and C is just a crapshoot. It depends on the project as well, for throwaway things I'm fine to just let it do whatever it wants, but for projects that I need to last more than a few days, I review everything.

A friend of mine has been doing embedded C stuff (making some kind of LCD wall) and has been blown away by how well it's been doing with C --- he went in an LLM skeptic (I've been trying to sell him for months, it finally clicked).

Huh, interesting. I wanted to turn an old rotary phone into a meeting headset, and I tried to get Opus to make me a sound card, but $35 in API costs later, I had no sound card.

Re: Vibing a non-trivial Ghostty feature

#105

Post by qudat »

piazz wrote: post #52

Such a useful walkthrough. It looks like Mitchell is using an agentic framework called Amp (I’d never heard of it) - does anybody else here use it or tried it? Curious how it stacks up against Claude Code.

I’m using it. It’s expensive but it’s awesome

Re: Vibing a non-trivial Ghostty feature

#106

Post by qudat »

simonw wrote: post #63
piazz wrote: post #52

Such a useful walkthrough. It looks like Mitchell is using an agentic framework called Amp (I’d never heard of it) - does anybody else here use it or tried it? Curious how it stacks up against Claude Code.

I haven't yet spent any time with it myself, but the impression I have been getting is that it is the most credible of the vendor-independent terminal coding agents right now. Claude Code, Codex CLI and Gemini CLI are all (loosely) locked to their own models.

As far as I know it only uses sonnet 4.5

Re: Vibing a non-trivial Ghostty feature

#107

Post by moderation »

jebarker wrote: post #80
mitchellh wrote: post #76

Earlier quoted context omitted.

Not sure exactly what you're referring to, but I'm guessing it may be this interview I did 2 years ago: https://youtu.be/rysgxl35EGc?t=214 (timestamp linked to LLM-relevant section) I'm genuinely curious because I don't quite remember saying the quote you're saying I did. I'm not denying it, but I'd love to know more of the context. :) But, if it is the interview from 2 years ago, it revolved more around autocomplete…

It was this one: https://sourcegraph.com/blog/dev-tool-time-mitchell-hashimot... They even used the quote as the title of the accompanying blog post. As I say, I didn’t mean this as a gotcha or anything- I totally agree with the change and I have done similarly. I’ve always disabled autocomplete, tool tips, suggestions etc but now I am actively using Cursor daily.

The CEO of Sourcegraph Quinn was pretty negative on coding agents and agentic development only about 10 months ago [0]. He had 'agentic stuff' in the Deader category (Used rarely, Reviewing it aint worth it). In fairness, he did say it was the future but 'is not there yet'. Since then, Sourcregraph's code assistant plugin Cody has been deprecated an they are all in on agents and agentic with Amp.

0.https://youtu.be/Up6WVA07QdE?si=xU_iu2rQAWoHXPpO&t=898

Re: Vibing a non-trivial Ghostty feature

#108

Post by blibble »

tptacek wrote: post #90
roughly wrote: post #73

Earlier quoted context omitted.

I'm not sure that's a fair take. I don't think it's an unfair statement that LLM-generated code typically is not very good - you can work with it and set up enough guard rails and guidance and whatnot that it can start to produce decent code, but out of the box, speed is definitely the selling point. They're basically junior interns. If you consider an engineer's job to be writing code, sure, you could read OP's post…

I can't get past the framing that "people who like the act and craftsmanship" feel AI is negative, which implicitly defines whatever Mitchell Hashimoto is doing as not craftsmanship, which: ghostty is pure craftsmanship (the only reason anyone would spend months writing a new terminal). No, I think my response was fair, if worded sharply. I stand by it.

> I can't get past the framing that "people who like the act and craftsmanship" feel AI is negative

self evident truths have a habit of doing that

Re: Vibing a non-trivial Ghostty feature

#109

Post by bsder »

JasonSage wrote: post #29
tptacek wrote: post #20

Earlier quoted context omitted.

Sounds like a very poorly managed team.

I have to agree. My experience working on a team with mixed levels of seniority and coding experience is that everybody got some increase in productivity and some increase in quality. The ones who spend more time developing their agentic coding as a skillset have gotten much better results. In our team people are also more willing to respond to feedback because nitpicks and requests to restructure/rearchitect are eva…

> My experience working on a team with mixed levels of seniority and coding experience is that everybody got some increase in productivity and some increase in quality.

Is that true? There have been a couple of papers that show that people have the perception that they are more productive because the AI feels like motion (you're "stuck" less often) when in reality it has been a net negative.

Re: Vibing a non-trivial Ghostty feature

#110

Post by yggdrasil_ai »

XenophileJKO wrote: post #26
j45 wrote: post #21

Earlier quoted context omitted.

Maybe it's possible to use AI to help review the PRs and claim it's the AI making the PR's hyperproductive?

Yes, this. If you can describe why it is slop, an AI can probably identify the underlying issues automatically. Done right you should get mostly reasonable code out of the "execution focused peer".

> If you can describe why it is slop, an AI can probably identify the underlying issues automatically

I would argue against this. Most of the time the things we find in review are due to extra considerations, often business, architectural etc, things which the AI doesn't have context of and it is quite bothersome to provide this context.

Post reply on HN