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

Google Antigravity

antigravity.google

Post reply on HN
Print view View original

421–430 of 1001 posts

Re: Google Antigravity

#421

Post by kccqzy »

modeless wrote: post #78
nateb2022 wrote: post #43

I went ahead and downloaded it, it looks to be a VSCode fork very similar to Cursor, with support for the following models: - Gemini 3 Pro (High) - Gemini 3 Pro (Low) - Claude Sonnet 4.5 - Claude Sonnet 4.5 (Thinking) - GPT-OSS 120B (Medium)

Thank you for saying what this entire blog post doesn't. It's actually disrespectful of Google to launch this without even a mention of the fact that it is based on VSCode.

Why credit? Come on, the world has moved on from 1990s-era 4-clause BSD licenses. If you recall, the 4-clause BSD license states that all advertising materials must display an acknowledgement. It’s widely considered to be a mistake and nobody uses this license any more. Not because of legal reasons (incompatibility with GPL) but because it is madness to require so many acknowledgements. Stallman was right.

Re: Google Antigravity

#422

Post by Maxatar »

modeless wrote: post #78
nateb2022 wrote: post #43

I went ahead and downloaded it, it looks to be a VSCode fork very similar to Cursor, with support for the following models: - Gemini 3 Pro (High) - Gemini 3 Pro (Low) - Claude Sonnet 4.5 - Claude Sonnet 4.5 (Thinking) - GPT-OSS 120B (Medium)

Thank you for saying what this entire blog post doesn't. It's actually disrespectful of Google to launch this without even a mention of the fact that it is based on VSCode.

And VSCode is based on Chromium (a fact you won't find on VS Code's website other than related to updates).

Re: Google Antigravity

#423

Post by Towaway69 »

What I don't get is why they didn't fork emacs, then build a VScode mode and then add AI to that.

I guess it must have been the GPL which isn’t compatible with their AI agents.

Oh, wait I was meant to take this announcement seriously?

Re: Google Antigravity

#424

Post by zahlman »

nova22033 wrote: post #250
jihadjihad wrote: post #51

> Spin up agents to tackle routine tasks that take you out of your flow, such as codebase research, bug fixes, and backlog tasks. The software of the future, where nobody on staff knows how anything is built, no one understands why anything breaks, and cruft multiplies exponentially. But at least we're not taken out of our flow!

The software of the future, where nobody on staff knows how anything is built Doesn't this apply to people who code in high level languages?

In the 00s I saw so many C codebases with hand-rolled linked lists where dynamically resized arrays would be more appropriate, "should be big enough" static allocations with no real idea of how to determine that size, etc. Hardly anyone seemed to have a practical understanding of hashes. When you use a higher level language, you get steered towards the practical, fundamental data structures more or less automatically.

Re: Google Antigravity

#426

Post by OtherShrezzing »

bradfa wrote: post #415
MangoToupe wrote: post #381

Earlier quoted context omitted.

Also, agents have no capacity to learn.

They have a capacity to "learn", it's just WAY MORE INVOLVED than how humans learn. With a human, you give them feedback or advice and generally by the 2nd or 3rd time the same kind of thing happens they can figure it out and improve. With an LLM, you have to specifically setup a convoluted (and potentially financially and electrical power expensive) system in order to provide MANY MORE examples of how to improve via…

I think it’s reasonable to say that different approaches to learning is some kind of spectrum, but that contemporary fine tuning isn’t on that spectrum at all.

Re: Google Antigravity

#427

Post by throwawaysleep »

uatec wrote: post #378

"Congratulations, you have been elevated to manager to agents." That's not exactly really where I hoped my career would lead. It's like managing junior developers, but without having nice people to work with.

As an introvert, that's a pro, not a con.

The burden of human interaction is removed from building.

Re: Google Antigravity

#428

Post by ethmarks »

bradfa wrote: post #415
MangoToupe wrote: post #381

Earlier quoted context omitted.

Also, agents have no capacity to learn.

They have a capacity to "learn", it's just WAY MORE INVOLVED than how humans learn. With a human, you give them feedback or advice and generally by the 2nd or 3rd time the same kind of thing happens they can figure it out and improve. With an LLM, you have to specifically setup a convoluted (and potentially financially and electrical power expensive) system in order to provide MANY MORE examples of how to improve via…

Depending on your definition of "learn", you can also use something akin to ChatGPT's Memory feature. When you teach it something, just have it take notes on how to do that thing and include its notes in the system prompt for next time. Much cheaper than fine-tuning. But still obviously far less efficient and effective than human learning.

Re: Google Antigravity

#429

Post by siliconc0w »

Trip report: I'm using it now to revamp a dashboard I'm working on. TBH it's not feeling much better than Codex - it couldn't figure out how to launch chrome with my default profile nor how to regenerate the css with tailwind. I'm also getting a lot of model quota errors like everyone else.

Re: Google Antigravity

#430

Post by 999900000999 »

Oh lucky, another free AI beta.

I'm going to treat this like Kiro, and just use it until they start charging for it and then probably switch back to VS code with its built-in agent support.

Eventually they're going to do a rug pull, and instead of paying $10 a month for tons of AI code request, it's going to be two or $300 for that. The economics just aren't there to actually make a profit, hopefully before the rug pool happens local models on normal hardware will be fast enough.

Post reply on HN