parasti

This got a laugh out of me. I guess I practiced vim commands in the most boring way - by using vim.

I had to stop reading the linked page once I realized it is melting my brain because it is repeating the same three/four one-liners in different formats.

Right, there is hardly any reason to start from "install this toolchain and these headers" when you can literally get a GLES-equivalent context with one line of JS in any browser from the past 10+ years.

TrenchBroom is great. I am the maintainer for Neverball (which is incidentally one of the games that does still use the old style texture projection). TrenchBroom was the first Quake editor that already had a functional Neverball gamepack without me having to do a thing.

To be fair Github refuses to render diffs after a couple of hundred lines.

Would not want to stand near this.

Weird take, considering it answered correctly despite disagreeing with consensus at the time.

Incidentally, this morning I asked Gemini Argon 4 on High to do deep research on the effect of the strikes on russian bot farms and after some hours (it got stuck) it generated the report starting with a surprising (to me at the time) paragraph:

"Rigorous digital forensic scrutiny and network telemetry analysis, however, demonstrate that the purported collapse of Russian social media propaganda bots is an analytical mirage born of misinterpreted network traffic data and fabricated viral graphics."

What were you doing if you can share.

Last I looked at it Safari on iOS couldn't even do element.requestFullscreen(), so no, thanks.

Knowledge cutoff isn't really what these models are used for.

This being the top comment posted mere 30 minutes after the article itself tells me the cynicism is not overwhelming.

Coolest thing I've read in months.

Only includes water in one type of meal?

They fixed it!

I spent five years on and off porting Neverball to the web by hand and waited for the perfect moment to announce it while tweaking the UX.

Fucking missed it.

https://play.neverball.org/

There is auto compaction at 250k? Having used agy for months with multi day sessions, I have never seen this.

And yet, there's been a glaring overflow bug on every repo page if the repo has a sponsor button on Firefox Android for months.

"doing dumb things" and "stupid shit" is an odd choice to describe tools that only trigger on explicit activation. Is a windshield being lowered by a switch being held a "dumb thing"?

Dumb things start to happen when you try to build Star Trek interfaces. When you build DWIM interfaces in real life, they are annoying and trigger unwanted and the implementation is without exception, by necessity, a growing ball of spaghetti.

At work, we use Claude Code with superpowers which organizes most work into planning followed by review followed by implementation and superpowers applies a fairly high level of scrutiny at every step. It's really quite effective, I don't generally read the code that comes out. In that workflow I have never used plan mode. I don't know what I'd use it for.

On my hobby projects I use Antigravity CLI and almost always start with /plan. This is a builtin skill - not a mode. It generates a plan artifact that I review and approve. Once approved the implementation speed is uncanny compared to Claude.

I tried using superpowers with Antigravity CLI but that slowed the agent down considerably without much benefit (plan quality was much worse).

> The people who love being connected never let the thing leave them be it pocket, palm, face-down next to the fork

I'm tired of convincingly presented falsehoods. I hate being on my phone. I also can't stop being on my phone.

I stopped reading when I realized the article reminds me of my own Claude-generated solutions at work - just an endless maze of special business logic on top of special business logic. You need an LLM to understand it. You need an LLM help write the documentation. You need an LLM to help read the documentation.

It's like that time when they "discovered" that babies can feel pain.

Similarities between languages is a much more intuitive approach for non-English speakers. This comment just explains an English idiom in terms of English idioms.

I'm on AI Pro and have had 3.8 since announcement day, both in the Android app and in Antigravity CLI.

Why even keep using zombie Twitter? I used to scroll Twitter a lot but deleted my account back when. Literally any other social network seems more appealing these days, and not by a small margin.

I just had to go check out your site with that comment. Nicely done! Was waiting for the physics simulation to either fully settle or explode, but really neither of those happened.

The blog post says that their gains come largely from the model trying harder. So, more tokens, more time spent.

I've done this. Ask Claude to permutate a prompt and run Claude Code or a subagent to observe effects until success. Claude will iterate on the prompt faster than I could and likely catch more edge cases, too. It's one of those tasks where the end goal is very clear and the agent just needs to iterate on permutations. Arguably this is a perfect match, too, because prompt engineering is really just text generation, so why not have a text generation machine do it.

Gemini is not bad at coding? Give it a /plan and it does a very decent job in a fraction of the time/tokens that Claude needs.