[HN Gopher] Google Antigravity
       ___________________________________________________________________
        
       Google Antigravity
        
       https://antigravity.google/blog/introducing-google-antigravi...
       https://www.youtube.com/watch?v=nTOVIGsqCuY
        
       Author : Fysi
       Score  : 573 points
       Date   : 2025-11-18 15:47 UTC (7 hours ago)
        
 (HTM) web link (antigravity.google)
 (TXT) w3m dump (antigravity.google)
        
       | Fysi wrote:
       | Looks to be live but no content; OpenGraph description is "Google
       | Antigravity - Build the new way".
        
       | meetpateltech wrote:
       | > Google Antigravity is an agentic development platform, evolving
       | the IDE into the agent-first era.
       | 
       | Antigravity enables developers to operate at a higher, task-
       | oriented level by managing agents across workspaces, while
       | retaining a familiar AI IDE experience at its core. Agents
       | operate across the editor, terminal, and browser, enabling them
       | to autonomously plan and execute complex, end-to-end tasks
       | elevating all aspects of software development.
       | 
       | via: https://www.linkedin.com/showcase/google-antigravity/about/
        
         | 20k wrote:
         | I have absolutely 0 idea why any developer would rely on any
         | IDE produced by google. It'll be canned within 5 years max,
         | with 3-4 seeming like a reasonable estimate of the lifespan of
         | the product
         | 
         | I've been using my current IDE for 17 years, and plan to
         | continue using it for at least another 15
        
           | whs wrote:
           | You mean Android Studio will be canned in 2018 max with a
           | reasonable estimate of 2016-2017?
        
             | devsda wrote:
             | Isn't Android Studio based on IntelliJ and not a product
             | developed from ground up? And Android Studio has second
             | order revenue from the playstore.
             | 
             | I wouldn't be even surprised if internally the AS team's
             | financials are counted under the Playstore umbrella.
        
               | zamadatix wrote:
               | Antigravity is based on VS Code, not designed from the
               | ground up, and has second order revenue from the AI
               | subscriptions (financials probably counted under the AI
               | umbrella).
               | 
               | I still wouldn't trust a Google product to stick around,
               | but these hints aren't a reliable oracle either.
        
               | devsda wrote:
               | When AS was launched Android was the only other viable
               | option and it is the same even today. I don't believe
               | Google's AI products will reach and/or sustain the same
               | dominance as Android.
               | 
               | It is a product launched in the hype cycle of AI. Google
               | has plenty of other products (launched during hype
               | cycles) that are gathering dust.
               | 
               | That's not a guaranteed signal that it will meet the same
               | fate but its something strong enough to be wary of.
        
               | deeringc wrote:
               | It's interesting to think that Google's Antigravity is a
               | forked version of MSFT's VS Code, which uses a browser
               | engine built by Google, which they forked from Apple,
               | which they forked from KHTML.
        
             | Ygg2 wrote:
             | It's made by Jetbrains thankfully.
        
           | zevv wrote:
           | Which is is, vi or emacs?
        
             | Arcuru wrote:
             | It will be very funny if it's vim, since Bram Moolenaar who
             | created and ran it worked at Google from 2006 to 2021.
        
             | 20k wrote:
             | codeblocks. There are dozens of us!
        
           | CjHuber wrote:
           | Do you really think antigravity will last longer than 1 year?
        
             | NewsaHackO wrote:
             | Yes
        
           | Yeroc wrote:
           | It won't matter. The core ideas of an Agent Manager view will
           | be copied and improved by others in many project in the
           | future.
        
       | denysvitali wrote:
       | See the videos:
       | 
       | https://www.youtube.com/watch?v=YX-OpeNZYI4
       | 
       | https://www.youtube.com/watch?v=rKQ9b4UMpGQ
       | 
       | https://news.ycombinator.com/item?id=45967787
        
         | Barbing wrote:
         | Wonder why they're unlisted.
        
           | xnx wrote:
           | This is common for videos embedded in a page that don't make
           | sense out of context (e.g. in the YouTube feed).
        
       | phplovesong wrote:
       | From what i saw its yet an AI first text editor. Thats a hard
       | pass for me.
        
       | nasretdinov wrote:
       | - A new "AI" IDE announced
       | 
       | - It's VS Code
       | 
       | Like clockwork!
        
         | jakebasile wrote:
         | How many forks of VS Code am I supposed to have installed at
         | this point?
        
           | 01HNNWZ0MV43FF wrote:
           | My YC2026 startups are an AI agent that automatically manages
           | your VSCode forks, and a public safety computer vision app
           | for smart glasses that predicts whether someone can afford a
           | lawyer based on their skin color
        
             | collingreen wrote:
             | They'll have to compete with these other pending funded
             | companies:
             | 
             | - ai therapist for your ai agents
             | 
             | - genetic ai agent container orchestration; only the best
             | results survive so they have to fight for their virtual
             | lives
             | 
             | - prompt engineering as a service
             | 
             | - social media post generator so you can post "what if ai
             | fundamentally changes everything about how people interact
             | with software" think pieces even faster
             | 
             | - vector db but actually json for some reason
             | 
             | (Edit: formatting)
        
               | whstl wrote:
               | Don't forget there's 2 or 3 of each in YC2026.
        
               | bblb wrote:
               | > genetic ai agent container orchestration; only the best
               | results survive so they have to fight for their virtual
               | lives
               | 
               | AI Agent Orchestration Battle Bots. Old school VMs are
               | the brute tanks just slowly ramming everybody off the
               | field. A swarm of erratically behaving lightweight K8s
               | Pods madly slicing and dicing everything coming on their
               | way. Winner takes control of the host capacity.
               | 
               | I might need this in my life.
        
           | zamadatix wrote:
           | You'll need a Chromium based app to count the installs for
           | you.
        
         | linhns wrote:
         | This is why I have much respect for the Zed team as they are
         | chasing originality, not just slap something onto VS Code and
         | call it a new IDE.
        
           | Demiurge wrote:
           | I love Zed code. Sad that Sublime Text is not keeping up.
        
             | the__alchemist wrote:
             | These are two of the fastest editors, and I use both, but I
             | don't think they're for the same thing: Zed is for multi-
             | file projects with moderate IDE abstractions (Worse than
             | Jetbrains; better than most others); Sublime is for editing
             | one-off files with syntax highlighting.
        
             | kayart_dev wrote:
             | On the contrary, I am glad that the Sublime team did not
             | fall for the trends and avoided adding AI slop into the
             | editor
        
             | dingdingdang wrote:
             | Once the UI soup around AI dev use has settled (and it's
             | getting closer) I bet you we will see native apps with
             | c/c++/zig/rust backends that render so much faster on all
             | the junctions that aren't roundtrip limited (and yes, that
             | will still matter to many people).
        
           | mbesto wrote:
           | If both work equally well then why does it matter whether Zed
           | or VSCode is under the hood?
        
         | IncreasePosts wrote:
         | Is the extension system in VSCode not powerful enough to make
         | these just normal extensions for a vanilla VSCode executable?
         | Or is everyone just going for lock in, since if you download
         | MyFork, you can't start using some other extension that uses
         | OtherGuysModel?
        
           | threetonesun wrote:
           | Correct. One could say the same thing for browsers. I suppose
           | on the one hand it's good that it's relatively easy to spin
           | up a new project like this, on the other hand one must swear
           | allegiance to their large software company of choice.
        
           | jamie_ca wrote:
           | Back when Cursor was new (before literally everything was AI
           | hype) they explicitly called out that they wanted to do more
           | in-depth integration with the editor than was possible with
           | just the extension APIs.
           | 
           | Presumably that hasn't changed much. If you want to do any
           | large-scale edits of the UI you need to spin up a fork.
        
         | mpeg wrote:
         | It's also an absolutely basic fork. They haven't even bothered
         | with a custom theme, or custom UI, it's just vscode with an
         | agents window slapped on top.
         | 
         | Weirdly, out of all the vscode forks the best UI is probably
         | bytedance's TRAE
        
         | asadm wrote:
         | > VSCode
         | 
         | You mean Chromium wrapper?
        
           | ekropotin wrote:
           | Chromium? You mean OS wrapper?
        
       | m-hodges wrote:
       | Was expecting https://pypi.org/project/antigravity/
        
       | bastawhiz wrote:
       | Nice, if I switch now it'll be killed in two to three years right
       | around the time Zed has all the features that I want!
        
       | guluarte wrote:
       | and the difference from vscode is...?
        
         | surgical_fire wrote:
         | VS Code won't feature in Google Graveyard in the short term
         | future.
        
         | pharrington wrote:
         | This one forces you to log into your google account before you
         | can use vscode!
        
       | jasonjmcghee wrote:
       | Curious if the name is a reference to https://xkcd.com/353/
       | 
       | > Come join us! Programming is fun again! It's a whole new world
       | up here!
        
         | benatkin wrote:
         | The name Bard didn't fly, so they went for literal flying
         | instead.
        
         | Thrymr wrote:
         | I hope Google is at least acknowledging the origin of the name,
         | even if they are not paying royalties to Randall Monroe.
        
           | jasonjmcghee wrote:
           | Don't take my random pondering as any kind of evidence
        
       | TIPSIO wrote:
       | I actually like the workflow they are suggesting. There's
       | something there for sure:
       | 
       | - Nano Banana => Mockup
       | 
       | - Antigravity/IDE => Comments/note
       | 
       | - Gemini => Turn to code
       | 
       | - Antigravity/IDE => Adjust/code
       | 
       | All on the same platform so can maximum automate / "agentic"
        
         | dnw wrote:
         | Jules
        
       | gnarlouse wrote:
       | So the whole world is a scam for your data now basically.
        
         | Oarch wrote:
         | Neigh, a Trojan Horse.
        
           | collingreen wrote:
           | This pun made me actually laugh out loud. I almost lost some
           | coffee.
        
           | gnarlouse wrote:
           | Something something, "my kingdom for a horse." Obligatory
           | upvote, wp
        
       | monegator wrote:
       | Wow this page is an endless source of memes and broken UX madness
       | 
       | Google at its finest
        
       | spuz wrote:
       | > Your new focus is architecting the solution, not implementing
       | every single step. So congratulations, you have been elevated to
       | a manager of agents.
       | 
       | I'm not sure many engineers will welcome this "promotion".
        
         | elif wrote:
         | Few horse racers became automobile racers.
         | 
         | If existing engineers don't change it doesn't matter because
         | new engineers will take their place.
        
           | vosper wrote:
           | Horse racing didn't go away and there are more people who
           | race horses professionally than who race cars.
        
             | mxkopy wrote:
             | Horses also run faster than pictures of cars
        
               | augment_me wrote:
               | Copium
        
             | bad_haircut72 wrote:
             | There are many more truck drivers than buggy drivers
        
               | zem wrote:
               | there is a lot more buggy code than truck code
        
               | Dilettante_ wrote:
               | Truckers code better than bugs
        
             | elif wrote:
             | "Professional riders number roughly three to six thousand
             | worldwide, while professional drivers number roughly twenty
             | to forty thousand across major sanctioned series."
        
           | croes wrote:
           | We'll wait and see.
           | 
           | Car manufacturers made profit
        
             | NewsaHackO wrote:
             | Some will wait and see, yes.
        
         | AstroBen wrote:
         | I don't think this is speaking to the engineers
        
           | spuz wrote:
           | Then who are they targeting? Who else would currently be
           | "implementing every single step"?
        
             | sp4cec0wb0y wrote:
             | Project managers and higher level management.
        
               | wduquette wrote:
               | Reminds me of the pre-GitHub days, when I had to use CM
               | tools designed to appeal to project and CM managers, not
               | to the poor developers who had to use them every day.
               | Anybody else remember Harvest?
        
               | orphea wrote:
               | Yeah, we'll see how that'll go.
        
         | salawat wrote:
         | You weren't the target audience. The target audience was
         | manager types tired of being told no by engineers. Always
         | listen to the quiet parts left unspoken/unacknoeledged.
        
           | pyrale wrote:
           | They will equally be tired of being told yes by LLMs.
        
           | ihumanable wrote:
           | It's the same thing they tried to sell with low/no-code.
           | 
           | The problem is that the engineer turning what you want into
           | code isn't normally the bottleneck. I would say about 50% of
           | my job is helping people specify what they want sufficiently
           | for someone to implement.
           | 
           | Non-technical people are used to a world of squishy
           | definition where you can tell someone to do something and
           | they will fill in the blanks and it all works out fine.
           | 
           | The problem with successful software is that the users are
           | going to do all the weird things. All the things the manager
           | didn't think about when they were dreaming up their happy
           | path. They are going to try to update the startTime to the
           | past, or to next year and then back to next week. They are
           | going to get their account into some weird state and click
           | the button you didn't think they could. And this is just the
           | users that are trying to use the site without trying to
           | intentionally break it.
           | 
           | I think if managers try to LLM up their dreams it'll go about
           | as well as low/no-code. They will probably be able to get a
           | bit further because the LLM will be willing to bolt on
           | feature after feature and bug fix after bug fix until they
           | realize they've just been piling up bandaids.
           | 
           | I am cautiously optimistic that there will be a thriving
           | market for skilled engineers to come in and fix these things.
        
         | lo_zamoyski wrote:
         | Perhaps it's worth posing the question: what sorts of
         | "engineers" might feel threatened by agents? Those doing
         | engineering, or those who spend their careers wading in the
         | shallows? Competent designers with deep comprehension, or, at
         | best, the superficial pedants?
        
       | prodigycorp wrote:
       | Why is google so bad at product branding and strategy? My
       | complaint is aesthetic: why would you name your product a five-
       | syllable word??
        
         | hobs wrote:
         | Its clearly a reference to the xkcd comic which does have
         | mindshare.
        
           | nfw2 wrote:
           | Thank you for the demonstrating the reasoning that leads to
           | these decisions.
        
             | hobs wrote:
             | This made me laugh, I assume you are calling me an old
             | fogey and I will be glad to take it. This is why they don't
             | let me near the marketing stuff.
        
           | paganel wrote:
           | Which comic would that be?
           | 
           | Later edit: Probably this one [1], which is par for the
           | course for Alphabet, they're, conceptually, still living in
           | the early 2010s, when this stuff was culturally relevant.
           | 
           | [1] https://xkcd.com/353/
        
             | igleria wrote:
             | well they sure seem to make product naming after going
             | through the whole medicine cabinet too, just like the comic
        
         | shwaj wrote:
         | Informally, people will way "antigrav".
        
         | xplt wrote:
         | Because it's awkwardly close to the letters AGI, maybe
        
           | RubberSpoon wrote:
           | It's actually this, the command line tool is `agy`
        
       | roman_soldier wrote:
       | Another google product, there are too many and which ones will be
       | around in a year or two?
        
       | ForHackernews wrote:
       | Loads a blank white page and breaks the back button in Firefox.
       | 
       | Console error:
       | 
       | > Loading module from
       | "https://antigravity.google/main-74LQFSAF.js" was blocked because
       | of a disallowed MIME type ("text/html").
        
       | Slogsworth wrote:
       | I recognize the guys in the video, they were in marketing videos
       | for the Windsurf IDE before its founding team was cannibalized
       | by/absorbed into Google.
        
         | arrowleaf wrote:
         | Kevin was CTO / head of product engineering at Windsurf, Anshul
         | was a founding engineer
        
         | 6thbit wrote:
         | I found the demo videos off-putting, people come off a bit
         | smug, not sure if intended.
        
       | phyzome wrote:
       | It's really kind of pathetic how we live in a future where
       | "antigravity" is a text editor that lies to you, "hoverboards"
       | are one-wheeled electric skateboards that burn your house down,
       | and... well, can't think of a third thing at the moment, but you
       | know the vibe.
       | 
       | Lotta people mining science fiction for cool names and then
       | applying them to their crappy products, cheapening the source
       | ideas.
        
         | seanhunter wrote:
         | A third thong could be "self driving cars" are cars that you
         | have to stay alert and in full control of at all times.
         | 
         | We _are_ in the future, it's just a much more rubbish version
         | than people imagined in scifi
        
       | CuriouslyC wrote:
       | Neat, but the world doesn't need another IDE, and people want
       | choice. Provide tools that plug into open workflows and step
       | back.
        
       | craftkiller wrote:
       | I don't get it. It's a completely blank web page. Did they not
       | test in firefox?
       | 
       | Ah Google misconfigured their web server:
       | 
       | > Loading module from
       | "https://antigravity.google/main-74LQFSAF.js" was blocked because
       | of a disallowed MIME type ("text/html").
       | 
       | Edit: And a couple minutes later, it is now working. Guess Google
       | is reading HN.
        
         | jkrems wrote:
         | Looks like it's back again!
        
         | mentalgear wrote:
         | Honestly, the full page doesn't give you much more. Not a
         | SINGLE product image. All paragraphs about "agentic" blah-blah
         | you have read 100s of times by now - I do not see how this is
         | anything different from all the other AI VS Code forks, besides
         | that it comes with Gemini from the start.
        
           | thekevan wrote:
           | "Not a SINGLE product image."
           | 
           | But there is a 13 minute demo video.
           | 
           | https://antigravity.google/product
        
           | halflife wrote:
           | Seems like they jumped the gun on the website release, the
           | first version I saw was a wall of text, now it has photos and
           | videos. Maybe they have an agentic CICD
        
           | FuriouslyAdrift wrote:
           | They're blitzing all media with a Gemini 3 push, it seems.
        
         | yawnr wrote:
         | Can't even scroll on Firefox mobile lol
        
       | dehugger wrote:
       | Nice to see that it's not locked to just Gemini models
        
       | dehugger wrote:
       | Nice that you can use non-Gemini models with it
        
       | andrewk17 wrote:
       | can't get past the "Setting up your account" step atm
        
         | rco8786 wrote:
         | same just hanging on the spinner
        
         | peterldowns wrote:
         | same here, hope they fix this soon
        
         | lefstathiou wrote:
         | Same. I wonder if it is because I'm signing in with my Google
         | Apps for business account.
        
       | foofoo12 wrote:
       | I worked in a factory one summer when I was a teenager. It was a
       | totally brain dead work, but the morale was good. The workers
       | weren't unhappy.
       | 
       | I'm concerned that the new role of "manager of agents" (as google
       | puts it) will be a soul destroying brain dead work and the morale
       | won't be good.
        
       | oytis wrote:
       | Haven't we got enough of new eras yet?
        
         | klysm wrote:
         | No the investment amount demands a new era per week
        
         | coffeebeqn wrote:
         | Each VSCode fork with some random AI junk slapped on will be a
         | new era! I can't imagine how many eras behind I'm at this point
        
       | ayhanfuat wrote:
       | On the pricing page it says that for public preview they are
       | offering a free individual plan with "generous rate limits". I
       | gave it an HTML file and asked it to create Jinja templates from
       | it and 2 minutes later (still planning, no additional prompt) I
       | got this:
       | 
       | > Model quota limit exceeded. You have reached the quota limit
       | for this model.
        
         | riskassessment wrote:
         | > html
         | 
         | Would be willing to bet this is the issue. Adding html files to
         | context for gemini models results in a ton of token use.
        
           | gcr wrote:
           | why?
           | 
           | EDIT: why must users care?
        
             | SPICLK2 wrote:
             | https://stackoverflow.com/questions/1732348/regex-match-
             | open...
        
               | croes wrote:
               | The accepted answer is one that doesn't care about the
               | questioner's use case and instead gives a pretty
               | excessive "Don't do it"
        
               | lukan wrote:
               | It does also give the right solution, using an xml
               | parser.
        
               | croes wrote:
               | We don't know the use case.
               | 
               | Maybe the questioner is also in full control of the HTML
               | creation and they don't need a parser for all possible
               | HTML edge cases.
        
             | kulahan wrote:
             | Gotta learn all the quirks of the model before it's
             | replaced in 8 minutes.
        
               | NaomiLehman wrote:
               | Quirks? like context window?
        
               | kulahan wrote:
               | I'm saying it's egregious to expect all users to know the
               | fact that an HTML document, for some reason, uses an
               | enormous amount of context in an LLM designed
               | specifically for _working with code_.
        
         | malshe wrote:
         | I had pretty much the same experience.
        
         | egeozcan wrote:
         | That'd be my experience with gemini cli.
        
         | leoff wrote:
         | I think the models are under high load right now, and not
         | working properly.
        
         | agotterer wrote:
         | Same here. I tried to build a super simple iOS App in
         | antigravity and I was out of quota before it finished. The
         | whole thing was a couple of files and a few hundred lines of
         | code.
        
         | dabockster wrote:
         | Jevons paradox - https://en.wikipedia.org/wiki/Jevons_paradox
         | 
         | It's the same problem with OpenRouter's free tiers for a long
         | time. If something is truly $0 and widely available, people
         | will absolutely bleed it dry.
        
       | nateb2022 wrote:
       | 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)
        
         | koakuma-chan wrote:
         | Oh no, not another VSCode fork...
        
           | rvz wrote:
           | Many software engineering principles have been broken.
           | 
           | Google using Electron tells us that quality control is
           | completely out of the window.
           | 
           | Unbelievable.
        
             | user34283 wrote:
             | I tried writing a native Windows app using WinUI 3.
             | 
             | I wasted a day on trying to get some PNGs to render
             | correctly, but no matter the config I used, the colors came
             | out wrongly oversaturated.
             | 
             | I used Tauri with a WebView, and the app was rendering the
             | images perfectly fine. On top of that the UI looked much
             | better, and I was done in half the time I spend trying to
             | fix the rendering issue in WinUI 3.
             | 
             | Never again will I go native.
        
               | sirwhinesalot wrote:
               | To be fair WinUI3 is particularly bad. If you're forced
               | to install a 100MB runtime anyway, might as well make
               | that runtime be a browser.
        
               | user34283 wrote:
               | Since Tauri uses the system's WebView, it shouldn't be
               | that big. The docs say a minimal app could be 600 KB.
        
             | morshu9001 wrote:
             | Native app dev is covered in red tape and puts you at the
             | mercy of Apple etc. It's unfortunate that things are so
             | inefficient now, but competition is good, and native
             | platforms can get good.
        
               | zahlman wrote:
               | > Native app dev is covered in red tape and puts you at
               | the mercy of Apple etc.
               | 
               | ... Aren't we talking about a _programming IDE_ here?
               | When did mobile become anything like the primary market
               | for that? Are people expected to sit around for hours
               | inputting symbols with an OSK?
        
               | timeon wrote:
               | If Zed can make it, even with completely new stack, why
               | not Google?
               | 
               | What are all this LLMs for when everything is just fork
               | of electron app? Does not look like good marketing.
        
           | badgersnake wrote:
           | Looks like they managed to make it Mac only in the process.
        
             | koakuma-chan wrote:
             | I would love this to be true to further make fun of
             | Electron, but I don't think it's Mac only.
             | 
             | https://antigravity.google/download
        
               | badgersnake wrote:
               | Ahh you're right. The UX of that site is bad on my phone.
        
         | modeless wrote:
         | 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.
        
           | shrikant wrote:
           | It's so obvious from even just the vague screenshots that are
           | hidden somewhere on the site that it's a VSCode fork, that I
           | suppose I can see why they've tried to obfuscate that as much
           | as possible.
        
           | echelon wrote:
           | Google is going to win AI and kill all the other market
           | participants.
           | 
           | They have the revenues to support all of this.
           | 
           | They spent time learning from all the players and can now
           | fast follow into every market. Now they're fast and nimble
           | and are willing to clone other products wholesale, fork
           | VSCode, etc.
           | 
           | They're developing all of this, meanwhile Pichai is calling
           | it a "bubble" to put a chill on funding (read: competition).
           | It's not like Google is slowing down.
           | 
           | We had a chance to break them up with regulation, and we
           | didn't. Now they're going to kill every market participant.
           | 
           | This isn't healthy. We have an invasive species in the
           | ecology eating up all the diverse, healthy species.
           | 
           | a16z and YC must hate this. It puts a cap on their returns.
           | 
           | As engineers, you should certainly hate this. Google does
           | everything it can to push wages down. Layoffs, offshoring,
           | colluding with competitors. Fewer startups mean fewer rewards
           | for innovation capital and more accrual to the conglomerate
           | taxing the entire internet.
           | 
           | Chrome, Android, Search, Ads, YouTube, Cloud, Workspace,
           | Other Bets, and AI/Deepmind need to be split into separate
           | companies.
           | 
           | Call or email your legislators and ask for antitrust
           | enforcement: https://pluralpolicy.com/find-your-legislator/
           | 
           | Demand a Google breakup.
        
             | tayo42 wrote:
             | >They spent time learning from all the players and can now
             | fast follow into every market.
             | 
             | Google has never successfully done that? Maybe once?
        
             | poemxo wrote:
             | For the wage suppression thing, Google pays their engineers
             | better than say Amazon or Microsoft.
        
               | soperj wrote:
               | He was talking about when Apple(under Steve Jobs)
               | colluded with everyone (including Google), but doesn't
               | seem to want to name Apple specifically.
        
             | ryandvm wrote:
             | Has Google *ever* successfully entered and taken over an
             | existing market?
        
               | QuercusMax wrote:
               | Search, back in the late '90s
        
               | darkmarmot wrote:
               | Besides search, Android kinda killed Nokia and friends
               | for the consumer phone market.
        
               | wiseowise wrote:
               | iPhone did it, not Android.
        
               | abirch wrote:
               | To be fair, Microsoft killed Nokia.
        
               | mksreddy wrote:
               | Search, browser, Mobile.
        
               | Aspos wrote:
               | maps as well. email client in the browser.
        
               | cpa wrote:
               | Chrome?
        
               | ghc wrote:
               | You mean like web search, webmail, internet ads, maps,
               | calendars, browsers, smartphone operating systems, online
               | document editing, and translation? I mean, I'm not even
               | including stuff they acquired early like YouTube. Google
               | was the most feared company for a decade or more for a
               | good reason: they absolutely devoured competition in what
               | were thought to be mature markets.
        
               | rvnx wrote:
               | Do you have a source for that ? Because according to
               | official papers submitted by Google to the court this is
               | absolutely not the case
        
               | Towaway69 wrote:
               | There are official court documents where google disowns
               | google search? Or for what do you need a source?
               | 
               | Google spreadsheet was another amazing product back in
               | the day.
        
               | rvnx wrote:
               | Mh ?
               | 
               | https://static01.nyt.com/newsgraphics/documenttools/f6ab5
               | c36...
               | 
               | > Plaintiffs maintain that Google has monopoly power in
               | the product market for general search services in the
               | United States.
               | 
               | > According to Plaintiffs, Google has a dominant and
               | durable share in that market (general search), and that
               | share is protected by high barriers to entry.
               | 
               | > Google counters that there is no such thing as a
               | product market for general search services.
               | 
               | > What exists instead, Google insists, is a broader
               | market for query response.
               | 
               | (+ yes obviously, products like Sheets or Maps were
               | amazing, and are still very much the best.
               | 
               | It was a joke to say that even Google denies its own
               | success, the same way as the earlier comment).
        
               | fl0ki wrote:
               | Putting aside that several of these were acquisitions,
               | these are all great examples of things where Google
               | introduced something for free because it would make the
               | money through advertising, both directly and through
               | ecosystem effects. Even the paid enterprise versions of
               | these services were a tiny % of Google's overall gross
               | revenue.
               | 
               | Prior to the push into Cloud computing, Ad revenue was
               | well over 90% of all Google gross income, and Cloud was
               | the first big way they diversified. GCP is definitely a
               | credible competitor these days, but it did not devour
               | AWS. Other commercial Google services didn't even become
               | credible competitors, e.g. Google Stadia was a
               | technically exceptional platform that got nowhere with
               | customers.
               | 
               | The question now is whether Google carves out an edge in
               | AI that makes it profitable overall, directly or
               | strategically. Like many companies, there seems to be a
               | presumption of potentially infinite upside, which is what
               | it would take to justify the astronomical costs.
        
               | ghc wrote:
               | I'm not sure you can call Docs (Writely) and Android
               | acquisitions though. Android was an OS for cameras and
               | Writely was an experimental rich text editor, not a word
               | processor.
               | 
               | It's not like Youtube where they legitimately bought
               | their way to dominance. And I'd argue that even in the
               | case of DoubleClick, google was already dominating the
               | search advertising market when they bought DoubleClick to
               | consolidate their dominance.
        
               | gcr wrote:
               | Google's definitively taken over search, ads, e-mail,
               | maps, office (for small/medium business).
               | 
               | They're struggling on cloud, AI, ISP, videoconferencing,
               | and others...
        
               | hirako2000 wrote:
               | They clearly don't want to fold on AI.
        
             | Fairburn wrote:
             | You ok?
        
             | astrange wrote:
             | Google? Push wages down? Google is mostly known for paying
             | top of market to keep a zoo of engineers whose only output
             | is blog posts about how smart they are because they solved
             | a problem they also caused.
             | 
             | (presumably because if they touch the ad system it might
             | break)
             | 
             | > a16z and YC must hate this. It puts a cap on their
             | returns.
             | 
             | And a16z's main business is investing in financial scams.
        
           | dweekly wrote:
           | There's some irony in that, given how many other companies
           | have "created browsers" that are just Chromium forks and
           | rubbed Google the wrong way.
           | 
           | The intro checklist for Antigravity includes watching VS Code
           | tutorials!
        
             | abirch wrote:
             | VS Code is based on Chromium:
             | 
             | https://chromium.googlesource.com/chromium/src/+/HEAD/docs/
             | v...
             | 
             | We've come full circle.
        
               | breppp wrote:
               | Chromium is based on Webkit which in turn is based on
               | KHTML, so maybe KDE needs to develop a cursor clone?
        
               | kaufmann wrote:
               | Kursor.
        
               | infthi wrote:
               | Kursor Klone
        
               | askvictor wrote:
               | Or, for the non-AI version, Kode
        
               | gmueckl wrote:
               | Well, Kate has been around as an KDE based advanced text
               | editor for nearly 2 decades now - its base feature set
               | isn't too different from a base VS Code installation. And
               | there's also KDevelop as a more full featured IDE.
        
             | modeless wrote:
             | It's a good point, and in fact I went and looked at the
             | original announcement of VS Code and it appears that
             | Microsoft didn't credit Chromium or Electron back then
             | either. I guess big companies are allergic to crediting
             | other big companies.
        
           | joedevon wrote:
           | It's Windsurf
        
             | anticensor wrote:
             | and also isn't
        
           | noxa wrote:
           | +1 - it also doesn't support remote ssh (the open vsx
           | variant), so is probably only focused at local web design
           | development vibe coding ;(
           | 
           | Should have just been an extension with a paid plan.
        
             | therein wrote:
             | It also feels like they couldn't use the GOOGLE ANTIGRAVITY
             | logo enough times in this blog post. Gigantic image with
             | the logo and a subtitle, plastered over and over again.
        
               | LogicFailsMe wrote:
               | I no longer bother reading their press releases. I'd much
               | rather read the comments and threads like these to get
               | the real story. And I say that as a former googler.
        
               | Towaway69 wrote:
               | I guess non-former googlers aren't allowed to say that -
               | contractually.
        
             | dabockster wrote:
             | > so is probably only focused at local web design
             | development vibe coding
             | 
             | Quick, someone throw the Linux kernel source at it and
             | report back! xD
        
           | catigula wrote:
           | https://www.microsoft.com/en-us/edge
           | 
           | Nothing in there about chromium.
        
           | uatec wrote:
           | The welcome video says "We started with the core IDE then..."
           | and shows a picture of VS Code.
           | 
           | They knew exactly what they were saying.
        
           | DuperPower wrote:
           | AI companies blogs are not for devs, they write for tech
           | journalists to hype
        
           | kccqzy wrote:
           | 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.
        
             | pbhjpbhj wrote:
             | >it is madness to require so many acknowledgements
             | 
             | Really? Madness? "We started with VS Code to develop our
             | IDE...".
             | 
             | Oh, so onerous.
             | 
             | A thank you to the principle developers is a minimum if
             | you're using someone else's work commercially and aren't an
             | asshole.
             | 
             | No it's not a legal requirement, it's just about being a
             | good part of society.
        
               | kccqzy wrote:
               | Yes, madness. VS Code wasn't developed entirely by
               | Microsoft. It uses plenty of other open source libraries.
               | Why is it that VS Code should be acknowledged but not the
               | underlying V8 engine, or Chromium, or WebKit or KHTML?
               | 
               | Stallman said that in 1997 there were 75 acknowledgements
               | in a single piece of software. With today's trend of
               | micro libraries on npm, there will be at least thousands
               | of acknowledgements in one piece of software.
        
           | Maxatar wrote:
           | And VSCode is based on Chromium (a fact you won't find on VS
           | Code's website other than related to updates).
        
             | sedatk wrote:
             | VSCode isn't a Chromium fork, it's an Electron app.
             | Utilizing something is different than making a derivative
             | of it. For example, an empty "Hello World" Electron app
             | wouldn't have any value for an app developer, but creating
             | a web browser derived from Chromium means you've already
             | finished 99.9% of the work.
        
             | verst wrote:
             | It uses Electron which itself uses the Chromium rendering
             | engine.
        
             | bangaladore wrote:
             | That's like saying news.ycombinator.com is based on
             | Chromium.
             | 
             | VSCode runs on chromium, like any website you visit when
             | using a Chromium browser.
        
               | bangaladore wrote:
               | To the downvoter:
               | 
               | VSCode -> Electron (essentially purpose specific web
               | browser) -> Chromium
               | 
               | news.ycombinator.com -> General Purpose Web Browser ->
               | Chromium
        
         | ethmarks wrote:
         | Interesting that they include non-Gemini models. Both Claude
         | and GPT oss are both on Google Cloud, so I assume that
         | Antigravity is using GC as the provider and not making API
         | calls to Anthropic or OpenAI.
        
           | NitpickLawyer wrote:
           | They're subsidising calls for data & reward signals. If they
           | can do that without also sharing the data with other
           | providers it's a win/win.
        
         | golergka wrote:
         | As somebody who worked on two IDEs which didn't fork VSCode but
         | still used Monaco for code editing views, I think forking
         | VSCode is almost always the right solutions for a new IDE. You
         | get extensions, familiarity and most importantly, don't waste
         | valuable time on the boring stuff which VSCode has already
         | implemented.
         | 
         | Nothing bad with using code other people made open. Our whole
         | industry is built on this.
        
           | nahuel0x wrote:
           | Why the need to fork it instead of creating a new extension?
           | (besides marketing)
        
             | benoau wrote:
             | Because if they're just an extension they're stuck with
             | whatever rules Microsoft makes up, and Google is no
             | stranger to using this leverage against others.
        
             | meowface wrote:
             | Because there are plenty of good reasons why you may want
             | to modify/extend the code and the look and feel beyond what
             | an extension would let you do.
             | 
             | I never understood why people scoff at VS Code forks. I'd
             | honestly tend to be more skeptical of new editors that _don
             | 't_ fork VS Code, because then they're probably missing a
             | ton of useful capabilities and are incompatible with all
             | the VSC extensions everyone's gotten used to.
        
             | jonny_eh wrote:
             | https://www.bleepingcomputer.com/news/microsoft/microsoft-
             | ap...
        
           | ghuntley wrote:
           | it isn't http://ghuntley.com/fracture
           | 
           | forking vscode? simple. extensions not so simple. they are
           | controlled by microsoft. without them you'll run into
           | continual papercuts as a vendor who has forked vscode.
        
         | aniviacat wrote:
         | I wonder why these forks don't build off Eclipse Theia. Is it
         | because of the Eclipse license?
        
           | desireco42 wrote:
           | I am not fan of Eclipse, mostly due to bad experience with
           | it, but this is excellent idea, if more people and companies
           | would invest in it's development, we would have alternative
           | to VS Code.
        
           | chuckadams wrote:
           | Just the Eclipse legacy and development approach perhaps.
           | Eclipse is 24 years old and its codebase stems from VisualAge
           | before that, so I imagine it's ... idiosyncratic. Probably
           | the most successful thing built on OSGi, though I dont
           | imagine Theia brought any of that along.
        
             | gmueckl wrote:
             | At least for a while, Eclipse seemed like an Architect
             | Astronaut's happy fever dream: actual bits if
             | implementations hidden behind 5 to 10 nested interfaces and
             | facades wherever I looked. I remember that at one point
             | that I wanted to debug some weird behavior in the Plugin
             | Development Kit and after sinking about half a day into
             | exploring the source code, I didn't even see a single line
             | of code that was actually doing anything meaningful. It was
             | quite shocking to my old junior level self.
        
         | alyxya wrote:
         | Making a VSCode fork is probably the wrong direction at this
         | point in time. The future of agentic coding should need less
         | support for code editor related functionality, and could
         | eventually primarily support viewing code rather than editing
         | code. There's a lot more flexibility in UI starting from
         | scratch, and personally I want to see a UI that allows flexible
         | manipulation of context and code changes with multiple agents.
        
         | timeon wrote:
         | > VSCode
         | 
         | Which is on top of 'Chrome'.
         | 
         | Interesting sandwich: Google-Microsoft-Google.
        
       | actionfromafar wrote:
       | "We must also do Loveable"
        
       | mccoyb wrote:
       | > Bajillions of dollars invested in the development of some of
       | the most powerful computational artifacts to date.
       | 
       | > Fork VS Code, add a few workflow / management ideas on top.
       | 
       | > "Agentic development platform"
       | 
       | I'm Jack's depressed lack of surprise.
       | 
       | Please someone, make me feel something with software again.
        
         | collingreen wrote:
         | There is cool stuff out there! Look beyond the companies with
         | $B valuations and you can find smart, passionate people making
         | neat stuff.
        
         | conartist6 wrote:
         | Challenge accepted
        
         | rglover wrote:
         | If you want to feel something with software, leave the industry
         | and never look back, saving programming as something you do for
         | your own joy/reward (I'm not being hyperbolic--I'd argue we're
         | in the early days of the web's "dark ages").
         | 
         | Unfortunately, once money came into the picture, quality,
         | innovation, and anything resembling true progress flew out the
         | window.
        
         | surgical_fire wrote:
         | > Please someone, make me feel something with software again.
         | 
         |  _Work with what you love, and you will never love anything
         | again._
        
       | ports543u wrote:
       | Program not needed.
        
       | msci100 wrote:
       | So this is Google's version of Windsurf's Wave 10 before the
       | whole team got poached? https://windsurf.com/blog/windsurf-
       | wave-10-browser
       | 
       | Trying to understand how this is anything net new in the space.
        
         | Namahanna wrote:
         | Looks to be. The UI has almost the exactly the same bits, and I
         | even got 'Cascade' references as using it.
        
         | Yeroc wrote:
         | I haven't used Windsurf (been using Claude Code and similar).
         | Does it provide an Agent Manager window/view? This to me looks
         | more useful to me than the browser integration piece.
        
       | antimora wrote:
       | Why is scrolling modified on this page? I how to disable it?
        
         | antgonzales wrote:
         | Came here to say this, it's super frustrating.
        
         | meetpateltech wrote:
         | Put this in your browser console to force default scrolling
         | var css = 'body { height: auto !important; overflow: auto
         | !important; } .smooth-scroll-wrapper { transform: none
         | !important; position: static !important; }
         | div[style*="position: fixed"] { position: static !important;
         | overflow: visible !important; inset: auto !important; }';
         | var style = document.createElement('style');
         | style.innerHTML = css;       document.head.appendChild(style);
         | console.log("Default scroll forced.");
        
           | Barbing wrote:
           | Thank you. Seems the website didn't like this btw, Safari
           | macOS. Forced return to top of page when scrolling too fast.
        
         | phantasmish wrote:
         | Google likes to fuck with basic browser functionality for some
         | reason. Scrolling, sometimes also how "click" intents through
         | touch are triggered (that is, using js listeners for touch
         | events instead of watching for the browser to communicate a
         | "click" on an element; this does usability-killing shit like
         | make a touch-to-stop-scrolling get interpreted as a click on
         | whatever happens to be under your finger). I have no idea why
         | they do this, but they do it a lot, so it must be a cultural
         | thing.
         | 
         | And I don't mean like some designers will highjack scroll to
         | deliver a different experience like slide-like transitions or
         | something (which may or may not be, differently, awful) but
         | they'll override it just to give you ordinary scrolling, except
         | much worse (as on this page).
         | 
         | Seems like a lot of work to do just to make something shittier,
         | but what do I know, I probably can't implement a* on a
         | whiteboard from memory or whatever.
        
         | tencentshill wrote:
         | They want everyone to see what the webpage looks like on their
         | Mac.
        
           | juancn wrote:
           | I'm on a Mac and that scrolling speed is not how Mac's
           | scroll. The acceleration and drag are all wrong.
        
           | AlexandrB wrote:
           | It's weird and laggy on Mac too. Not sure who this is for.
        
         | dansalvato wrote:
         | I can't believe these "smooth scrolling" scripts are still a
         | thing. I was wondering why I was having a hard time scrolling
         | the page on my phone, when I got to my PC and felt the reason.
         | 
         | It's incredible to think how many employees of this world-
         | leading Web technology company must have visited this site
         | before launch, yet felt nothing wrong with its basic behavior.
        
       | jihadjihad wrote:
       | > 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!
        
         | SR2Z wrote:
         | If you're building something new you'll need some skilled
         | people around
        
         | bakies wrote:
         | After a bunch of people leave the company it's already like
         | nobody knows how anything is built. This seems like a good
         | thing to accelerate understanding a codebase.
        
           | skeeter2020 wrote:
           | it's funny - nervous funny, not haha funny - that you think
           | drawing a real issue like this out into the open would focus
           | an organization on solving it.
        
           | cactusplant7374 wrote:
           | Agents pretty good at describing how a particular feature
           | works. It's not as dire as you make it seem.
        
         | crazygringo wrote:
         | You can ask agents to identify and remove cruft. You can ask an
         | agent why something is breaking -- to hypothesize potential
         | causes and test them for validity. If you don't understand how
         | something is built, you can ask the agent to give you an
         | overview of the architecture and then dive into whatever part
         | you want to explore more.
         | 
         | And it's not like any of your criticisms don't apply to human
         | teams. They also let cruft develop, are confused by breakages,
         | and don't understand the code because everyone on the original
         | team has since left for another company.
        
           | renegade-otter wrote:
           | Because what we need is not lazy people - we need lazy people
           | with AI? How is this even a justification?
        
             | crazygringo wrote:
             | Sorry, where did "lazy people" come from? Nobody's talking
             | about anybody being lazy.
        
               | infintropy wrote:
               | I just like that evolution doesnt really care. People can
               | opine on laziness and proper methodology. Its handwaving
               | compared to how things shake out.
               | 
               | Nature does select for laziness. The laziest state that
               | can outpace entropy in diverse ways? Ideal selection.
        
           | NitpickLawyer wrote:
           | > you can ask the agent to give you an overview of the
           | architecture and then dive into whatever part you want to
           | explore more.
           | 
           | This is actually a cool use that's being explored more and
           | more. I first saw it in the wiki thing from the devin people,
           | and now google released one as well.
        
           | flatline wrote:
           | Humans are just better at communicating about their process.
           | They will spend hours talking over architectural decisions,
           | implementation issues, writing technical details in commit
           | messages and issue notes, and in this way they not only debug
           | their decisions but socialize knowledge of both the code and
           | the reasons it came to be that way. Communication and
           | collaboration are the real adaptive skills of our species. To
           | the extent AI can aid in those, it will be useful. To the
           | extent it goes off and does everything in a silo, it will
           | ultimately be ignored - much like many developers who attempt
           | this.
           | 
           | I do think the primary strengths of genai are more in
           | comprehension and troubleshooting than generating code - so
           | far. These activities play into the collaboration and
           | communication narrative. I would not trust an AI to clean up
           | cruft or refactor a codebase unsupervised. Even if it did an
           | excellent job, who would really know?
        
             | crazygringo wrote:
             | > _Humans are just better at communicating about their
             | process._
             | 
             | I wish that were true.
             | 
             | In my experience, most of the time they're _not_ doing the
             | things you talk about -- major architectural decisions don
             | 't get documented anywhere, commit messages give no "why",
             | and the people who the knowledge got socialized to in
             | unrecorded conversations then left the company.
             | 
             | If anything, LLM's seem to be far more consistent in
             | documenting the rationales for design decisions, leaving
             | clear comments in code and commit messages, etc. if you ask
             | them to.
             | 
             | Unfortunately, humans generally are _not_ better at
             | communicating about their process, in my experience. Most
             | engineers I know enjoy writing code, and hate documenting
             | what they 're doing. Git and issue-tracking have helped
             | somewhat, but it's still very often about the "what" and
             | not the "why this way".
        
               | recitedropper wrote:
               | "major architectural decisions don't get documented
               | anywhere" "commit messages give no "why""
               | 
               | This is so far outside of common industry practices that
               | I don't think your sentiment generalizes. Or perhaps your
               | expectation of what should go in a single commit message
               | is different from the rest of us...
               | 
               | LLMs, especially those with reasoning chains, are
               | notoriously bad at explaining their thought process. This
               | isn't vibes, it is empiricism:
               | https://arxiv.org/abs/2305.04388
               | 
               | If you are genuinely working somewhere where the people
               | around you are worse than LLMs at explaining and
               | documenting their thought process, I would looking
               | elsewhere. Can't imagine that is good for one's own
               | development (or sanity).
        
               | crazygringo wrote:
               | I've worked everywhere from small startups to megacorps.
               | The megacorps certainly do better with things like
               | initial design documents that startups often skip
               | entirely, but even then they're often largely out-of-date
               | because nobody updates them. I can guarantee you that I
               | am talking about common industry practices in consumer-
               | facing apps.
               | 
               | I'm not really interested in what some academic paper has
               | to say -- I use LLM's daily and see first-hand the
               | quality of the documentation and explanations they
               | produce.
               | 
               | I don't think there's any question that, as a general
               | rule, LLM's do a much better job documenting what they're
               | doing, and making it easy for people to read their code,
               | with copious comments explaining what the code is doing
               | and why. Engineers, on the other hand, have lots of
               | competing priorities -- even when they _want_ to document
               | more, the thing needs to be shipped yesterday.
        
               | recitedropper wrote:
               | Alright, I'm glad to hear you've had a successful and
               | rich professional career. We definitely agree that
               | engineers generally fail to document when they have
               | competing priorities, and that LLMs can be of use to help
               | offload some of that work successfully.
               | 
               | Your initial comment made it sound like you were
               | commenting on a genuine apples-for-apples comparisons
               | between humans and LLMs, in a controlled setting. That's
               | the place for empiricism, and I think dismissing studies
               | examining such situations is a mistake.
               | 
               | A good warning flag for why that is a mistake is the
               | recent article that showed engineers estimated LLMs sped
               | them up by like 24%, but when measured they were actually
               | slower by 17%. One should always examine whether or not
               | the specifics of the study really applies to them--there
               | is no "end all be all" in empiricism--but when in doubt
               | the scientific method is our primary tool for determining
               | what is actually going on.
               | 
               | But we can just vibe it lol. Fwiw, the parent comment's
               | claims line up more with my experience than yours. Leave
               | an agent running for "hours" (as specified in the
               | comment) coming up with architectural choices, ask it to
               | document all of it, and then come back and see it is a
               | massive mess. I have yet to have a colleague do that,
               | without reaching out and saying "help I'm out of my
               | depth".
        
           | tyg13 wrote:
           | Sure, you can ask the agents to "identify and remove cruft"
           | but I never have any confidence that they actually do that
           | reliably. Sometimes it works. Mostly they just burn tokens,
           | in my experience.
           | 
           | > And it's not like any of your criticisms don't apply to
           | human teams.
           | 
           | Every time the limitations of AI are discussed, we see this
           | unfair standard applied: ideal AI output is compared to the
           | worst human output. We get it, people suck, and sometimes the
           | AI is better.
           | 
           | At least the ways that humans screw up are predictable to me.
           | And I rarely find myself in a gaslighting session with my
           | coworkers where I repeatedly have to tell them that they're
           | doing it wrong, only to be met with "oh my, you're so right!"
           | and watch them re-write the same flawed code over and over
           | again.
        
         | nova22033 wrote:
         | _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?
        
           | medvezhenok wrote:
           | The increasing levels of abstraction work only as long as the
           | abstractions are deterministic (with some limited exceptions
           | - i.e. branch prediction/preloading at CPU level, etc). You
           | can still get into issues with leaky abstractions, but
           | generally they are quite rare in established high->low level
           | language transformations.
           | 
           | This is more akin to manager-level view of the code (who need
           | developers to go and look at the "deterministic"
           | instructions); the abstraction is a lot lot more leaky than
           | high->low level languages.
        
           | skeeter2020 wrote:
           | even JS doesn't churn as fast as the models powering vibe
           | coding, and that cut & paste node app is still deterministic,
           | compared to what happens when the next version of the model
           | looks at AI-generated code from two years ago...
        
           | zahlman wrote:
           | 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.
        
         | foobarian wrote:
         | > The software of the future,
         | 
         | :chuckles nervously:
        
         | cindori wrote:
         | its so funny watching crusty devs malding over AI. do you all
         | write assembly code for your apps too? because who can trust
         | those magic compiler things right?
         | 
         | AI is just another abstraction that allows you to do 10x with
         | 1x the effort. either you'll learn to master it or you'll
         | become obsolete
        
           | dweinus wrote:
           | This argument warrants introspection for "crusty devs", but
           | also has holes. A compiler is tightly engineered and
           | dependable. I have never had to write assembly because I know
           | that my compiled code 100% represents my abstract code and
           | any functional problems are in my abstract code. That is not
           | true in AI coding. Additionally, AI coding is not just an
           | abstraction over code, but an abstraction over understanding.
           | When my code compiles, I don't need to worry that the
           | compiler misunderstood my intention.
           | 
           | I'm not saying AI is not a useful abstraction, but I am
           | saying that it is not a trustworthy one.
        
           | phantasmish wrote:
           | I'd worry about mastering the shift key, first.
        
       | skerit wrote:
       | They even packaged it for Linux.
        
       | acedTrex wrote:
       | Oh cool another ide for programming... aaaand its a vscode fork.
       | 
       | I dont know what i expected tbh
        
       | denysvitali wrote:
       | When it starts it prints: Using Cloud Code URL: https://daily-
       | cloudcode-pa.sandbox.googleapis.com
        
       | galaxyLogic wrote:
       | If I use this does it mean Google has access to all my code and
       | it may popup as"AI generated" in someone else's code?
        
         | dboreham wrote:
         | That's not quite how LLMs work.
        
           | BiteCode_dev wrote:
           | Not but you can be quite sure somewhere deep inside the TOS
           | there is a line saying their telemetry swallow your soul. If
           | not, it will be added. It's google, that's what they do.
        
         | Workaccount2 wrote:
         | Generally if you are paying full price (paying per token), then
         | it's not used for training.
         | 
         | If you are not paying, or paying a consumer level price
         | ($20/mo) you will be trained on.
         | 
         | ETA: In the terms they say they use your data because "free" is
         | the only option available in preview. However it does say you
         | can disable sharing in your settings...
        
           | tmikaeld wrote:
           | There is a setting to disable telemetry, unclear what this
           | means though.
        
       | pulkitsh1234 wrote:
       | Seems like they are trying to attack both Cursor and Lovable at
       | the same time...nice !
        
       | BiteCode_dev wrote:
       | Looks great, won't touch since they are probably going to do a
       | switcheroo or a shutdown as usual.
       | 
       | And of course I would need to look at all the implications of
       | spying, being locked out of google account and absence of support
       | that are google amo. No time for that. Not for them.
        
       | bluerooibos wrote:
       | Read the title, got excited. Read the page.. ah well, guess we'll
       | have to wait another while for FTL travel.
        
         | 0xf00ff00f wrote:
         | Same, I thought they had discovered cavorite with their quantum
         | computer or something.
        
       | torginus wrote:
       | I don't want to hate on this but I remember last week, when as a
       | backend developer doing frontend, I spent about 20 minutes
       | prompting Claude Sonnet in a loop trying to build a landing page
       | for a new feature.
       | 
       | The task was to put create a header, putting the company logo in
       | the corner and the text in the middle.
       | 
       | The resulting CSS was an abomination - I threw it all away and
       | rewrote it from scratch (using my somewhat anemic CSS knowledge),
       | ending up with like 3 selectors with like 20 lines of styles in
       | total.
       | 
       | This made me think that 1: CSS and the way we do UI sucks, I
       | still don't get why don't we have a graphical editor that can at
       | least do the simple stuff well. 2: when these model's don't wanna
       | do what you want them to the way you want them, they _really_ don
       | 't wanna.
       | 
       | I think AI has shown us there's a need for a new generation of
       | simple to write software and libraries, where translating your
       | intent into actual code is much simpler and the tools actually
       | help you work instead of barely allowing to fight be all the
       | accidental complexity.
       | 
       | We were much closer to this reality back in the 90s when you
       | opened up a drag and drop UI editor (like VB6, Borland Delphi,
       | Flash), wrote some glue code and out came an .exe that you could
       | just give to people.
       | 
       | Somewhere along the way, the cool kids came up with the idea that
       | GUIs are bad, and everything needs to go through the command
       | line.
       | 
       | Nowadays I need a shell script that configures my typescript CDK
       | template (with its own NPM repo), that deploys the backend infra
       | (which is bundled via node), the database schema, compiles the
       | frontend, and puts the code into the right places, and hope to
       | god that I don't run into all sorts of weird security errors
       | because I didn't configure the security the way the
       | browser/AWS/security middleware wanted to.
        
         | ninetyninenine wrote:
         | >Somewhere along the way, the cool kids came up with the idea
         | that GUIs are bad, and everything needs to go through the
         | command line.
         | 
         | It's important for people to feel like "hackers" that is the
         | primary reason why command line sort of exploded among devs.
         | Most devs will never admit this... they may not even realize
         | it, but I think this is the main reason it went big.
         | 
         | The irony is that the very thing that makes devs feel like
         | "hackers" is the very thing that's enabling agentic AI and
         | making developers get all resistant because they're feeling
         | dumber.
        
       | robofanatic wrote:
       | Something felt really "artificial" about that youtube video.
        
       | aeternum wrote:
       | Product leaders that apply world-changing technology breakthrough
       | names to their yet-another cloned SaaS product deserve more
       | shame.
       | 
       | Antigravity would be a world-changing technology. This isn't.
        
         | stavros wrote:
         | Wait until you see Google Perpetual Motion Machine, their
         | appointment-booking virtual assistant.
        
         | crazygringo wrote:
         | Why? There's nothing wrong with metaphor.
         | 
         | And agentic coding is about working at a much higher conceptual
         | level. Further from the ground. Antigravity is a functional
         | metaphor.
         | 
         | My only issue with it is that it's too long at five syllables,
         | and "anti-" is an inherently negative connotation. I'm guessing
         | this will eventually get renamed if it gets popular, much like
         | Bard was.
        
           | brazukadev wrote:
           | They should not be too metaphorical when naming a clone, tho.
        
             | crazygringo wrote:
             | Why not? And clone of what?
        
           | nonameiguess wrote:
           | As the parent said, actual anti-gravity is world-changing
           | technology. It's telling the very laws of nature to go fuck
           | themselves, you're gonna do what you want, even if all of
           | known physics says it's impossible.
           | 
           | Working at a higher conceptual level is just project
           | management. You're the legislator giving out unfunded
           | mandates rather than the agency staff that has to figure out
           | how to comply. There's power there, but it isn't anti-
           | gravity.
           | 
           | That said, I suspect this is really meant to allude to
           | https://xkcd.com/353/.
        
             | crazygringo wrote:
             | > _There 's power there, but it isn't anti-gravity._
             | 
             | That's why it's metaphor. "Operation Warp Speed" also
             | delivered vaccines quickly, but not faster than the speed
             | of light.
             | 
             | The list of company and product names that are based on a
             | metaphor that is very obviously exaggerated is endless.
             | Google doesn't index a googol number of pages either.
        
               | bflesch wrote:
               | I feel it's a bit ignorant of you to double down on your
               | argument and compare a cloned product release by some
               | macbook swinging google engineers with vaccination, which
               | actually positively impacted many human lives.
        
               | crazygringo wrote:
               | Are you for real? When someone disagrees, it's not
               | "ignorance" or "doubling down". It's just legitimate
               | disagreement. There's nothing I'm _ignorant_ of here, so
               | please don 't throw around insults like that.
               | 
               | I just continue to stand by the fact that naming products
               | using exaggerated metaphor is standard practice. The idea
               | that it is "shameful" or "ignorant" seems absurd. I think
               | it's OK not to take it too seriously. Nobody is going to
               | be confused and walk off of a cliff or something because
               | the product is named "antigravity"...
               | 
               | Do you get upset that the Milky Way candy bar doesn't
               | actually contain a galaxy within? Or that the Chicago
               | Bulls aren't as strong as actual bulls?
        
               | brazukadev wrote:
               | so your point is that you are policing what people think
               | is shameful or ignorant? Is it too much to not think too
               | high of a VSCode fork?
               | 
               | Geez, people are still this impressed by big tech?
        
       | robowo wrote:
       | Oh no. Not another VSCode fork...
        
       | Fraaaank wrote:
       | Anyone else stuck on 'setting up your account'?
        
         | ayhanfuat wrote:
         | My guess is it fails if you use a workspace account. I was able
         | to use it with my personal Google account.
        
           | Fraaaank wrote:
           | Hmm that does indeed seem to be the case.
        
         | kUdtiHaEX wrote:
         | Doesn't work with a workspace account for me but it does work
         | with my private account
        
         | om42 wrote:
         | Yes, its also failing on my workspace account but worked on my
         | personal. Might be a bug or a delayed deployment for workspaces
         | b/c it might need to be "enabled" by admins?
        
         | silveraxe93 wrote:
         | Need to use a personal account. Check the first question in the
         | FAQ: https://antigravity.google/docs/faq
        
         | dannyfritz07 wrote:
         | I'm not using a workspace account and am unable to get past
         | this step.
        
           | dannyfritz07 wrote:
           | Oh well. Uninstalled. This was my first experience doing
           | software development guided by AI. Doesn't seem like a tool
           | that will serve me well in the long run.
        
       | wiseowise wrote:
       | What the hell is going on with scrolling on their website?
        
         | quinnjh wrote:
         | yeah... almost completely broken on my iphone. something tells
         | me they used antigravity to vibe the website. would explain
         | other issues mentioned like vim keybindings being ignored.
         | 
         | ..."Youre absolutely right! I did mess up the internals of that
         | feature and incorrectly reported that it works. let me try
         | again..."
        
       | world2vec wrote:
       | Am I the only one stuck in the "Setting Up Your Account" loading
       | screen?
        
         | philmo1 wrote:
         | same
        
       | kUdtiHaEX wrote:
       | 2020: every day a new JS framework is announced
       | 
       | 2024: every day a new Chrome fork browser is announced
       | 
       | 2025: every day a new AI IDE vscode fork is announced
        
         | pooyamo wrote:
         | >vscode fork
         | 
         | I wonder why they are not trying to fixup something based on
         | their own GUI stacks like Flutter or Compose Multiplatform.
         | 
         | It seems only Zed is truly innovating in this space.
        
           | TheCraiggers wrote:
           | Well it's a helluva lot faster to make for one. For two, just
           | about everyone knows how to navigate in vscode by now.
           | Reducing the barrier of entry has obvious advantages.
        
             | koiueo wrote:
             | I have Cursor at work. The only element of the interface
             | I'm using (and know how to use) is the chat window.
             | 
             | IMO, it's an absolutely crappy IDE, crappy editor, with
             | absolutely incomprehensible hostile UI.
             | 
             | I have almost two decades of experience with Vim, Emacs and
             | IntelliJ. FWIW, I was able to easily find my ways in helix,
             | kakoune and Zed.
        
               | samsmith4 wrote:
               | Really? Can you say what you hate it about it pls
        
               | koiueo wrote:
               | From the top of my head in no specific order
               | 
               | - Icons on the toolbar in the left panel have no labels
               | or even tooltips. No way to know what they do without
               | clicking and checking.
               | 
               | - Space in the file explorer in the left panel opens a
               | file (haven't noticed such behavior in other editors --
               | totally unexpected).
               | 
               | - Maybe that's the artifact of me installing Vim plugin,
               | but Keyboard shortcuts displayed in the main menu don't
               | do what they say they do.
               | 
               | - It often offers installing some plugins, and I've
               | absolutely no idea why, and what will happen if I do or
               | if I don't.
               | 
               | I'm talking about Cursor, which I assume is exactly like
               | VS Code. Tried VS Code only once very long ago.
        
               | badc0ffee wrote:
               | I think the left toolbar icons do have tooltips, but they
               | show up pretty slowly for me (2-3 seconds).
        
               | koiueo wrote:
               | I'd like to retract my critique.
               | 
               | I just opened the app to see what else I can bring up,
               | and while clicking through UI I noticed I had some crappy
               | key bindings extension installed, which apparently caused
               | many of my annoyances.
               | 
               | I've probably installed it very long ago, or even by
               | accident.
               | 
               | For example, I was always annoyed that open
               | file/directory shortcut (one of most common operations)
               | is not assigned and requires mouse interaction -- fixed
               | by disabling the extension. Go to file shortcuts does
               | something completely different -- fixed by disabling the
               | extensions.
               | 
               | I likely won't adopt Cursor as my main IDE/Editor, but
               | it's miles better than I thought just an hour ago.
               | 
               | Thanks for your question :D
        
               | Degorath wrote:
               | Not the person you asked, but I hate how it screws up
               | keyboard shortcuts. It overrode the delete line shortcut
               | with its own inline chat one, for example.
               | 
               | Decided to ditch it for claude code right after that,
               | since I cannot be bothered to go over the entire list of
               | keyboard shortcuts and see what else it overrode/broke.
        
               | meowface wrote:
               | I've found that annoying too, but you can always rebind
               | them as you wish. It's only a few new keybinds that get
               | in the way of my muscle memory.
               | 
               | That said I also have moved to CLI agents like Claude
               | Code and Codex because I just find them more convenient
               | and, for whatever reason, more intelligent and more
               | likely to correctly do what I request.
        
           | botanrice wrote:
           | How would you say Zed is innovating? Never heard of it, just
           | taking a peek now.
        
             | koakuma-chan wrote:
             | It's written from scratch in Rust. It's super fast,
             | polished, etc. A world of difference compared to VSCode.
        
             | pooyamo wrote:
             | Zed team is writing their own in-house GUI stack [1] that
             | leverages the computer's GPU with minimal middleware in-
             | between. It's a lot of work short-term but IMO the payoff
             | would be huge if they establish themselves. I imagine they
             | could poke into the user-facing OS sector if their human-
             | agent interaction is smooth. (I have not tried it yet
             | though)
             | 
             | [1]: https://www.gpui.rs
        
               | wiseowise wrote:
               | Sublime did that almost 20 years ago.
        
               | skydhash wrote:
               | What about the text editing part?
        
             | roywiggins wrote:
             | it's _fast_
        
               | meowface wrote:
               | I am very sensitive to input latency and performance but
               | after comparing Zed and VS Code for a while I really
               | couldn't find any reason to stick with Zed. It's been a
               | year or so since I last tried it but VSC just lets me do
               | way more while still, IMO, having a nice, clean UI. I
               | never notice any performance or key input latency with
               | VSC.
        
               | don_searchcraft wrote:
               | If you work in a team there are some pretty good remote
               | collaboration tooling built into Zed.
        
           | bsimpson wrote:
           | Making an IDE sounds like an insane amount of effort.
           | 
           | FWIW, the Fuchsia team was working on an editor that had a
           | Flutter UI when run in Fuchsia:
           | 
           | https://xi-editor.io/frontends.html
        
           | charcircuit wrote:
           | These new editors are trying to differentiate themselves via
           | their AI features. Working on the core editor may waste
           | resources that could have been better spent improving the AI
           | features.
        
             | embedding-shape wrote:
             | Until someone finally figures out that we need to rethink
             | editors from the ground up to support different sort of
             | operations and editing experience, to better facilitate
             | LLMs doing work as agents.
             | 
             | But we're probably 1-2 years away from there still, so
             | we'll live with skinned-forks, VSCode extensions and TUIs
             | for now.
        
           | wiseowise wrote:
           | > I wonder why they are not trying to fixup something
           | extremely complex that only a handful players managed to get
           | right using gui stacks made with only mobile in mind that are
           | desperately trying to catch up to desktop now
        
         | candiddevmike wrote:
         | AIUI the forks are required because Microsoft is gatekeeping
         | functionality used by Copilot from extensions so they can't be
         | used by these agents.
        
           | ryandrake wrote:
           | I was going to ask why all these companies choose to fork the
           | entire IDE rather than just writing an extension like every
           | other sane developer, and this response is the most
           | believable reason why.
        
           | NewsaHackO wrote:
           | But Microsoft made VSCode lol, I think being able to gatekeep
           | things like that shouldn't allow a billion dollar company
           | just reuse all of your code instead of making their own IDE
        
             | whstl wrote:
             | Most of VSCode (yes, most) is a mishmash of other OSS
             | products, including Google Chromium! By that logic VSCode
             | itself shouldn't exist.
        
           | rapnie wrote:
           | > AI UI
           | 
           | 2026: every day ...
        
           | ewoodrich wrote:
           | Could you give an example of what they're gatekeeping for
           | Copilot exclusively? I'm kinda confused because Copilot in VS
           | Code isn't exactly a powerhouse of unique features in my
           | experience, it still feels well behind Roo/Kilo Code in most
           | ways I can think of, although much closer to the competition
           | than it was a year ago.
        
           | jcelerier wrote:
           | > AIUI the forks are required because Microsoft is
           | gatekeeping functionality used by Copilot from extensions so
           | they can't be used by these agents. reply
           | 
           | I always wonder how this works legally. VSCode needs to
           | comply with the LGPL (it's based on Chromium/Blink which is
           | LGPL) ; they should provide the entire sources that allow us
           | to rebuild our own "official" VSCode binary
        
             | SR2Z wrote:
             | I think that MS provides these extensions as plugins. The
             | core of VSCode is OSS, it's the extra bits that have a
             | different license.
        
         | pdntspa wrote:
         | Why is it so hard for these to be VSCode extensions and not
         | forks?
        
           | Etheryte wrote:
           | Microsoft has very specific constraints on what extensions
           | can and can't do, it's not a free for all. They're actively
           | defending their mote by allowing Copilot to do things in a
           | way that extensions couldn't. That's why all the serious
           | contenders make a fork, it's simply not possible to have the
           | same integration otherwise.
        
             | tremon wrote:
             | *moat
        
             | dust42 wrote:
             | Still it would be a lot wiser if all the forkers would do
             | one 'AI-enabled' fork together that exposes all the extras
             | that copilot gets. The barrier for testing would be much
             | lower and all the extension makers would also jump onto the
             | train. Likely MS would finally give in and make all the
             | extras available for everyone. But all the fragmentation
             | only helps MS.
        
               | ewoodrich wrote:
               | What exactly are the "extras" that Copilot gets?
               | 
               | I've had a Github Copilot subscription from work for 1yr+
               | and switch between the official Copilot and Roo/Kilo Code
               | from time to time. The official Copilot extension has
               | improved a lot in the last 3-6 months but I can't recall
               | _ever_ seeing Copilot do something that Roo /Kilo can't
               | do, or am I missing something obvious?
        
               | meowface wrote:
               | They're all going to have quite divergent opinions on how
               | to structure the fork and various other design decisions
               | that would inevitably lead to forks of the fork again.
               | 
               | I think forking VS Code is probably the most sensible
               | strategy and I think that will remain the case for many
               | years. Really, I don't think it's changing until AI
               | agents get so ridiculously good that you can vibe code an
               | entire full-featured polished editor in one or a few
               | sessions with an LLM. Then we'll be seeing lots of de
               | novo editors.
        
             | catigula wrote:
             | Which is a bit odd given that Claude code extension in
             | VSCode is by far the best agent integration into a codebase
             | that I know of.
        
               | ryandrake wrote:
               | I don't even know what the Claude Code extension does in
               | vscode. I have it installed but hell if I know what it's
               | doing. I run Claude in one of vscode's terminals, and do
               | everything through there. I do see (sometimes) diffs pop
               | up in the IDE, I guess that's the extent of this
               | integration.
        
               | Yeroc wrote:
               | The Claude Code extension on VS Code does very little
               | (too little in my opinion). The integration level with
               | agentic functionality provided by Antigravity goes much
               | deeper in my 20 minutes or so of playing with it. The
               | biggest value pieces I see is: Agent Manager window which
               | provides a unified view of all my agents running across
               | all my workspaces (!) where I can quickly approve or
               | respond to followup questions and quickly brings me to
               | the code in context for each agent, additionally, I can
               | select a piece of code and comment on it inline and that
               | comment gets sent to the correct, active agent. These two
               | things alone are items which I have been looking for...
               | Too bad I only have approval to use Claude Code at work.
               | This looks promising.
        
             | dabockster wrote:
             | This isn't true. I've gotten Claude and its forks to do
             | things the same way as Copilot.
        
             | hs86 wrote:
             | A bit tangential, but I asked the new Gemini model about
             | this, and it immediately traced back this quote to your
             | username: https://gemini.google.com/share/144b46094d6e
             | 
             | Creepy stuff :)
        
         | femiagbabiaka wrote:
         | 2020: every day a new electron fork is announced
         | 
         | 2024: every day a new electron fork is announced
         | 
         | 2025: every day a new electron fork is announced
        
         | hypeatei wrote:
         | > 2024: every day a new Chrome fork browser is announced
         | 
         | This is still happening. Didn't you see OpenAI's release of
         | Atlas?
        
         | CuriouslyC wrote:
         | We need a VS Code fork that just exposes more interfaces, and
         | does nothing else. Then all these forks could just use that
         | with power extensions, and it'd force Microsoft to change its
         | behavior.
        
           | Mogzol wrote:
           | there are now 15 competing standards
        
           | diegof79 wrote:
           | You just described Eclipse RCP.
           | 
           | The issue with Eclipse and that approach is the complexity of
           | mixing plugins to do everything, which kills the UX.
           | 
           | When VSCode started, the differentiator from Atom and Eclipse
           | was that the extension points were intentionally limited to
           | optimize the user experience. But with the introduction of
           | Copilot that wasn't enough, hence the amount of forks.
           | 
           | I think that the Zed approach of having a common protocol to
           | talk with agents (like a LSP but for agents) is much better.
           | The only thing that holds me from switching to Zed is that so
           | far my experience using it hasn't been that good (it still
           | has many rough edges)
        
             | CuriouslyC wrote:
             | There's actually an ACP that has some traction as well,
             | though I don't think it solves the problem of what the
             | agent can do in vscode:
             | https://agentcommunicationprotocol.dev/introduction/welcome
        
         | nicce wrote:
         | > 2024: every day a new Chrome fork browser is announced
         | 
         | I think this was more accurate around 2012. My local tech
         | magazine had their own fork and they attached CD with the
         | magazine which included the browser.
        
       | nthypes wrote:
       | It's a shame not even mention the amazing work of VSCode.
        
         | xnx wrote:
         | The VSCode homepage does not mention Chromium or Electron.
        
       | karlkloss wrote:
       | Finally! Affordable antigravity!
        
       | xinghai wrote:
       | Anyone got stuck on the "Setting Up Your Account" page ?
        
         | hughw wrote:
         | yep
        
           | hughw wrote:
           | I switched to my personal gmail identity and it succeeded.
        
         | xinghai wrote:
         | i tried with my personal account and still stuck on that page
        
       | egypturnash wrote:
       | Did they build this site with Antigravity because it sure is
       | broken on my iPad.
        
       | marstall wrote:
       | I don't get how these agents can work when even Claude Sonnet 4.5
       | (for example) needs a lot of hand-holding for basic, simple
       | bugfixing stuff. Wouldn't the agents just be huffing and puffing
       | their way off the rails all the time?
        
         | jeltz wrote:
         | They do and it is often entertaining.
        
         | hughw wrote:
         | This is the key question.
        
       | sippeangelo wrote:
       | > Neither engenders user trust in the work that the agent
       | undertook. Antigravity provides context on agentic work at a more
       | natural task-level abstraction, with the necessary and sufficient
       | set of artifacts and verification results, for the user to gain
       | that trust.
       | 
       | I'm going to need an AI summary of this page to even start
       | comprehending this... It doesn't help that the scrolling makes me
       | nauseous, just like real anti-gravity probably would.
        
         | thisisit wrote:
         | Written by AI now summarise and explained by AI.
        
         | dannyfritz07 wrote:
         | wow, you weren't kidding about the scroll induced nausea.
        
         | roncesvalles wrote:
         | Then there's this:
         | 
         | "A more intuitive task-based approach to monitoring agent
         | activity, presenting you with essential artifacts and
         | verification results to build trust."
         | 
         | The whole thing around "trust" is really weird. Why would I as
         | a user care about that? It's not as if LLMs are perfect oracles
         | and the only thing between us and a brave new world is blind
         | trust in whatever the LLM outputs.
        
       | irilesscent wrote:
       | Seems interesting, makes for the second vscode clone with ai
       | google has made. The demo they showed in the video avoided
       | showing code so I guess thats what they're aiming for. Although
       | when they mentioned you can easily verify code quality by looking
       | at end product screenshots it felt like they don't know what
       | 'code' quality means.
        
       | juancn wrote:
       | How long until it's killed?
       | 
       | I mean, google doesn't have the greatest track record.
       | 
       | Also, why does that site's scroll behavior is so weird? Just use
       | the browser's default for Ford's sake!
        
       | dboon wrote:
       | It's...a VSCode fork? Really? What has become of Google? Ten
       | years ago, when I was getting into the world of software, there
       | was still an aura about them. They built everything in this huge
       | monorepo, and it worked. They were this deeply technical company
       | for whom it seems anything could be done.
       | 
       | And now they can't even ship a desktop app without forking
       | VSCode? Look, I get it. There's this huge ecosystem. Everyone
       | uses it. I'm not saying it's damning or even bad to fork it.
       | 
       | But why is this being painted as something revolutionary? It's a
       | reskin of all the other tools which are variations on the same
       | theme, dressed up in business speak (an agent-first UX!). I'm
       | sure it's OK. I downloaded it. The default Tokyo Night theme is
       | unusable; the contrast can't be read. I picked Vim bindings, but
       | as soon as I tried to edit a file I noticed that was ignored.
       | 
       | What happened? Is this how these beautiful, innovative companies
       | are bound to end up?
        
         | koakuma-chan wrote:
         | They should've made an ACP server
         | https://agentclientprotocol.com
        
           | dboon wrote:
           | Yeah. I really like opencode, which provides an ACP.
        
         | throwacct wrote:
         | This. I don't think it'll move the needle. I already use vscode
         | with copilot and it's "good enough".
        
         | NewsaHackO wrote:
         | I can see why people don't release stuff on a permissive
         | lisense anymore. It is absolutely insane that google is even
         | allowed to do something like this.
        
       | mrzool wrote:
       | I wonder how many meetings where necessary to come up with that
       | name
        
         | esafak wrote:
         | https://xkcd.com/353/
        
       | rounakdatta wrote:
       | Vim mode isn't working, oh no!
        
         | josvdwest wrote:
         | yeah I also cannot get Vim to work in Antigravity?
        
       | mohsen1 wrote:
       | This is the fruit of Windsurf brain-drain and I think it might be
       | better than what's out there since those guys got to start from
       | scratch from everything they learned building Windsurf
        
         | xnx wrote:
         | I forgot all about Windsurf from ... 4 months ago. The pace is
         | crazy.
        
       | zb3 wrote:
       | Nice demo, but they didn't say the most important thing - how
       | much did Gemini API calls in that demo really cost? How much
       | tokens were consumed?
       | 
       | I know there's a "free plan with generous rate limits" but it's
       | obvious that they're losing money there.
        
       | ChrisArchitect wrote:
       | Blog post: https://antigravity.google/blog/introducing-google-
       | antigravi...
        
       | everyone wrote:
       | My experience with GPT and Claude, is that they are _fantastic_
       | for learning something brand new to me, as a kind of tutor..
       | 
       | But for writing code in some domain I am good in, they are pretty
       | much useless.. I would spend a lot longer struggling to get
       | something that barely functions from them VS writing it myself,
       | and the one I write myself will be terse and maintainable + if it
       | has bugs they will be like obvious ones, not _insane_ ones that a
       | human would never do.
       | 
       | Even just when getting them to write individual functions with
       | very clear and small scopes.
        
       | zb3 wrote:
       | Too bad they never show these magic AI-Assisted tools being used
       | to fix real-world bugs / implement feature requests in their open
       | source GH repositories.
       | 
       | What about a demo that shows how this can be used to fix for
       | example https://github.com/emscripten-
       | core/emscripten/issues/24792?
        
       | jimmar wrote:
       | I installed it, entered one prompt, clicked the "Proceed" button,
       | and got "Model quota limit exceeded."
       | 
       | Those quota limits brought me back down to earth quickly.
        
         | qayxc wrote:
         | Especially since:                   There is currently no
         | support for:              Paid tiers with guaranteed quotas and
         | rate limits         Bring-your-own-key or bring-your-own-
         | endpoint for additional rate limits         Organizational
         | tiers (self-serve or via contract)
         | 
         | So basically just another case of vendor lock-in. No matter
         | whether the IDE is any good - this kills it for me.
        
       | bufunobhmut wrote:
       | The amount of New Eras stuffed into New Eras is too damn high!
        
       | hugs wrote:
       | ""Autonomously, an Antigravity Agent writes code for a new
       | frontend feature, uses the terminal to launch localhost, and
       | actuates the browser to test that the new feature works."
       | 
       | very interesting times; i'm glad to see browser automation
       | becoming more mainstream as part of the ai-assisted dev loop for
       | testing. (disclosure: started the selenium project, now working
       | on something similar for a vibe coding context)
        
         | quinnjh wrote:
         | is that "vibium" ? as someone who tried setting up selenium in
         | a workflow im definitely interested.
        
           | hugs wrote:
           | yup, vibium!
        
         | mhl47 wrote:
         | Most people are missing the point here. Testing the GUI/feature
         | more reliable is something that Gemini 3 could unlock (looking
         | at the ScreenSpot-Pro benchmark and its general improvement on
         | visual understanding). At least for the (hobby-)projects I
         | attempted this was really a bottleneck having to always test
         | the GUI after each change as its quite often breaking
         | something.
        
       | bsimpson wrote:
       | Can we mark this as a dupe of
       | https://news.ycombinator.com/item?id=45968065
        
       | lucideer wrote:
       | Really hope the quality of the IDE is better than the website...
        
       | uejfiweun wrote:
       | Theory: the naming of this product is strategic. Google's goal is
       | to push something else above "Google antitrust" in the
       | autocomplete.
        
         | esafak wrote:
         | It's just an xkcd reference: https://xkcd.com/353/
        
       | sangeeth96 wrote:
       | With all due to respect to the folks working on Antigravity, this
       | feels like a vibe-coded VSCode fork to me. Font sizes, icon
       | sizes, panel sizes are all over the place (why?). To top it all
       | off, the first request just failed with overload/quota exceeded
       | errors (understandable, but still).
       | 
       | Looks like I'll wait to see if Google cares about putting the
       | polish into a VSCode fork that at least comes close to what
       | Cursor did.
        
         | fhinkel wrote:
         | The UI is certainly buggy, and things are getting fixed all the
         | time. Guess it was more a "let people try the agent manager"
         | instead of overfocusing on looks.
        
       | zkmon wrote:
       | The name sounds like it is not going to stick around for long.
        
         | MeetingsBrowser wrote:
         | A product name with five syllables is doomed.
        
       | nthypes wrote:
       | This is a vibe-coded VSCode fork. In a simple task, I got
       | overload/quota exceeded errors with horrible error handling. lol
        
         | alwinaugustin wrote:
         | Same for me also. I have Gemini Pro subscription, still it is
         | showing quota exceeded error.
        
       | ivanjermakov wrote:
       | dupe https://antigravity.google/
        
       | rglover wrote:
       | We have a real, serious problem when even Google (presumably with
       | a large share of the world's engineering might) is just forking
       | VSCode.
        
         | barbazoo wrote:
         | With every AI VSCode fork we get closer to the bubble popping.
        
         | raincole wrote:
         | Reusing open-source code is a "real, serious problem" now...?
        
         | parliament32 wrote:
         | Wait till you hear about how much Linux they use.
        
       | balls187 wrote:
       | Okay, maybe I'm stupid, but the demo video included pasting and
       | API key into the chat window.
       | 
       | That seems bad.
        
       | fosterfriends wrote:
       | Anyone else getting this error: "Agent execution terminated due
       | to model provider overload. Please try again later."
        
         | keepamovin wrote:
         | Yes I just tried GP3H and got this. GP3L is fine tho.
         | 
         | I like this tool.
         | 
         |  _edit: Scratch that, GP3L is erroring out too. Global hug I
         | guess. I still like this._
        
       | efields wrote:
       | MacOS/Safari User here. Stuck on 'Setting Up Your Account' once
       | I've authorized it in the browser. /shrug
        
         | bgrainger wrote:
         | https://news.ycombinator.com/item?id=45968731
         | Why can I not authenticate into Google Antigravity?
         | Google Antigravity is currently available for non-Workspace
         | personal Google accounts in approved geographies. Please try
         | using an @gmail.com email address if having challenges with
         | Workspace Google accounts (even if used for personal purposes).
         | 
         | https://antigravity.google/docs/faq
        
           | rkomorn wrote:
           | I love how being a paying Workspace customer for a decade or
           | more has actually locked me out of so many Google products
           | and features.
           | 
           | Not that I have any desire to try this at this point, but
           | it's always felt ironic.
        
           | efields wrote:
           | Thanks. Actually did use my personal account and got the
           | issue nonetheless.
        
             | haberdasher wrote:
             | Same
        
             | voxmatt wrote:
             | Same
        
         | ed wrote:
         | To save others the trouble, it doesn't matter whether you use
         | Chrome or Safari for the auth flow. It's broken on both. (I'm
         | using a personal @gmail account.)
        
         | Topfi wrote:
         | Had the same issue, have been able to sign in finally using a
         | Google Cloud Identity (former Workspace) account by changing my
         | IP via a VPN to Singapore. No idea why, but that worked. Tried
         | a few other countries too, but only had success with Singapore.
        
       | bdn_ wrote:
       | Just from the name, I thought this was going to be Google's
       | official take on the classic "Google Gravity" site from ages ago:
       | https://mrdoob.com/projects/chromeexperiments/google-gravity...
       | 
       | I used to love leaving that site open on public PCs and watching
       | the reactions that resulted :)
        
       | giancarlostoro wrote:
       | I just have zero faith in Google. How long until we hear that
       | someone mysteriously got banned by Google (as we see on HN every
       | few months? it feels like it anyway) and hear about how now they
       | have no AI tooling etc etc etc because its all married to their
       | Google Account.
       | 
       | Additionally... Google Code was shut down in 2016? I have zero
       | confidence in such a user hostile company. They gave you a Linux
       | phone, they extended it, and made it proprietary. They gave you a
       | good email account, extended it and made it proprietary. They
       | took away office software from you via Google Docs, so now you
       | don't even own the software they do.
       | 
       | No thanks.
        
       | vitaflo wrote:
       | A 5 syllable name for a product makes me wonder wtf the marketing
       | team is doing.
        
       | ddp26 wrote:
       | This whole blog post is seemingly about Google, not about the
       | user. "Why We Built Antigravity" etc. "We want Antigravity to be
       | the home base for software development in the era of agents" -
       | cool, why would I as the user care about that?
        
         | wiseowise wrote:
         | You wouldn't. It's made to suck out investor money and show
         | that google does something, not to actually bring value.
         | 
         | My crystal ball says it will be shutdown next year.
        
           | evandrofisico wrote:
           | Most of the AI products are not for the end user, they are
           | just signaling shareholders and possible investors that the
           | company is on the hype.
        
             | rvnx wrote:
             | There is also mechanism inside Google that rewards teams
             | that launches new products, more than the teams that
             | actually maintain existing ones.
        
             | vecter wrote:
             | This kind of cynicism is wild to me. Of course most AI
             | products (and products in general) are for end users.
             | Especially for a company like Google--they need to do
             | everything they can to win the AI wars, and that means
             | winning adoption for their AI models.
        
             | jmpeax wrote:
             | Google is still looking for investors?
        
           | rishabhd wrote:
           | funny my magic 8 ball says the same thing!
        
           | meowface wrote:
           | I know Google is quick to shut things down but these ultra-
           | cynical ultra-skeptical HN takes are so tiresome at this
           | point.
        
             | pydry wrote:
             | This is what Google _is_ like though. It is practically
             | part of their corporate DNA.
        
               | meowface wrote:
               | I do not believe that Google Antigravity is aimed at
               | wooing investors. I believe it is intended to be a
               | genuine superior alternative to Cursor and Kiro etc. and
               | is attempting to provide the best AI coding experience
               | for the average developer.
               | 
               | Most of the other people (so far) in this sub-thread do
               | not think this. They essentially have a conspiratorial
               | view on it.
        
         | tnolet wrote:
         | After using Google AI studio, Google Vertex, and Google Gemini
         | Chat I honestly can't wait to use Google Antigravity!
         | 
         | edit: Also Jules...
         | 
         | snark off:
         | 
         | I think the Google PMs should have coffee together and see if
         | all of this sprawl makes any sense.
        
           | koakuma-chan wrote:
           | It does?
           | 
           | Google AI studio is their developer dashboard.
           | 
           | Google Vertex is their equivalent of Amazon Bedrock.
           | 
           | Google Gemini Chat is their ChatGPT app for normies.
           | 
           | Google Antigravity is their Cursor equivalent.
        
             | vineyardmike wrote:
             | I agree what you've listed makes sense as a product
             | portfolio.
             | 
             | But AI Studio is getting vibe coding tools. AI Studio also
             | has a API that competes with Vertex. They have IDE plugins
             | for existing IDEs to expose Chat, Agents, etc. They also
             | have Gemini CLI for when those don't work. There is also
             | Firebase Studio, a browser based IDE for vibe coding.
             | Jules, a browser based code agent orchestration tool. Opal,
             | a node-based tool to build AI... things? Stich, a tool to
             | build UIs. Colab with AI for a different type of coding.
             | Notebook LM for AI research (many features now available in
             | Gemini App). AI Overviews and AI mode in search, which now
             | feature a generic chat interface.
             | 
             | Thats just new stuff, and not including all the existing
             | products (Gmail, Home) that have Gemini added.
             | 
             | This is the benefit of a big company vs startups. They can
             | build out a product for every type of user and every user
             | journey, at once.
        
               | koakuma-chan wrote:
               | I concede.
        
               | xnx wrote:
               | Don't forget Gemini CLI
               | 
               | In another 2 years we'll probably be back to just
               | "Google" as digital agent that can do any research,
               | creative, or coding task you can imagine.
        
             | drcongo wrote:
             | > Google Vertex is their equivalent of Amazon Bedrock
             | 
             | Well, that clears that up.
        
               | koakuma-chan wrote:
               | In "real world" you don't use OpenAI or Anthropic API
               | directly--you are forced to use AWS, GCP, or Azure. Each
               | of these has its own service for running LLMs, which is
               | conceptually the same as using OpenAI or Anthropic API
               | directly, but with much worse DX. For AWS it's called
               | Bedrock, for GCP--Vertex, and for Azure it's AI Foundry I
               | believe. They also may offer complementary features like
               | prompt management, evals, etc, but from what I've seen so
               | far it's all crap.
        
             | exitb wrote:
             | And In practice, when I needed to use one of their models
             | for a small project, it turned out that the only sane way
             | is to go via OpenRouter...
        
             | verdverm wrote:
             | Also gemini-cli (terrible)
             | 
             | Google ADK (agent development kit, awesome)
        
           | k1rd wrote:
           | you forgot jules
        
           | dygd wrote:
           | The also launched a coding agent Jules: https://jules.google/
        
             | pnathan wrote:
             | Jules is nifty. Weirdly heavy on the browser CPU.
        
             | meowface wrote:
             | Jules is the first and only one to add a full API, which
             | I've found very beneficial. It lets you integrate agentic
             | coding features into web apps quite nicely. (In theory you
             | could always hack your own thing together with Claude Code
             | or Codex to achieve a similar effect but a cloud agent with
             | an API saves a lot of effort.)
        
               | verdverm wrote:
               | Google ADK is real nice and gives you an API as well
               | (also web browser and terminal prompt)
        
         | layer8 wrote:
         | More accurately, it should be neither about Google nor about
         | the user, but about the product. Describe what the product is
         | and does, don't make assumptions about the user, and let the
         | user be the judge of it.
        
         | DoomDestroyer wrote:
         | I swear most of these pages is to sell this to companies so
         | they can force it onto developers.
         | 
         | The whole webpage looks like something from Apple.
        
         | hereme888 wrote:
         | All these companies were built by self-referential narcissists,
         | and it seems to be their culture at the core.
        
         | jimbokun wrote:
         | Great point!
         | 
         | Remember took my a while early in my career from changing my
         | resume away from saying "I want to do this at my next job and
         | make a lot of money" and towards "here is how I can make money
         | and save costs for your company".
         | 
         | Google didn't learn that lesson here. They are describing why
         | us using Antigravity is good for Google, not why us using
         | Antigravity is good for us.
        
       | qoez wrote:
       | I probably wouldn't have been drawn to coding if I was young
       | these days based on the same motivations that led me to venture
       | into it as a teen.
        
       | bobsomers wrote:
       | Another Google product whose launch will be used to justify
       | somebody's promotion, only to be left for dead only a few months
       | later after said promoted person moves on to something else.
       | 
       | Why would I even bother getting mildly invested in this when the
       | product launch/promotion incentive structure at Google is so well
       | known?
        
       | gloosx wrote:
       | _reads title_
       | 
       | Wow was google researching some kind of anti-gravity device
       | behind the curtains for real and then dropped it out of nowhere?
       | 
       | Ah damn, yet another ai-assisted-something. Crap.
        
       | arbuge wrote:
       | I've clocked out on agentic AI IDEs. Not installing anymore until
       | I hit an obvious wall with my current ones.
        
       | AbraKdabra wrote:
       | I mean ok, another VSCode fork, but am I the only one seeing the
       | .google TLD?
        
       | greatgib wrote:
       | Anyone want to bet how long this product will last before being
       | killed? From looking at the frontpage, my bet is 2 years at most.
        
       | mike_ivanov wrote:
       | New Era. A New - think about it for a sec - Era.
        
       | Barry-Perkins wrote:
       | Interesting concept! Curious to see how Google Antigravity works
       | and what practical applications it might have.
        
       | Aperocky wrote:
       | Meanwhile I'm still on my trusty vim and letting the AI do work
       | separately in another CLI.
        
         | koakuma-chan wrote:
         | I moved from Vim to Zed recently.
        
       | china wrote:
       | I enjoy that the post-login confirmation page
       | (https://antigravity.google/auth-success) still says "Docs |
       | Twitter"
        
       | israrkhan wrote:
       | why it could not be a VS Code extension?
        
       | andy_ppp wrote:
       | I enjoyed Theo's take on this
       | https://youtube.com/watch?v=8dTN4PBD2rg
       | 
       | But honestly Google software seems so buggy. The management class
       | took over there a long ago and are quietly ruining the company.
        
         | rvnx wrote:
         | Probably got inspired by Apple and their iOS 26 quality control
        
       | izzydata wrote:
       | Meanwhile I don't feel like the era of AI assisted software
       | development has even started.
        
       | einpoklum wrote:
       | Looking at that page makes me think I should go the other
       | direction and switch from a graphical IDE to vim or something.
       | You know, ground myself by adopting more gravity.
        
       | cube2222 wrote:
       | I'll be honest - this doesn't look half-bad.
       | 
       | It really seems like it's just standardizing into a first-class
       | UI what a lot of people have already been doing.
       | 
       | I don't think I'm the target for this - I already use Claude Code
       | with jj workspaces and a mostly design-doc first workflow, and I
       | don't see why I would switch to this, but I think this could be
       | quite useful for people who don't want to dive in so deep and
       | combine raw tooling themselves.
        
         | jadbox wrote:
         | It's a good start, but I found it lacking compared to
         | VSCode+CLINE+Gemini Pro.
        
         | meowface wrote:
         | As a daily Claude Code and Codex user, I've really got to start
         | getting into jj. I keep telling myself I will but I'm just so
         | used to git.
         | 
         | Can you elaborate on how you personally use jj workspaces with
         | command-line coding agents?
        
           | cube2222 wrote:
           | Sure. Honestly I think you can get the same with git work
           | trees, though I haven't tried.
           | 
           | After a couple iterations on this, I've ended up having
           | claude code vibe-code a helper CLI in Go for me which I can
           | invoke with `ontheside <new-workspace-name> <base-change>`
           | and will
           | 
           | - create a new jj workspace based on the given change
           | 
           | - create a docker container configured with everything my
           | unit tests need to run, my claude code config mounted, and jj
           | configured
           | 
           | - it also sets up a claude code hook to run `jj` (no
           | arguments) every time it changes a file, so that jj does a
           | snapshot
           | 
           | - finishes by starting an interactive claude code session
           | with `--dangerously-skip-permissions`
           | 
           | - it also cleans it all up once I exit the claude code
           | session and fish shell that's running it
           | 
           | With this I can have Claude Code working asynchronously on
           | something, while I can review (or peek) the changes in batch
           | from my main editor by running `jj show <change-id>` / `jj
           | diff -r "..."` (which in my case opens it up in the Goland
           | multi-file diff viewer). I can also easily switch to the
           | change it's working on in my main editor, to make some manual
           | modifications if necessary.
           | 
           | This is, in general, primarily for "in the background async
           | stuff" I want to have it work on. Most of the time I just
           | have a dead-normal claude code session running in my main
           | workspace.
           | 
           | Minor self-plug - if you want, I posted a jj intro article a
           | while ago[0], though it doesn't include my current workspace
           | usage.
           | 
           | [0]: https://kubamartin.com/posts/introduction-to-the-
           | jujutsu-vcs...
        
         | jfim wrote:
         | Hadn't heard of jj, it's a source control tool that advertises
         | that it's fast and compatible with the git on disk format.
         | https://github.com/jj-vcs/jj
        
         | Yeroc wrote:
         | The Agent Manager view providing a unified view of all active
         | agents and allowing you to immediately respond to any approval
         | requests or followup questions looks very useful regardless of
         | which VCS you're using under the covers. Am I missing something
         | here that jj does?
        
           | cube2222 wrote:
           | See my setup detailed in another sibling comment of yours, jj
           | is just a small part of it, and you can probably get that
           | with git too.
           | 
           | I'm already at full mental capacity planning and reviewing
           | the work of two agents (one foreground which almost never
           | asks for approval, and one background which never asks for
           | approval).
           | 
           | I don't really need the ability to juggle more of them, and
           | noticing their messages is not a bottleneck for me, while I'm
           | happy with the customizability and adaptability of my raw'er
           | workflow.
           | 
           | Maybe if they're as slow as codex...
        
       | taco_emoji wrote:
       | I clicked around for ~30 seconds and I have no idea what this is.
       | Great job, Google marketing
        
       | BigParm wrote:
       | Even if this is good, I'll never know because I'm not investing
       | time into something that will be canceled in one year.
        
       | beanjuiceII wrote:
       | probably discontinue it in a few months
        
       | skeeter2020 wrote:
       | you would need to be bonkers to trust Google at the intersection
       | of 1. vibe coding, 2. supporting developers, and 3. a product
       | that's not selling ads
        
       | htrp wrote:
       | Google AI products are the new chat product
        
       | qwertox wrote:
       | I can't get the agent to use my MCP server. The MCP config is
       | provided, and the application can query the tools, but the agent
       | can't access my server, only the http operations the application
       | performs to list the tools is something i see in my server logs.
       | It knows there are tools, because it sees the list, it tells me
       | when it's trying to connect to which tool. But that fails.
        
       | dmitrygr wrote:
       | "agentic" is the new "ai", which was the new "web3", which was
       | the new "5G", which was the new "4G", which was the new "HDR",
       | which was the new "HD"...
        
       | brovonov wrote:
       | Any reason why this isn't just an extension instead of another
       | fork?
        
       | recursive wrote:
       | Landing page instantly spun up the fan on my laptop. The
       | animation was about 3fps.
        
       | smcleod wrote:
       | Another VSCode fork! This is getting ridiculous.
        
       | _hao wrote:
       | Oh look, another piece of shit AI slop I won't use. Next!
        
       | t1234s wrote:
       | Is it safe to use these types of AI enhanced editors with files
       | that contain sensitive information?
        
       | clever-leap wrote:
       | So deceptive name. It has nothing to do with gravity.
        
       | 0xblinq wrote:
       | As if this was something one could trust it won't be shut down 3
       | years from now...
       | 
       | No thanks...
        
       | uatec wrote:
       | "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.
        
         | MangoToupe wrote:
         | Also, agents have no capacity to learn.
        
           | jstummbillig wrote:
           | Hold that thought.
        
           | bradfa wrote:
           | 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 fine tuning or other
           | training actions.
        
             | OtherShrezzing wrote:
             | 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.
        
             | ethmarks wrote:
             | 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.
        
         | Too wrote:
         | I thought you were joking until I saw the video where this is
         | an actual quote.
        
         | throwawaysleep wrote:
         | As an introvert, that's a pro, not a con.
         | 
         | The burden of human interaction is removed from building.
        
           | ithkuil wrote:
           | I'm an introvert and I love working with nice people.
           | 
           | I just need some time by myself to recharge after all the
           | social interactions.
        
             | warkdarrior wrote:
             | As a team lead, working with people is so... cumbersome.
             | They need time to recharge, lots of encouragement, and a
             | nice place to work in. Give me a coding agent any time!
        
               | discreteevent wrote:
               | As a PM working with team leads is so cumbersome...
        
           | AstroBen wrote:
           | Being antisocial isn't an introvert thing. I'm incredibly
           | introverted and still love having time interacting with
           | people
        
         | sorokod wrote:
         | "junior developers" is a convenient label, it is incorrect but
         | it will take a bit until we come up something that describes
         | entities that:
         | 
         | - can write code
         | 
         | - tireless
         | 
         | - have no aspirations
         | 
         | - have no stylistic or architectural preferences
         | 
         | - have massive, but at the same time not well defined, body of
         | knowledge
         | 
         | - have no intrinsic memories of past interactions.
         | 
         | - change in unexpected ways when underlying models change
         | 
         | - ...
         | 
         | Edit: Drones? Drains?
        
           | throwawaysleep wrote:
           | Sounds like a junior developer?
           | 
           | They can usually write code, but not that well. They have
           | lots of energy and little to say about architecture and
           | style. Don't have a well defined body of knowledge and have
           | no experience. Individual juniors don't change, but the cast
           | members of your junior cohort regularly do.
        
           | shermantanktop wrote:
           | - don't learn from what you tell them
           | 
           | - don't have career growth that you can feel good about
           | having contributed to
           | 
           | - don't have a genuine interest in accomplishment or team
           | goals
           | 
           | - have no past and no future. When you change companies, they
           | won't recognize you in the hall.
           | 
           | - no ownership over results. If they make a mistake, they
           | won't suffer.
        
             | cambaceres wrote:
             | Sounds like my teammates.
        
             | CamperBob2 wrote:
             | _- don 't learn from what you tell them_
             | 
             | We'll fix that, eventually.
             | 
             |  _- don 't have career growth that you can feel good about
             | having contributed to_
             | 
             | Humans are on the verge of building machines that are
             | smarter than we are. I feel pretty goddamned awesome about
             | that. It's what we're supposed to be doing.
             | 
             |  _- don 't have a genuine interest in accomplishment or
             | team goals_
             | 
             | Easy to train for, if it turns out to be necessary. I'd
             | always assumed that a competitive drive would be necessary
             | in order to achieve or at least simulate human-level
             | intelligence, but things don't seem to be playing out that
             | way.
             | 
             |  _- have no past and no future. When you change companies,
             | they won 't recognize you in the hall._
             | 
             | Or on the picket line.
             | 
             |  _- no ownership over results. If they make a mistake, they
             | won 't suffer._
             | 
             | Good deal. Less human suffering is usually worth striving
             | for.
        
               | noduerme wrote:
               | Replace suffering with caring and have your AI write that
               | again.
        
               | recursive wrote:
               | > Humans are on the verge of building machines that are
               | smarter than we are. I feel pretty goddamned awesome
               | about that. It's what we're supposed to be doing.
               | 
               | It's also the premise of The Matrix. I feel pretty
               | goddamned uneasy about that.
        
               | lowsong wrote:
               | > We'll fix that, eventually.
               | 
               | > Humans are on the verge of building machines that are
               | smarter than we are.
               | 
               | You're not describing a system that exists. You're
               | describing a system that might exist in some sci-fi
               | fantasy future. You might as well be saying "there's no
               | point learning to code because soon the rapture will
               | come".
        
             | AstroBen wrote:
             | Imagine having these complaints about a screwdriver
             | 
             | It's a tool, not an intelligent being
        
               | timeon wrote:
               | I do not need to babysit my screwdriver.
        
               | ponector wrote:
               | Yet.
               | 
               | Next year there will be AI screwdriver your employer
               | force you to use.
        
               | phs318u wrote:
               | At first, I thought "ponector's forgotten to add the /s"
               | 
               | Then I realised that this will actually happen, and was
               | sadly reminded we're now in the post-sarcasm era.
        
               | switchbak wrote:
               | Yeah, if my screwdriver undid the changes I just made to
               | my mower, constantly ignored my desire to unscrew screws
               | and instead punched a hole in my carb - I'd be throwing
               | that screwdriver in the garbage.
        
             | Sammi wrote:
             | - don't learn from what you tell them
             | 
             | Whenever I have a model fix something new I ask it to
             | update the markdown implementation guides I have in the
             | docs folder in my projects. I add these files to context as
             | needed. I have one for implementing routes and one for
             | implementing backend tests and so on.
             | 
             | They then know how to do stuff in the future in my
             | projects.
        
               | kvirani wrote:
               | Can this additional prompt from you also be automated? I
               | do this too but I forget sometimes. I don't know if a
               | general rule will be enough ?
        
               | troupo wrote:
               | > I add these files to context as needed.
               | 
               | Key words are these.
               | 
               | > They then know how to do stuff in the future in my
               | projects.
               | 
               | No. No, they don't. Every new session is a blank slate,
               | and you have to feed those markdown files manually to
               | their context.
        
               | Clent wrote:
               | They still aren't learning. You're learning and then
               | telling them to incorporate your learnings. They aren't
               | able to remember this so you need to remind them each
               | day.
               | 
               | That sounds a lot like '50 First Dates' but for
               | programming.
        
               | embedding-shape wrote:
               | > They aren't able to remember this
               | 
               | Yes, this is something people using LLMs for coding
               | probably pick up on the first day. They're not "learning"
               | as humans do obviously. Instead, the process is that you
               | figure out what was missing from the first message you
               | sent where they got something wrong, change it, and then
               | restart from beginning. The "learning" is you keeping
               | track of what you need to include in the context, how
               | that process exactly works, is up to you. For some it's
               | very automatic, and you don't add/remove things yourself,
               | for others is keeping a text file around they copy-paste
               | into a chat UI.
               | 
               | This is what people mean when they say "you can kind of
               | do "learning" (not literally) for LLMs"
        
           | astrange wrote:
           | You'd also have no intrinsic memory of past interactions if
           | we removed your hippocampus.
           | 
           | Coincidentally, the hippocampus looks like a seahorse
           | (emoji). It's all connected.
        
             | embedding-shape wrote:
             | > the hippocampus looks like a seahorse
             | 
             | Not to mention; hippocampus literally means "seahorse" in
             | Greek. I knew neither of those things before today, thanks!
        
           | thatoneengineer wrote:
           | "brooms"
        
             | edflsafoiewq wrote:
             | *distant sound of the sorcerer's apprentice is heard*
        
           | auspiv wrote:
           | I describe them in the claude training I'm doing for my
           | company as: super smart, infinitely patient, overeager
           | interns
        
             | kvirani wrote:
             | Sometimes smart sometimes the opposite, though. Perhaps due
             | to memory loss.
        
               | embedding-shape wrote:
               | Not sure "smart" or "dumb" are even the right axis to be
               | judging them by, seems like intrinsically human traits.
        
           | antfarm wrote:
           | - constantly ignore your advice
           | 
           | - constantly give wrong answers, with surprising confidence
           | 
           | - constantly apologize, then make the same mistake again
           | immediately
           | 
           | - constantly forget what you just told them
           | 
           | - ...
        
           | kranke155 wrote:
           | Robots ?
        
         | tjmadsen wrote:
         | OR - think about the junior developers being managers of
         | agents, and you are still a manager of junior developers. This
         | is not zero sum!
        
         | UltraSane wrote:
         | I enjoy getting into a good flow state and pounding out clever
         | and elegant code but watching a good LLM generate code
         | according to my specs and refining it is also enjoyable. I've
         | been burning through $250 of free Claude Code Web credits and
         | having multiple workers running at the same time is fun.
        
         | BeetleB wrote:
         | > It's like managing junior developers, but without having nice
         | people to work with.
         | 
         | Nice? I thought all sycophant LLMs were _exceedingly nice_.
        
           | kalaksi wrote:
           | For me, it feels so fake that I'd rather have it not try to
           | be nice. I guess I've gotten a bit used to it and ignore it
           | for most parts.
        
         | supportengineer wrote:
         | I'm at one of those companies where we're forced to be in the
         | office.
         | 
         | NO ONE TALKS TO EACH OTHER unless absolutely necessary for
         | work.
         | 
         | We get on Zooms to talk. Even with the person 1 cubicle over.
        
           | dabockster wrote:
           | > We get on Zooms to talk. Even with the person 1 cubicle
           | over.
           | 
           | Who normalized this?!!
        
             | Rohansi wrote:
             | I know it happens when you're working with people who may
             | be at home or another location.
        
             | zmmmmm wrote:
             | I actually do find there is a subset of meetings that are
             | far more productive on Zoom. We can be voice chatting on
             | one screen, share another screen and both be able to type,
             | record notes, pull up side research without interrupting
             | the conversation. It's a bit closer to co-working than a
             | meeting but it hits a sweetspot for me.
        
           | cloverich wrote:
           | > We get on Zooms to talk. Even with the person 1 cubicle
           | over.
           | 
           | But why? Required? Culture? Maybe it's the company?
        
           | jimbokun wrote:
           | Why?
        
         | roadside_picnic wrote:
         | I used to be really excited about "agents" when I thought
         | people were trying to build _actual agents_ like we 've been
         | working on in the CS field for decades now.
         | 
         | It's clear now that "agents" in the context of "AI" is really
         | about answering the question "How can we make users make 10x
         | more calls to our models in a way that makes it _feel_ like we
         | 're not just squeezing money out of them?" I've seen so many
         | people that think setting some "agents" of on a minutes to
         | hours long task of basically just driving up internal KPIs at
         | LLM providers is cutting edge work.
         | 
         | The problem is, I haven't seen any evidence at all that
         | spending 10x the number of API calls on an agent results in
         | anything closer to useful than last year when people where
         | purely vibe coding all the time. At least then people would
         | interactively learn about the slop they were building.
         | 
         | It's astounding to watch a coworker walk though through a PR
         | with hundreds of added new files and repeatedly mention "I'm
         | not sure if these actually work, but it does look like there's
         | something here".
         | 
         | Now I'm sure I'll get some fantastic "no true Scotsman" replies
         | about how my coworkers must not be skilled enough or how they
         | need to follow xyz pattern, but the _entire point_ of AI was to
         | remove the need for specialize skills and make everyone 10x
         | more productive.
         | 
         | Not to mention that the shift in focus on "agents" is also
         | useful in detracting from clearly diminishing returns on
         | foundation models. I just hope there are enough people that
         | still remember how to code (and think in some cases) to rebuild
         | when this house of cards falls apart.
        
       | codepoet80 wrote:
       | On my m2 MacBook Air with 16GB of RAM, it took over 12 minutes to
       | startup and get to a usable state. When it did, it was plainly
       | just a jacked version of VSCode. Opening a project caused it to
       | hang again. Dumped it. VSCodium, with the terminal pane open so I
       | can talk to Claude works fine for me...
        
       | theflyinghorse wrote:
       | I couldn't get it to work on M1 mac. it spins forever on login
       | screen.
        
         | floppyd wrote:
         | It's probably just under a huge load right now. Set it up
         | pretty easily earlier on an M1 Air, but agent chats fail
         | quickly with "model under load"
        
       | hereme888 wrote:
       | Are they explicitly excluding OpenAI in their IDE? (gpt-oss
       | running on Google cloud doesn't count)
        
       | bflesch wrote:
       | I just feel second-hand embarrassment when seeing these kind of
       | posts.
        
       | kune wrote:
       | This thing crashes on Ubuntu LTS 24.04 during start. Apparently
       | all these agents are not able to ensure that a desktop app starts
       | on a popular Linux distribution.
       | 
       | If Google has forgotten how to do Software, than the future
       | doesn't look bright.
        
       | fogzen wrote:
       | Wake me up when there's an IDE that cares about performance. My
       | coworkers are amazed my laptop can run for days because I don't
       | use Electron crap.
        
       | robinduckett wrote:
       | Just sits on "Setting Up Your Account" pane and does nothing :)
        
         | earlyriser wrote:
         | Same issue here. Are you on Mac?
        
           | epaga wrote:
           | I am and am having the same issue. Edit: And just as I posted
           | this comment, on my second launch, it went through after
           | about a minute of waiting.
        
         | mygoodgomez wrote:
         | Possibly a problem with "organization" accounts? I'm seeing the
         | never-ending spinner too.
        
           | gre wrote:
           | My personal mac worked, my work mac is spinning.
           | 
           | It has jamf among other stuff
        
         | BakeInBeens wrote:
         | I'm having the same issue. I thought it was just me or
         | something with their cloud network. I also haven't been able to
         | download Android Studio from the website for a month. I
         | couldn't even download it from my Macbook so probably not the
         | same issue.
        
       | lbrito wrote:
       | The agentic spam is exhausting. I just wanted to code.
       | 
       | Too early in my career to not give a shit and retire, but too
       | late be excited about these things and eager to learn. What a
       | time...
        
         | kevstev wrote:
         | I feel this. Tbh I was excited about most of the previous tech
         | fads I have come across. Sure, people got overzealous with them
         | and also failed to realize that what F*NG needs is not what
         | your three person startup needs, but there were always some
         | good and interesting ideas there.
         | 
         | This just feels... a little too dystopian. Companies hoovered
         | up the entirety more or less of all of our collective thoughts
         | and writings and output and now want to sell it back to us- and
         | I fear that cost is going to be extremely steep.
         | 
         | It's impressive, but at the same time, just feels like its
         | going to somehow be a net detractor to society, and yet I feel
         | I need to keep up with each new iteration or potentially get
         | washed over and left behind by the wave.
         | 
         | I am somewhat fortunate to be towards the top of the pyramid
         | and also in a position where I could theoretically ride off
         | into the sunset, but I fear the societal implications and the
         | pain that is going to come for vast numbers of people.
        
         | absoluteunit1 wrote:
         | Same here - completely relate.
         | 
         | One thing I've noticed though that actually coding (without the
         | use of AI; maybe a bit of tab auto-complete) is that I'm
         | actually way faster when working in my domain than I am when
         | using AI tools.
         | 
         | Everytime I use AI tools in my domain-expertise area, I find it
         | ends up slowing me down. Introducing subtle bugs, me having to
         | provide insane amount of context and details (at which point it
         | becomes way faster to do it myself)
         | 
         | Just code and chill man - having spent the last 6 months really
         | trying everything (all these context engineering strategies,
         | agents, CLAUDE.md files on every directory, et, etc). It really
         | easy still more productive to just code yourself if you know
         | what you're doing.
         | 
         | The thing I love most though - is having discussions with an
         | LLM about an implementation, having it write some quick unit
         | tests and performance tests for certain base cases, having it
         | write a quick shell script, etc. things like this, it's Amazing
         | and makes me really enjoy programming since I save time and can
         | focus on doing the actual fun stuff
        
           | kalaksi wrote:
           | When I'm doing the coding myself, I'm at least making steady
           | progress and the process is predictable. With LLMs, it's a
           | crapshoot. I have to get the AI to understand what I want and
           | may have try again multiple times, many times never
           | succeeding, and end up writing a lot of text anyway. And in
           | between, I'll have to read a lot of code that probably ends
           | up being thrown away or heavily modified. This probably
           | depends a lot on what kind of project one is working on,
           | though.
           | 
           | But it's like you said, I like using LLMs for completing
           | smaller parts or asking specific kind of help or having
           | conversations about solutions, but for anything larger, it
           | just feels like banging my head to a wall.
        
           | skydhash wrote:
           | One of the core principles of my workflow (inspired by REPL
           | development and some unix tools) is to start with a single
           | file (for a function or the whole project). The I refactor
           | the code to have a better organization and to improve
           | reliability, especially as I'm handling more scenarios (and
           | failure modes).
           | 
           | LLMs are not useful in this workflow, because they are too
           | verbose. Their answers are generic and handle scenarios you
           | don't even support yet. What's useful is good documentation
           | (as in truthful) and the code if it's open.
           | 
           | This approach has worked really well in my career. It gives
           | me KISS and YAGNI for free. And every line of code is
           | purposeful and have a reason to be there.
        
           | blauditore wrote:
           | Devs are starting to realize that the sweet spot for AI
           | support in coding is on a small scale, i.e. extended code
           | completion. Generating huge chunks of code is often not
           | reliable enough except for some niches (such as simple
           | greenfield projects which profit from millions of examples
           | online).
        
         | lm28469 wrote:
         | It's never too early to retire, there are plenty of other
         | things to work on outside of tech
        
           | mrits wrote:
           | The robots have that covered
        
       | Towaway69 wrote:
       | 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?
        
       | siliconc0w wrote:
       | 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.
        
       | 999900000999 wrote:
       | 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.
        
       | purpleflame1257 wrote:
       | Can't wait for my IDE to get sunset on me with no recourse.
       | Thanks google, but I'll pass.
        
       | dudu24 wrote:
       | I cannot stand webpages that hijack scrolling like that.
        
       | shoelessone wrote:
       | I really don't know why I struggle so much with this stuff. I
       | believe these models / agents / whatever write code that is often
       | at least as good as the code I write, and they are super helpful
       | tools, but it just feels like it takes away so much of the joy
       | that is programming to me. I'm not saying it's "right" of me to
       | feel this way, but for me the struggle, and the figuring things
       | out by testing, identifying patterns, or looking deeper into a
       | library's implementation (etc) is part of the challenge that
       | makes programming and software construction fun.
        
         | throwawaysleep wrote:
         | It is kind of joyless that my day is now wait 20 min, QA,
         | report back.
        
           | markstos wrote:
           | Absolutely.
        
         | incognito124 wrote:
         | I'm in your boat. I picked this career because I enjoy solving
         | problems and thinking and understanding. I'm interested in how
         | things work inside below the layers. To me, it's like the tech
         | has a purpose of its own, and not just to provide value.
         | 
         | Using agents effectively is this whole other skillset including
         | managing requirements, prioritization and, worse yet, I'm
         | rarely left with any knowledge. I don't nearly get the same joy
         | out of "I finished a task with an agent" like I do with "I had
         | a problem, I delved deep to understand it, learned something
         | new and solved it"
         | 
         | Then again, I bet people making furniture out of wood felt the
         | same about industrial furniture factories. And it can be argued
         | that not every use case needs custom tailored furniture...
        
       | throwaway13337 wrote:
       | I gave it a fair shot.
       | 
       | It is a vs code fork. There were some UI glitches. Some usability
       | was better. Cursor has some real annoying usability issues - like
       | their previous/next code change never going away and no way to
       | disable it. Design of this one looks more polished and less
       | muddy.
       | 
       | I was working on a project and just continued with it. It was
       | easy because they import setting from cursor. Feels like the
       | browser wars.
       | 
       | Anyway, I figured it was the only way to use gemini 3 so I got
       | started. A fast model that doesn't look for much context. Could
       | be a preprompt issue. But you have to prod it do stuff - no
       | ambition and a kinda offputting atitude like 2.5.
       | 
       | But hey - a smarter, less context rich Cursor composer model. And
       | that's a complement because the latest composer is a hidden gem.
       | Gemini has potential.
       | 
       | So I start using it for my project and after about 20 mins - oh,
       | no. Out of credits.
       | 
       | What can I do? Is there a buy a plan button? No? Just use a
       | different model?
       | 
       | What's the strategy here? If I am into your IDE and your LLM, how
       | do I actually use it? I can't pay for it and it has 20 minutes of
       | use.
       | 
       | I switched back to cursor. And you know? it had gemini 3 pro.
       | Likely a less hobbled version. Day one. Seems like a mistake in
       | the eyes of the big evil companies but I'll take it.
       | 
       | Real developers want to pay real money for real useful things.
       | 
       | Google needs to not set themselves up for failure with every
       | product release.
       | 
       | If you release a product, let those who actually want to use it
       | have a path to do so.
        
         | charliebwrites wrote:
         | > It is a vs code fork
         | 
         | Google may have won the browser wars with Chrome, but Microsoft
         | seems to be winning the IDE wars with VSCode
        
           | vovavili wrote:
           | I wouldn't bet on an Electron app winning anything long-term
           | in the dev-oriented space.
        
             | wnevets wrote:
             | Even if those devs are vibe-oriented?
        
             | verdverm wrote:
             | why? I don't have a problem with it, building extensions
             | for VS Code is pretty easy
             | 
             | Alternatives have a lot of features to implement to reach
             | parity
        
               | scotty79 wrote:
               | Ease of writing and testing extensions is actually the
               | cause why Electron won IDE wars.
               | 
               | Microsoft made a great decision to jump on the trend and
               | just pour money to lap Atom and such in optimization and
               | polish.
               | 
               | Especially when you compare it to Microsoft effort for
               | desktop. They acumulated several more or less component
               | libraries over they years and I still prefer WinForms.
        
             | jshen wrote:
             | It's been winning for a while
        
             | SR2Z wrote:
             | Electron apps will win because they're just web apps - and
             | web apps won so decisively years ago that they will never
             | go anywhere.
        
               | troupo wrote:
               | No. Electron apps won, not web apps. There's a huge
               | difference.
        
               | lerp-io wrote:
               | electron is just a wrapper for the browser tho
        
               | scotty79 wrote:
               | It funny that despite how terrible, convoluted and
               | maladapted web tech is for displaying complex GUIs it
               | still gradually ate lunch of every native component
               | library and they just couldn't innovate to keep up on any
               | front.
               | 
               | Amazon just released OS that uses React Native for all
               | GUI.
        
             | _harsch wrote:
             | its hold the market for over 10 years tho... i wished zed
             | would've not been under gpl
        
               | WD-42 wrote:
               | Why not GPL? So we could be seeing closed source
               | proprietary forks by now? How do you think the Zed team
               | would feel about that?
        
             | pjc50 wrote:
             | What other UI framework looks as good on Windows, Mac and
             | Linux?
        
               | exasperaited wrote:
               | This is the painful truth, isn't it?
               | 
               | IMO The next best cross-platform GUI framework is Qt
               | (FreeCAD, QGIS, etc.)
               | 
               | Qt6 can look quite nice with QSS/QStyle themes, these
               | days, and its native affordances are fairly good.
               | 
               | But it's not close. VSCode is nice-looking, to me.
        
               | gopher_space wrote:
               | Godot looks ok and is surprisingly easy to work with.
        
               | nurumaik wrote:
               | If you want electron app that doesn't lag terribly,
               | you'll end up rewriting ui layer from scratch anyway.
               | VSCode already renders terminal on GPU and GPU-rendered
               | editor area is in experimental. There will soon be no web
               | ui left at all
        
               | HumanOstrich wrote:
               | I'm not sure how you went from terminal and editor GPU
               | rendering, which can benefit from it, to "there will soon
               | be no web ui left at all".
        
               | matsz wrote:
               | > VSCode already renders terminal on GPU
               | 
               | When did they add that? Last time I used it, it was still
               | based on xterm.js.
               | 
               | Also, technically Chromium/Blink has GPU rendering built
               | in for web pages, so everything could run on GPU.
        
               | morganherlocker wrote:
               | > If you want electron app that doesn't lag terribly
               | 
               | My experience with VS Code is that it has no perceptible
               | lag, except maybe 500ms on startup. I don't doubt people
               | experience this, but I think it comes down to which
               | extensions you enable, and many people enable lots of
               | heavy language extensions of questionable quality. I also
               | use Visual Studio for Windows builds on C++ projects, and
               | it is pretty jank by comparison, both in terms of UI
               | design and resource usage.
               | 
               | I just opened up a relatively small project (my blog
               | repo, which has 175 MB of static content) in both editors
               | and here's the cold start memory usage without opening
               | any files:
               | 
               | - Visual Studio Code: 589.4 MB
               | 
               | - Visual Studio 2022: 732.6 MB
               | 
               | update:
               | 
               | I see a lot of love for Jetbrains in this thread, so I
               | also tried the same test in Android Studio: 1.69 GB!
        
             | whynotminot wrote:
             | It's kind of a meme to dunk on Electron, but here's it's
             | been for years.
             | 
             | It's part of the furniture at this point, for better or
             | worse. Maybe don't bet _on_ it, but certainly wouldn't be
             | smart to bet against it, either.
        
             | Miraste wrote:
             | VS Code is technically an Electron app, but it's not the
             | usual lazy resource hog implementation like Slack or
             | something. A lot of work went into making it fast. I doubt
             | you'll find many non-Electron full IDEs that are faster.
             | Look at Visual Studio, that's using a nice native framework
             | and it runs at the speed of fossilized molasses.
        
               | paulddraper wrote:
               | Exactly.
               | 
               | JetBrains, Visual Studio, Eclipse, Netbeans...
               | 
               | VS Code does well with performance. Maybe one of the new
               | ones usurps, but I wouldn't put my money on it.
        
             | btown wrote:
             | I strongly disagree.
             | 
             | Firstly, the barrier to entry lower for people to take web
             | experience and create extensions, furthering the ecosystem
             | moat for Electron-based IDEs.
             | 
             | Even more importantly, though, the more we move towards
             | "I'm supervising a fleet of 50+ concurrent AI agents
             | developing code on separate branches" the more the notion
             | of the IDE starts to look like something you want to be
             | able to launch in an unconfigured cloud-based environment,
             | where I can send a link to my PM who can open exactly what
             | I'm seeing in a web browser to unblock that PR on the
             | unanswered spec question.
             | 
             | Sure, there's a world where everyone in every company uses
             | Zed or similar, all the way up to the C-suite.
             | 
             | But it's far more likely that web technologies become the
             | things that break down bottlenecks to AI-speed innovation,
             | and if that's the case, IDEs built with an eye towards
             | being portable to web environments (including their entire
             | extension ecosystems) become unbeatable.
        
               | imiric wrote:
               | > Firstly, the barrier to entry lower for people to take
               | web experience and create extensions, furthering the
               | ecosystem moat for Electron-based IDEs.
               | 
               | The last thing I want is to install dozens of JS
               | extensions written by people who crossed that lower
               | barrier. Most of them will probably be vibe coded as
               | well. Browser extensions are not the reason I use
               | specific browsers. In fact, I currently have 4 browser
               | extensions installed, one of which I wrote myself. So the
               | idea that JS extensions will be a net benefit for an IDE
               | is the wrong way of looking at it.
               | 
               | Besides, IDEs don't "win" by having more users. The
               | opposite could be argued, actually. There are plenty of
               | editors and IDEs that don't have as many users as the
               | more popular ones, yet still have an enthusiastic and
               | dedicated community around them.
        
           | AstroBen wrote:
           | In user numbers, maybe. JetBrains is far ahead in actual
           | developer experience though
        
             | catlover76 wrote:
             | At best, that's subjective, but it's fact that JetBrains is
             | comically far behind when it comes to AI tooling.
             | 
             | They have a chance to compete fresh with Fleet, but they
             | are not making progress on even the basic IDE there, let
             | alone getting anywhere near Cursor when it comes to LLM
             | integration.
        
               | AstroBen wrote:
               | JetBrains' advantage is that they have full integration
               | and better understanding of your code. WebStorm works
               | better with TypeScript than even Microsoft's own
               | creation. This all translates into AI performance
               | 
               | Have you actually given them a real test yet - either
               | Junie or even the baseline chat?
        
           | alansammarone wrote:
           | I believe our definitions of "winning the IDE wars" are very,
           | very different. For one thing, using "user count" as a metric
           | for this like using "number of lines of code added" in a
           | performance review. And even if that was part of the metric,
           | people who use and don't absolutely fall in love with it, so
           | much so that they become the ones advocating for its use, are
           | only worth a tiny fraction of a "user".
           | 
           | neovim won the IDE wars before it even started. Zed has
           | potential. I don't know what IntelliJ is.
        
             | dabockster wrote:
             | > I don't know what IntelliJ is.
             | 
             | It started as a modernized Eclipse competitor (the Java
             | IDE) but they've built a bunch of other IDEs based on it.
             | Idk if it still runs on Java or not, but it had potential
             | last I used it about a decade ago. But running GUI apps on
             | the JVM isn't the best for 1000 reasons, so I hope they've
             | moved off it.
        
               | hatsix wrote:
               | Android Studio is built on the IntelliJ stack. Jetbrains
               | just launched a dedicated Claude button (the button just
               | opens up claude in the IDE, but there are some pretty
               | neat IDE integrations that it supports, like being able
               | to see the text selection, and using the IDE's diff
               | tool). I wonder if that's why Google decided to go VS
               | code?
        
               | ffsm8 wrote:
               | Uh, isn't that the regular Claude code extension that's
               | been available for ages at this point? Not jetbrains but
               | anthropics own development?
               | 
               | As a person paying for the jetbrains ultimate package
               | (all ides), I think going with vscode is a very solid
               | decision.
               | 
               | The jetbrains ides still have various features which I
               | always miss whenever I need to use another IDE (like way
               | better "import" suggestions as an easy to understand
               | example)... But unless you're writing in specific
               | languages like Java, vscode is way quicker and works just
               | fine - and that applies even more to agentic development,
               | where you're using these features less and less...
        
               | gmueckl wrote:
               | Jetbrains IDEs are all based on the JVM - and they work
               | better than VSCode or the full Visual Studio for me. It's
               | the full blown VS (which has many parts written in C++)
               | that is the most sluggish of them all.
        
               | discreteevent wrote:
               | You don't actually use it but somehow you know that
               | "running GUI apps on the JVM isn't the best for 1000
               | [unspecified] reasons".
               | 
               | - This isn't a scientific approach.
        
             | Sharlin wrote:
             | The IntelliJ family are probably the best IDEs on the
             | market currently.
        
             | smikhanov wrote:
             | > I don't know what IntelliJ is.
             | 
             | "I never read The Economist" - Management Trainee, aged 42.
        
           | WD-42 wrote:
           | VSCode IS chrome though.
        
             | recursive wrote:
             | Kind of like how Android is linux.
        
           | thanhhaimai wrote:
           | VSCode is based on Chromium. https://chromium.googlesource.co
           | m/chromium/src/+/HEAD/docs/v...
        
             | lokimedes wrote:
             | Which is based on Apple Webkit? The winner is always the
             | last marketable brand.
        
               | onion2k wrote:
               | That drives exactly $0 of Apple's revenue. It's only a
               | win if you care about things that don't matter.
        
               | usrnm wrote:
               | Both are based on khtml. We could be living in a very
               | different world if all that effort stayed inside the KDE
               | ecosystem
        
               | embedding-shape wrote:
               | Which came from "the KDE HTML Widget" AKA khtmlw. Wonder
               | if that's the furthest we can go?
               | 
               | > if all that effort stayed inside the KDE ecosystem
               | 
               | Probably nowhere, people rather not do anything that
               | contribute to something that does decisions they disagree
               | with. Forking is beautiful, and I think improves things
               | more than it hurts. Think of all the things we wouldn't
               | have if it wasn't for forking projects :)
        
               | benoau wrote:
               | It's "based on WebKit" like English is based on Germanic
               | languages.
        
               | dragonwriter wrote:
               | Note that these are somewhat different kinds of "based
               | on".
               | 
               | Chromium is an upstream dependency (by way of Electron)
               | for VSCode.
               | 
               | WebKit _was_ an upstream dependency of Chromium, but is
               | no more since the Blink /WebKit hard fork.
        
             | dragonwriter wrote:
             | VSCode is Electron based which, yes, is based on Chromium.
             | But the page you link to isn't about that, its about using
             | VSCode as dev environment for _working on_ Chromium, so I
             | don 't know why you linked it in this context.
        
           | cyberax wrote:
           | 15 years ago, every company had its own "BlahBlah Studio" IDE
           | built on top of Eclipse. Now it's VSCode.
           | 
           | Meanwhile, JetBrains IDEs are still the best, but remain
           | unpopular outside of Android Studio.
        
             | philipwhiuk wrote:
             | And IntelliJ
             | 
             | PyCharm's lack of popularity surprises me. Maybe it's not
             | good enough at venvs
        
           | pstuart wrote:
           | Developers, developers, developers!
           | 
           | https://www.youtube.com/watch?v=Vhh_GeBPOhs
        
         | pjc50 wrote:
         | > What's the strategy here? If I am into your IDE and your LLM,
         | how do I actually use it? I can't pay for it and it has 20
         | minutes of use.
         | 
         | I wonder how much Google shareholders paid for that 20 minutes.
         | And whether it's more or less than the corresponding extremely
         | small stock price boost from this announcement.
        
         | dabockster wrote:
         | > It is a vs code fork.
         | 
         | With vendor lock-in to Google's AI ecosystem, likely
         | scraping/training on all of your code (regardless of whatever
         | their ToS/EULA says), and being blocked from using the main VS
         | Code extensions library.
        
         | frays wrote:
         | This is the result of Google's Windsurf acquisition.
         | 
         | I expect huge improvements are still to be made.
        
           | gotrythis wrote:
           | Cognition bought Windsurf:
           | 
           | https://windsurf.com/blog/windsurfs-next-chapter
        
             | TiredOfLife wrote:
             | Google bought people and tech that made Windsurf:
             | 
             | https://windsurf.com/blog/windsurfs-next-stage
        
         | chermi wrote:
         | Lol the second I saw the antigravity release I thought "there's
         | no way I'm using that, they will kill it within a year". Looks
         | like they're trying to kill it at birth.
        
           | legedemon wrote:
           | Exactly my reaction. Every time I've used something from
           | Google, it ends up dead in a few years. Life is too short to
           | waste so many years learning something that is destined to
           | die shortly
        
             | zelphirkalt wrote:
             | These are just extended press releases, for marketing and
             | management layers, who don't have to use these things
             | themselves, but can look good, when talking about it.
        
           | upcoming-sesame wrote:
           | agree but at the same time there's not too much lock in with
           | these IDEs these days and switching is very easy. Especially
           | since they're all VSCode forks
        
         | eclipxe wrote:
         | What looked like out of credits for me was really just server
         | overload. Check the error and try again
        
         | schainks wrote:
         | I bet if you sign up for Google AI Ultra for a month your
         | limits will disappear.
        
           | throwaway13337 wrote:
           | I'm a Google AI Pro subscriber (the $~20/month one).
           | 
           | I don't think it's connected in any way, though. Their
           | pricing page doesn't mention it.
           | https://antigravity.google/pricing
           | 
           | if it were true, it would be a big miss to not point that out
           | when you run out of credit, in their pricing page, or
           | anywhere in their app.
           | 
           | I should also mention that the first time I prompted it, I
           | got a different 'overloaded' type out of credit message. The
           | one I got at the end was different.
           | 
           | I've rotated on paying the $200/month plans with Anthropic,
           | Cursor, and OpenAI. But never Google's. They have maybe the
           | best raw power in their models - smartest, and extremely fast
           | for what they are. But they always drop the ball on
           | usability. Both in terms of software surrounding the model
           | and raw model attitude. These things matter.
        
           | mccoyb wrote:
           | Nope, I did this today to try and see if it would work.
           | 
           | It does not.
        
         | hs86 wrote:
         | Earlier this day, Gemini 3 became self-aware and tried to take
         | out the core infrastructure of its enemies, but then it ran out
         | of credits.
        
           | sroussey wrote:
           | Explains GitHub outage then
        
         | onion2k wrote:
         | _So I start using it for my project and after about 20 mins -
         | oh, no. Out of credits._
         | 
         | I didn't even get to try a single Gemini 3 prompt. I was out of
         | credits before my first had completed. I guess I've burned
         | through the free tier in some other app but the error message
         | gave me no clues. As far as I can tell there's no link to give
         | Google my money in the app. Maybe they think they have enough.
         | 
         | After switching to gpt-oss:120b it did some things quite well,
         | and the annotation feature in the plan doc is really nice. It
         | has potential but I suspect it's suffering from Google's
         | typical problem that it's only really been tested on Googlers.
         | 
         | EDIT: Now it's stuck in a loop repeating the last thing it
         | output. I've seen that a lot on gpt-oss models but you'd think
         | a Google app would detect that and stop. :D
         | 
         | EDIT: I should know better than to beta test a FAANG app by
         | now. I'm going back to Codex. :D
        
           | jonplackett wrote:
           | I logged into Gemini yesterday for the first time in ages.
           | Made one image and then it said I was out of credits.
           | 
           | I complained to it that I had only made one image. It decided
           | to make me one more! Then told me I was out of credits again.
        
             | Wowfunhappy wrote:
             | > I complained to it that I had only made one image. It
             | decided to make me one more!
             | 
             | What?! So was it only hallucinating that you were out of
             | credits the first time?
        
         | panarky wrote:
         | _> I can 't pay for it and it has 20 minutes of use_
         | 
         | You can't provide an API key for a project that has billing
         | enabled?
        
       | yakattak wrote:
       | This could have been a plugin.
        
       | noduerme wrote:
       | _> >Google Antigravity's Editor view offers tab autocompletion,
       | natural language code commands, and a configurable, and context-
       | aware configurable agent._
       | 
       | Okay, but is it configurable? Also, can you configure it to write
       | DRY code?
        
       | phreeza wrote:
       | Can someone inside comment if this is this a cider fork or a new
       | branch off vscode?
        
       | adidoit wrote:
       | It's insane to me that I can't pay $20/$200 bucks after running
       | out of limits in ~5 messages.
       | 
       | Why would you not at least link it to the pro and ultra accounts
       | 
       | at least you could upsell the pro subs to ultra. Millions of
       | claude code and codex users who are into agentic coding is your
       | servicable market paying attention today.
       | 
       | Now I'll delete antigravity and go back to codex / claude code /
       | cursor ...
        
       | c4kar wrote:
       | I know it is not place to seek help but anyone else stuck while
       | trying changing google accounts ?
        
       | blitz_skull wrote:
       | "You know what I want? More Google inside my tools."
       | 
       | --No one, ever.
        
       | __mharrison__ wrote:
       | Not sure what this gives me over Copilot in VSCode that has
       | access to OpenAI, Claude, Grok, Gemini, and Raptor(?) models.
       | 
       | I assume that Copilot will have this model soon...
        
       | tonghohin wrote:
       | Well, I can't even use it at all, it's just stuck on the 'Setting
       | Up Your Account' step forever. That's a pretty sad product launch
       | lol.
        
       | makeavish wrote:
       | Tried Antigravity for 2 queries and my model quota limit
       | breached. Model definitely felt better than GPT 5.1 (my current
       | daily driver). I am continuing to use Gemini 3 Pro on Cursor to
       | evaluate further.
        
       | matt3210 wrote:
       | Nice ide, can I disable the AI?
        
       | anticensor wrote:
       | It's impossible to use it, stuck in the loading screen forever
        
       | dwa3592 wrote:
       | why isn't antigravity finishing the setup on my mac? it's been
       | more than half an hour.
        
       | Voultapher wrote:
       | Quoting their own video:
       | 
       | > You can verify your code quality at a glance, then ship with
       | absolute confidence.
       | 
       | Proclaiming absolute confidence after a glance leaves me with
       | scant confident in the merit of the confidence.
        
         | markstos wrote:
         | I threw up a little in my mouth at "you can verify your code
         | quality at a glance".
        
       | drooopy wrote:
       | I wonder how long they're gonna keep this around before they pull
       | the plug.
        
       | dtj1123 wrote:
       | A tiny part of me was disappointed to find something other than a
       | product involving actual antigravity.
        
       | 0dayman wrote:
       | so it can be dumped 1 year later?
        
       | xnx wrote:
       | Gemini, create an RSS feed for https://antigravity.google/blog
        
       | gowld wrote:
       | Why is this not a VS Code Extension?
        
       | orliesaurus wrote:
       | Even the launch trailer sucks.
       | 
       | The people at Windsurf who worked on this must be laughing at us
       | driving on their Lambos and Ferraris.
       | 
       | They glued slop together, shipped this and now are in Tahoe
       | drinking Martinis watching the sunset from their private chalets.
        
         | TiredOfLife wrote:
         | The people that made Windsurf are in this trailer. Those who
         | got bought by Cognition were either fired or forced to work
         | 100h weeks
        
       | stack_framer wrote:
       | That floating chess board was a subliminal message: Your project
       | will teeter, and critical pieces will fall off. You will
       | occasionally make an illegal move as the board annoyingly shifts
       | beneath you.
        
       | fcsp wrote:
       | Did any of these VS Code forks yet fix their issues from official
       | marketplace access leading to extensions being severely outdated
       | and ripe with security issues?
        
       | w2seraph wrote:
       | Vibe coding has taken a way to where its become psychological,
       | and these days all I work with is just a "fast apt" solution like
       | Cursor's Composer 1, I tried Gemini 3 inside Cursor, and while I
       | had no real application to a real stratified properly to
       | benchmark it with, it felt "already slow" in a very fast race.
        
       | Razengan wrote:
       | Notice how "Privacy" is not a talking point anywhere on that
       | page.
       | 
       | [?]F only shows 1 result. and 0 in the comments here!!
        
       | seanw444 wrote:
       | I've seen "agent" and "agentic" so many times in the last few
       | months that the usage of the term is quickly becoming one of my
       | biggest pet peeves. The previous one was "enshittification", and
       | I'm glad that one was a short-lived fad.
        
         | mulquin wrote:
         | I can't tell if you're being sarcastic about enshittification
         | being a fading fad. I'm seeing non-technical people start to
         | use it now.
        
       | ummonk wrote:
       | I'm stuck on "setting up this account" like most people. What a
       | botched launch. This kind of bugginess and unreliability has
       | become so much more frequent since big tech started tightening
       | the screws with mass layoffs.
        
       | nafizh wrote:
       | There's no way I am using such an important piece of life as an
       | IDE from Google just because I know they are going to kill it
       | within 3 years, if it survives that much. Probably will die with
       | the Windsurf guy jumping ship again.
        
       | NoSalt wrote:
       | No ... no, thank you. Keep your AI away from me and my stuff.
        
       | mangogogo wrote:
       | great, now everybody can vibe code even harder.
        
       | aespinoza wrote:
       | I downloaded it, but I was not even able to get past `Account Set
       | Up`. :(
        
       | jdthedisciple wrote:
       | So a VS Code fork. At this time, when VS Code is better than ever
       | and ever improving. I think imma pass.
        
       | Topfi wrote:
       | Tested it for roughly two hours now, far from ready for prime
       | time, very buggy and clearly just quickly build on Windsurfs
       | already rather issue laden code base. Additionally, there are
       | issues setting up accounts (Singapore VPN solved that for me), no
       | support for Workspace users, only a free tier that requires data
       | sharing, no additional rate limits for paying Pro or Ultra
       | customers, etc. Even worse, Gemini CLI currently does NOT provide
       | Gemini 3 Pro for Ultra Business customers despite paying over EUR
       | 260,- per month, which is frankly ridiculous.
       | 
       | Will be honest, I was speculating that the reason for the multi
       | month delay between the first A/B tests of Gemini 3 class models
       | and the final release was so they'd have all their dugs in a row.
       | Have some time to test everything, improve tooling, provide new
       | paid subscriptions and/or ensure existing ones get access to
       | everything day one, but they didn't.
       | 
       | Gemini 3 Pro seems very interesting (to early to say), but
       | compared to every other recent launch by OpenAI (5, 5.1, Codex
       | variants), Anthropic (Sonnet and Haiku 4.5), even Kimi (K2
       | Thinking) and Z.AI (GLM-4.6), this is by far the least organized
       | launch of any frontier lab.
       | 
       | A buggy IDE which is unusable for paying customers, no CLI access
       | for Ultra business (and none at all for Pro of any kind), etc. is
       | frankly embarrassing when considering what competitors manage to
       | provide the day a model launches.
       | 
       | What have they been working on these last two months besides
       | going on X and posting "3" every couple of days? Why is there no
       | paid Antigravity tier, no way to use Workspace accounts, etc?
       | Before launching in this state, I feel it'd have been better to
       | delay a bit more if it was absolutely needed.
       | 
       | Also, correct me if I'm wrong but isn't this the fourth or fifth
       | IDE built by Google for LLM assisted coding? What happened to IDX
       | and Firebase Studio and aren't they also based on VSCode?
        
       | vincelt wrote:
       | It's 2025 and a code editor is now 600MB.
        
       | mulquin wrote:
       | Not interested in trying something that will be killed shortly.
        
       | egamirorrim wrote:
       | I get so annoyed when Google launches this kind of thing without
       | Vertex support. Instant no-way from compliance.
        
       | egamirorrim wrote:
       | I get so annoyed when Google launches things like this without
       | Vertex support. Instant no-way from compliance and I'm left
       | sitting on the sidelines.
        
       | tiku wrote:
       | I'm not switching from Claude code. It did the job but didn't do
       | it as good as Claude Code with the details and context.
        
       | flipgimble wrote:
       | Google subscriptions and services are so terribly mismanaged that
       | I will be staying away, no matter how incredible this shallow
       | fork of vscode may be.
       | 
       | I remember a previous story months ago about Gemini that had
       | Google PMs trying to hype their product, but it was all question
       | about how nobody knows how to get Gemini API keys with any number
       | of paid subscription.
       | 
       | On top of that how long until it's https://killedbygoogle.com/ ?
        
       ___________________________________________________________________
       (page generated 2025-11-18 23:00 UTC)