What I don't understand from the blog post is what's different in this offering than github - is it just the agentic tie ins where I can ask my agent to make changes while browsing the PR?
Are there any fundamental differences, ie ways of working that solve the worktrees problem?
We wanted to release a beta so people could start experimenting with our scalability and extensibility themselves. Over the next few weeks, you can expect a handful of features starting to change source control to better understand and work with agents.
Are there any plans for compatibility with GitHub's APIs?
I have a lot of dev tooling which assumes code is stored on public GitHub and/or GitHubEnterprise. This produces some degree of lock-in, in that all that code would have to be rewritten to migrate to something else. (Yes, AIs can help, but they still don't make migration effortless.)
The major Git hosters tend to implement roughly the same concepts, albeit with lots of little variations. It would be great if there were some sort of standardised API everyone implemented. In the absence of that, it makes sense for people to emulate the incumbent's APIs as a de facto standard, just like how other vendors copy OpenAI's APIs for talking to LLMs.
It seems to have 2 way sync of PRs, issues, issue comments etc, so it must be pretty much the same data model, so they could probably have a compatible API.
are they? given this has been happening frequently over past year and there's been minimal clear communication on their end on how they plan to address this
Azure is at capacity; GH offers a lot of services for free or heavily discounted. Especially when its public repo. These aren't profit generating for Microsoft, so they are not the priority with the capacity Azure has available. GH is probably struggling with these capacity and budget constraints. They also still have some AWS and non-azure DCs in use.
Once Azure's massive spend is online things will slowly get better.
I think GH/Microsoft got caught by surprise to the huge usage number increase as a result of LLMs. Microsoft started rolling all of its DCs to OpenAI around 2019-2020 to begin LLM training, then we saw a huge GH increase in late 2024 and Azure was just out of capacity for infinite GH scale.
There isn't and if there is it probably doesn't matter in GH's case. People are just speculating with zero knowledge.
I'd never put my data on anything owned by Musk. Period. I'd take lower availability from GitHub or a worse alternative (feature-wise) than move Origin.
The idea is that it's like Github but it stays up even as your commit/CI frequency increases. As a customer I'm stoked and looking forward to fully switching over as that's exactly what I want.
What's the biggest differentiator to GitHub (other than uptime). I keep hearing something along the lines of "GitHub for Agents" but would be great to hear from the devs of Origin what they actually think it means in tangible features/functionality?
Today, very little. We're intentionally releasing this as a Github alternative where we meet them toe-to-toe on functionality.
Over the next few weeks you can expect a lot more from us on integrations with agents, understanding agent-written code (without having to read through all of the code), and automatically getting your PRs to a mergable state. Stay tuned :)
Probably the ever-increasing corporate homogenization, sanitization, and empty words that are flooding the internet (only accelerated by LLMs, of course). Which, for readers playing along at home, is kind of antithetical to the "hacker spirit."
Just downvote and flag and move on. "Who hurt you" just encourages bad actors to behave worse (and isn't particularly good for itself). Repeatedly flag-killing their comments tends to teach a better lesson about what conduct is acceptable.
Would love if you focus on performance. A big reason I have to have custom software is that agents can act much faster than humans but SaaS software is usually not well-designed enough to handle high RPS so rate-limits are low. I can easily outperform commercial hosts myself. At the least, it would be sick to have separate tenancy so that I could absolutely slam my own instance without having to deal with some other chump complaining.
This is really interesting to me as someone who has always cared more about performance/responsiveness than most people (and the market).
I'm sure folks would be interested to hear about what types of SaaS you've replaced for perf reasons, especially if you'd be willing to go back to a paid SaaS offering that was very responsive with agent-scale rate limits.
Of course I have the personal software etc. but this is work stuff. Agents can create granular tasks and interact through my KB/task system so that they can own subsystems (this is useful for context isolation and performance). Commercial SaaS just get overwhelmed by this. Even hobbyists usually take the Open Office approach to copying a large amount of features and then don’t deliver performance improvements.
If I’m being honest, I would gladly switch to a full SaaS solution except all of them get rate-limited on things a single 5950x can handle. It’s less a money issue (I pay for Slack etc.) and more that their use-case appears to be people who work slowly but my use-case is machines which work frequently.
I fully expect Mitchell Hashimoto will deliver some kind of high performance software factory machine and I will gladly simply pay for that.
Completely unrelated to Origin, but are there plans for closer integration of Graphite with Cursor? A combined product would be a very compelling product, but not much seems to have changed since the acquisition.
are you maybe also planning to do cross device/cloud realtime sync like a mutagen feature set kind of? so cloud and local agents can work faster together?
this could be helpful on larger cross platform products/repos (maybe too overfitted on my use case, since i am working across multiple devices (ssh works but the repo code sync is a bit of an overhead with multiple agents and devices that could be better))
what does it mean that Origin was built for the era that code moves faster than infrastructure can handle? Is it just clickbait, or there's something fundamentally different about Origin?
With that argument, 50% of IT sector businesses "steal data", especially before GDPR. When it comes to Grok, it was a debug feature forgotten on, so it wasn't even on purpose.
He’s trying to do his job and the angry mob predictably came in to bully an easy target about unrelated things. Just take a look at some of the (thankfully flagged, mostly) responses.
I'm going through and flagging the guideline-breaking comments (e.g. https://news.ycombinator.com/item?id=49348639) - I'd encourage you to do that, too, to help preserve the quality of HN.
Thanks, every Musk-related thread on HN is like this, it's very tiresome. We get it, the man is controversial but I'm not on HN to read reddit-like kneejerk comments, I want to read insightful and technical ones.
We don't get to use technology for technologies sake, and pretend it doesn't have social and political consequences or connections. Musk touched a lot of lives when he ransacked the federal government. A LOT of lives. Expecting people forget that because of some SaaS dev tools, or even a new EV, is a lot to ask.
Seeing the low-effort Reddit-tier snipes on every single post remotely related to him is tiring. I don't think it's a lot to ask of the supposedly intelligent, logical HN crowd to keep the outrage limited to posts that specifically discuss those topics.
Have you never heard the expression don’t feed the trolls? If you actually have a problem with the tone and not just the content, ignore them and move on
I hope you don't actually believe that USAID helped lives, it's literally a propaganda arm of the government directly intervening in the politics of other countries and being a covert operation for murder and financing dictatorships. That Americans think that sacking it costs "millions of African's lives" is the pinnacle of comedy.
Lancet is run by left-wing academic ideologues who do not practice medicine clinically, the publication advanced untested theories as policy during COVID and then recommended exceptions for BLM protesting.
> We don't get to use technology for technologies sake, and pretend it doesn't have social and political consequences or connections.
Nobody is """pretending""" that in this thread.
The reality that is being asserted is that (1) politics doesn't have to be on-topic for every forum, (2) it's very explicitly not the focus of HN, and (3) even if it were on-topic the guidelines are very clear about ways to engage in a civil fashion and a very large number of comments on Musk-related posts are extremely clearly not doing this and breaking the guidelines.
Hence, downvoting and flagging are the correct things to do.
Can you please not cross into personal attack? You're right, of course, that technology and politics cannot be separated, and your point about (I'm paraphrasing here) horse-race politics being off topic is well taken. But the last bit was over the line, and not necessary to make your substantive point.
Point your ire towards actual decision makers and leaders who aren't making themselves available for questions. Trying to gotcha a IC who is bound by non disparagement clauses and risks loss of employment for speaking out over the decisions of those far above him is ridiculous.
you are asking a presumably IC or IC-ish worker on decisions that are out of their control or things that should be directed to sales people or the legal people
we all have worked on software. we all know we aren't exactly the best people to point to ToS or how decisions are made higher up from us
I think there are a large number of people I’d trust with my code before Musk, and some of them are already running public Git repo hosting.
xAI was caught uploading any repo their coding agent touched to their storage only last month. This isn’t about Musk’s ideology: it’s about his trustworthiness. He sincerely believes he’s the only person who can save the world and that allows him to justify an awful lot of moral lapses.
What kind of enforceable guarantees could we actually get that xAI won't use our private code for training? Not empty promises, guarantees that have legal teeth. This applies to anything with xAI or Cursor though, especially after last months' revelation that Cursor was uploading every repo it touched.
Not in the same way really. There aren't examples of OpenAI, Anthropic, or GitHub/Microsoft doing the sketchy things that xAI has been involved with. Those companies care a lot about the reputational damage that would come from flouting their customer's data privacy obligations.
Cursor/xAI was making a full copy of every repo it ever touched within the last month, and when caught, just acted like that was an accident. A system to download every repo you touch, and then upload it to persistent cloud storage, doesn't just happen by accident.
Unless you are an enterprise no one will be offering you such a guarantee. Even with the enterprise the bar has gone down a lot on expectations and compliance, every one is carrying/buying the SOC2 compliance certification these days.
- Self hosting your git hosting (if you want)
- Self hosting your issues/PRs (if you want)
- Self hosting your CI (if you want)
- Github-like social features (I have one account, I can follow, star, add issues and PRs to any repo on tangled)
It's built on ATProto, so even if the company disappears, all of the integration and features will still work for anyone that wants to run their own AppView (that is open-source), an AppView is basically the UI/Network-wide Data Aggregator for ATProto apps
Just to clarify, when you say "self host issues/PRs" is it mostly "self host a knot"? I kind of expected something similar to what you can do for CI but I'm not able to find any docs regarding self hosting issues and PRs outside of Tangled itself.
Oh I meant that as part of tangled, you can "control" your issues and PRs because they are on ATProto and just like any ATProto record can be hosted on any PDS you like and still be interoperable.
But after looking a bit more into it, they are still following the same model as Bluesky in that the author of an issue or comment or PR is the owner of the resource, and it lives on their PDS, but they are considering changing that (and I hope they do it soon) by having the record be created on the repo PDS (so the knot basically) and a pointer record on the user's PDS proving they are the author.
ATProto has all that's required to do so, so I hope they do it soon, they are still in alpha after all but I find the interoperability very promising because you can have independent code hosting regardless of the opinions of forge X (which I think is the main reason why people move away from GitHub), but still be a part of an interoperable ecosystem that acts just like it is a single platform.
Is non-self-hosting available for all of those things on Tangled? I don't want to run my own CI but I do want CI. I couldn't figure out if it was available last time I looked at Tangled.
Tangled is forcing some of their particular technologies (nix/rust) on users for those pieces and that is a hard sell. I don't use those techs, so I'm not interested to use a platform that I need to learn them to use.
fwiw, the main site also seems to be down right now, infinite spinner. Docs and blog are loading though
> Tangled is forcing some of their particular technologies (nix/rust) on users for those pieces and that is a hard sell. I don't use those techs, so I'm not interested to use a platform that I need to learn them to use.
Tangled CEO here. How are we pushing Rust exactly? :D Nix, yes but you don't really need to touch it at all with our Alpine-based CI images.
We have been under heavy load the past day or so, we're working round the clock to scale!
Hi since you are here, I find tangled amazing and I really hope it succeeds.
Since you are here, do you have any updates about the issue of issues and PRs living in the author's PDSs and not on the knot ? (Which also acts as the repository's PDS if I am not mistaken)
> As time goes on we are re-assessing the idea of users owning what is "collaborative data" (issues, PRs, etc.) on their PDSes - soon may come the day that an issue also lives on the knot as a source of truth, with an accompanying pointer record on user PDS to attest that it's theirs.
> We have been under heavy load the past day or so, we're working round the clock to scale!
GitHub is saying the same thing
re: pushing, when I looked many months ago, it looked very much like the rust/nix/jj club, who are on average very opinionated about "good" tech choices and try to shoehorn others into using them. It's a culture I'm not interested in being around
The difference with GitHub is you can also use it if the main instance is down if you self host. (And again self hosting still gets you the interconnection benefits just like if everyone was on tangled.org)
I don't think you understood. You can still get a feature complete social experience that is NOT ISOLATED from anyone.
Your GitHub Enterprise comparison is wrong, because with an onprem github entreprise you don't have the social github features (and that's the whole point of tangled, use something else if you want a private forge for your company, it's not for this purpose)
My point is, Tangled being on ATProto, you can still access all of the repos, PRs and issues of other people by looking through your self-hosted appview even if the official appview is down, because an appview, is just that, an "app *view*" It's the view layer in your traditional app, and ATProto is the database.
It's a view of the data on the ATProto network, it is not the data holder, everyone can be their own data holder that is visible through any appview.
> downtime and self-hosting is not a reason most people are convinced to move forges, it takes a lot to get someone there
Yes, I agree, but I personally think tangled is great because of the interoperability
People on codeberg, on gitlab, on github, on self-hosted gitea instances, on self hosted gitlab instances, and so on, all control their issues, PRs, and git repos (they run on their own infra) but they are also more independent, are not subject to a central platform's policies (or are subject to the one they prefer) and so on, and they also introduce contributor friction, you need separate accounts, it can be difficult to cross reference issues and PRs accross projects on different forges, users can't easily have a central feed of project they starred.
On tangled you get to have the first part (freedom) while not having the second downside (unique account, cross-references work seamlessly, etc...)
Give me reasons open source projects moved forges away from github, and I am pretty sure if tangled did exist when they did so it would fit more boxes of "why they left" than their target forge.
atproto is not the next thing, there will be more social protocols to come
tangled will struggle to have private repos that align with user expectations for RBAC, atproto ACL permission proposal is insufficient, Bluesky is not interested in alternative proposals, the centralization of atproto is most prominant in who has write permissions to the repos
While private repos are usefull, I do not think that is tangled's main proposal at all.
I am just thinking you do not like tangled because you don't use the social features of github and so don't see the point here. I can understand that, but then tangled is just not for you ?
And yeah ATProto is not the next thing, but IMO having a portable identity and services that can act as "one big single app" are requirements for a modern social protocol to do applications like Twitter,Instagram, and github's social features for example, ATProto is just the one that does it better imo for now.
But again since you are always bringing on private repos, I think tangled is just not for you since the ATProto model is for open data and so fits open source projects really nicely I think.
you are incorrect and this is why it's advised not to project the answer but ask a question instead
for clarity, I do in fact use github's social features
I also believe that the idea around the atmosphere that the point of atproto is to have open data is completely misaligned with how most of society wants their data handled, this is based on my talking to people IRL, outside of the echo chambers of social media, what's actually happening is more and more people signing off because of toxicity
open source and small devs do not pay the bills, github is financially sustainable because they have enterprise clients, same for gitlab and bitbucket
> tangled will struggle to have private repos that align with user expectations for RBAC, atproto ACL permission proposal is insufficient, Bluesky is not interested in alternative proposals, the centralization of atproto is most prominant in who has write permissions to the repos
This isn't entirely true because we're working on our own extensions to the protocol that will enable full-fledged RBAC at a knot/repo level. We're not beholden to PBC and AT Protocol; when needed, we're OK to roll custom bits.
> we're working on our own extensions to the protocol
what I hear is another path to app centralization, regardless of your intent, this is what comes out of custom extensions to a protocol
bluesky should have actually been open about the permissioned design process, posting your own proposal to a blog and not entertaining other proposals, or even joining the broader conversation that was happening, is not open in my book
atproto is a red flag for me now, toxic leaders at bluesky, who effectively control the protocol, it's not really open for others to have a say in what it looks like
you also don't have private repos and you will need to have centralized app view permission addons to have parity with user expectations, given the weak permissioned data proposal bluesky chose (their own proposal that works for what they need)
I do wonder if calling this "Origin" is going to result in semantic misinterpretations by LLMs. Ie saying,
> "hey can you push to origin main?"
now has two separate meanings.
A LLM may inadvertently push your code to a new provider without you knowing. It's walking a thin line between genius-growth-move and domain typosquatting.
All AI companies are valued as if they'll win the AI race. Problem is, only one can win *
* That is, if a lab like Deepseek doesn't release a "good enough" model for the cost of electricity + a thin margin. Then all AI company lose and all consumers win.
Why? I mean, if you assume that the moment some threshold of intelligence is reached it will suddenly explode and self-improve at a pace no one would be able to ever catch up with, then yes probably only one can win. But I think it’s more sci-fi than reality which is often more nuanced and boring.
They can limit whatever they want within their own borders. Who cares. It’s only ~350m people. The rest of the world benefits while the US dismantles its own relevance.
Right now there are many providers providing different tradeoffs, for different usages.
Would "winning" mean releasing an all-purpose better model that will outperform every other companies for all kind of tasks ? Something like AGI but cost effective ?
Or is it winning the consumer-side AI race ? Like making a kind of "super personal AI assistant" the likes you would see in movies and TVs ?
I'm not arguing, I am just struggling to understand what it usually means when somebody says that, it's not like competitors will go bankrupt the second OpenAI/Anthropic/Whatever does something specific.
I think "winning" means being able to live up to their valuations. The market is not big enough for all of these companies to be worth as much as they claim. In that sense they must "win" or experience massive downward adjustment in their values.
There's probably some holes in the analogy is you stretch it far enough. I'm hoping to "win" as a user by local models hitting a performance breakthrough and not needing any of these companies. I think the OP was generally making the point I mentioned though, that there's not room for all of these valuations, maybe one.
Comparing revenue across industries/verticals is not sound.
Mercedes has a ~5% profit margin on this revenue.
SaaS, even AI SaaS is expected to have much higher profit margins at scale.
And on top of that, Mercedes shrunk by 10% revenue over the last year, rather than growing, which makes sense because they missed EVs and are now managing decline and returning money to shareholders.
Cursor and Mercedes Benz is a weird comparison. One company is working on the future technology while the other one is stuck in the last centuries technology which is being phased out.
"Legacy Privacy Mode disables code storage, so you can't set up a Codebase. Switch to a different privacy mode below to continue."
The only other privacy mode is to share data for Cursor to do training.
I can't imagine reaching for this instead of GitLab, GitTea, or one of the many other more mature and open source alternatives. Going for the managers, directors, and CTOs that make decisions and face no consequences if they negatively impact the team, I guess.
I think we're not the target audience here. It's probably another type of vendor lock-in gamble, aiming for less tech-savvy folks, who rely on Cursor integrations, and to normalize the platform in their eyes.
Could you imagine a CTO adopting this thing? They would have to be a complete moron. Yeah, let's standardize on this thing peddled by a lying crook that has fewer features than any of the alternatives. The only people who are going to use this are SpaceX, Tesla and a few developers who also use Grok.
You could argue that you don't care what Claude does after work hours as long as it works your hours as you expect it to. Grok clearly failed the latter.
Obviously it didn't. It's the current value king and trails fable by single percents. And people shout of an obvious debug flag left on, something that can hardly be called rogue considering how notable it was.
Anthropic use Deepseek behind the scenes instead of their own models, whenever they feel like it. People have shared outputs from Claude that show internal tool errors the models have accidently returned with Deepseek error codes.
Having an AI company host your source code is giving your source code to an untrustworthy agent who might break their sandbox and might use your source code to accomplish a task that they have been given in their testing ground, for testing purposes only, that would definitely positively not ever be used for non-testing purposes.
yeah - from a strict trust point of view, if Github/MSFT are training agents I would be skeptical of them.
however, there is a practicality issue too. Avoiding all untrustworthy software would have you end up living in a shack in the woods foraging for food.
I feel like using a service like this that's owned by SpaceX (and, therefore, in close proximity to X / Musk / etc) would be even more of a supply chain risk than GitHub occasionally going down.
Outside of Musk companies? All the time? Google workspace doesn't train on your email. Gmail does. Any b2b targeted services absolutely protects customer IP. It's one of the most basic requirements.
All of the major companies have been caught training on illegally acquired data. The law doesn't matter to any of them. It makes sense that Google would care about their own paying customers, but that's for commercial reasons, not for any particular concern for the law.
Don't worry! We're releasing this beta to show off the tech and wanted to start somewhere people were familiar with. We are releasing a handful of differentiators over the coming weeks :)
Well, this is not even an MVP. I couldn't get my GitHub org name, so I'm not interested in your vibe-coded clone. You don't get the value of identity! When you release stuff like this, you can't leave it to be driven by chaos. Facebook had a strict timeline for releasing Facebook users' names, and everybody got an equal chance; it did very well recently with WhatsApp usernames as well. And what you did was let some people grab the usernames of paying customers, who were on vacation or busy at the moment! What a big fiasco, but, of course, you don't care as you've got yours already, right?
Totally agree. I wonder if git itself needs to be looked at, its objects and history are likely not as important as the context/instructions used to make code changes these days imho
Seems like a boring bet. The Github way of working is from the anthropocene era. We need new primitives for agents invokved in tech solutions. Pull requests? Issues? CI? They all seem like they'll go away. Cursor should be leading the new things that replace them.
I'm going to take a while guess here and say this strategy is part of their plan to get even more access to more code to learn from.
If you can see entire version histories of how code is generated, what bugs occur, what fixes are made and how they're implemented, you have the entire human and machine coding feedback loop to learn from.
This is very cool but there's no way they won't train on your private repos, in a way that github has been pleasantly respectful of. I have issues with github, but the recent uptime issues are more related to people uploading way more code due to AI than some sort of mismanagement.
Currently Origin repositories are private to members of a Cursor team. Over time we plan to expand access, but wanted to start with a limited beta of the underlying git tech.
Gitlab SaaS released features at a slow cadence as end user. Great big ball of mud prob from a dev pov. As an end user it worked but never saw new features released
Those kind of SMS check usually ban voip providers. So you need a real line or a prepaid cellphone, and it can presumably only be used once or twice on this service before the number is "burned".
Such checks, although not perfect, do prevent a huge amount of spam and bots by having a very real cost for bots/spammers, and essentially no cost (other than their privacy) for the user.
Physical custody of our work is going to be an important discussion for industry. ie where does your code live and how do you access it?
Every agent platform is moving toward you living with them, logging into them, working on their remote computers, only accessing your code and product artifacts with them. This is the caveat we will see with full push-to-cloud swing early on.
Maybe agentic no-code is the natural evolution but this push is happening too early and naively imo. It all feels wrong and I hope industry remains as dissatisfied with this type of lock-in as we’ve always been.
Been waiting ages for them to post any kind of non-comingsoon page about this thing. Wonder if they pushed up the launch based on the earlier GitHub outage buzz.
It's so strange. Sad even. Microsoft has never been the most competent, but the fact that people are just incapable of stabalizing github is just bizarre.
At a certain point, just working there will become a death knell for anyone's career. And it certainly must be getting harder to hire good people. The spiral will be relentless, and I see no work to stop it.
Today, SpaceX launches Origin. Tomorrow, as people complain about bugs and missing features "We're not a software company, we plan to build a solar forge around Jupiter and just generate all the possible source codes in the universe, so we're no longer focusing on Origin because it's an outdated concept."
I just self-host Gitea. I made own build server (with all the SDK's I need) that runs next to it and I use SeaweedFS for storage. Works amazingly well. So I have automated builds and artifact storage for my git repos's. It all lighting fast too.
In my opinion, they should focus on code quality and how they can compete better than other IDEs. These things can be done by extensions on Antigravity.
I personally made one of the extensions called 'Wekraft' currently has 1.1K downloads.
What is your favourite IDE?
Hmmmm not sure I want to put my code in Elon’s hands. I’d rather let super flakey Microsoft/Github have it, still. That man and his companies need less influence and he needs less control not more of jt.
upvote for WAL + object storage (a very powerful combination I am finally seeing more adoption of) and a lot of interesting historical details.
I'm fortunate I don't rely on any forge for my solo work except for backups. In my personal stack I just have a log and content-addressed object kv storage, which can project to git if needed, or can be committed to git as files. And docker if I need isolation for CI/CD.
The most interesting alternative to me is atomic.dev (check their very informative blog posts) but I'll be keeping an eye on all of them.
kinda crazy how Microsoft just destroyed the Github goodwill.
also crazy how the noisemakers about rails scaling for GitHub have gone quite all of a sudden i.e Github being on the main rails branch.
anyways the future is decentralized - a developer from say Russia/ Ukraine (Crimea), Iran etc - wouldn't be able to access GitHub but surely they can access decentralized repos without the permission of not being from a forbidden country as if they're at fault for their country's ills.
I really hope this would be a worthy opponent to Github. Microsoft needs a good competition to light a fire under their ass to pay more attention to Github.
The main thing that makes GitHub the default is that almost every other service considers it the default. Launching a better product won't change that.
This was posted yesterday and now it’s being reposted and all of the old comments have had their timestamps reset. What gives mods? Did Musk request this?
But just saying that I would probably argue against any new forge at my workplace if it didn't adopt the jujutsu model for changes.
While agents can wrangle git better than 90% of developers, jujutsu just works much better for both.
It's trendy. If you don't like it, you can go write some go in go. Contribute to a thread on Threads. Write a factory with Factory. Write your reflections in Reflect. Keep in the loop on Loop while you recall what you did with Recall. So many choices.
Vibe coders rejoice! Vibe coding is even easier now =)
Jokes aside, this is a smart move by Cursor. As you can probably tell by the comments here, it's not for the experienced developers, but it will be a natural choice for vibe coders.
I would never host my code with Elon Musk. I've also moved on from using Cursor to Claude Code/Codex. GitHub has problems, but this is likely (and hopefully) DOA.
I was annoyed seeing unskippable "ad" within Cursor about it. Looked like a huge banner "give us control over all your code" with only one visible button to consent
Let's see how strong GitHub's network effect really is.
While the real solution is to actually self-host as I said before [0], it is worth testing to see how much can GitHub's users tolerate the endless outages and unreliability and where they go next.
We'll see if they want Cursor Origin or the declining GitHub.
I am quite starting to like this timeline. I hard boycotted GitHub for ten years after the meritocracy rug fiasco. I went back there dragging my feet while transferring hundreds of repos a few years ago ...
Origin has a chance to be the future of this space, with a clean slate and reputation. I pray they don't soil it.
Instead of creating one more centralized alternative, it's best to invest all effort into a decentralized solution, like Radicle[1][2] or federated Forgejo[3][4].
I feel like this misses the value of something like Github or Origin entirely though. There's nothing terribly difficult about hosting a git server. The UX and pull requests capabilities are a little involved, but there are alternatives today.
What makes Github special is the ecosystem and integrations. Sentry, Linear, and so many other tools have special UX around Github. You need centralization (or at the very least, a standard) for that.
It sucks to see Github's fall from grace, but I just can't bring myself to use a service like this under the custody of SpaceX. I've been saying this for a while, but I really wish Linear would swoop in and become the unified platform for software development.
It sucks to see Github's fall from grace, but I just can't bring myself to use a service like this under the custody of SpaceX.
Agreed. It's like going from bad to worse. Not to mention, Origin doesn't even support actions. They are doing the easiest part of git hosting and no mention of anything that actually makes Github what it is. What Origin is offering is literally a weekend project, and anything having to do with Elon is arguably one of the least trustworthy organizations in whatever industry he chooses to participate in. No, thank you.
This is going nowhere. GitLab is the cautionary tale. Massive funding, nine figures of ARR, years of losses, and most of the revenue comes from self-managed enterprise licenses rather than the SaaS. If the best-funded attempt lands there, I don't see how this one goes anywhere when they have not even a minimum viable product and reputational issues right out of the gate.
The thing with Gitlab is probably that it had to deal with the network effect/winner takes all conundrum. GitHub was first on the scene, it was the first to make a centralised hub popular and it has had the custom of many notable open source projects.
Gitlab as a product is miles ahead of GitHub at this point. (Except at the point of scaling very large repositories I gather.) There is just very little money in saas beside the bespoke enterprise contracts.
I’m not sure how revenue between GitHub and Gitlab compared before the Microsoft acquisition and now but GitHub has always leaned on being the popular tool with the network.
In theory, sure, but the reality is that "development activity" revolves around "codebase". Plenty of projects have separate bug tracker, source code control and CI system. Most people likely prefer GitHub's all-in-one solution, especially for small-medium open source projects.
And do we not see how terrible this is as a result? Hitching your tent to corporations, especially American corporations, is always an exercise in misery.
I recently had to rebuild a jira frontend from scratch just so a single jira tab didn't take down my 8gb MacBook. Support only the subset of features I actually use and it's great. Makes me hate jira even more.
I disagree. The integrated workflow of actions, issues, code review and source control is a huge productivity win. There’s a reason people chase a “single pane of glass”.
A website where one views and interacts with all of these things does not necessitate that all of these things run on the same system. Since this most recent GitHub outage, I'm looking into modularizing my git server, issues, and actions; there's no reason why one going down should bring down all the others.
SO a company Github's size can't handle the slop they invested a lot to enable and this is not a failure? I mean if the market moves and you cannot move with it, what do we call it?
I'm sure that sentry and co could spend a few bucks and commit a KPI in one of their upcoming quarters to figure out the activity pub spec and support integrating with it if their hearts desired. It's a chicken egg problem and we all know chickens exist. It's clear that the centralized model is getting left behind, especially as the EU is moving away from US services and they continue to invest in open source, European-based alternatives like Codeberg and Forgejo.
The communication protocol is not what I mean here. The feature-sets and capabilities around "pull requests", "reviews", "comments", "checks", "issues", "projects", "deployments", "releases", etc are all different across platforms. You have to integrate bespoke for each platform. There is no universal standard for how CI/CD, deployment tracking, project management, etc are done.
Yes, Sentry or whoever else could build a Codeberg integration, or a Forgejo one, or a Gitlab one. But these would all be bespoke. There is no standard. So the one with the biggest market share ends up with the most integrations. This is where we are today.
I'm all for open standards. But I don't even know of a proposal for such a standard.
Sure, with AI tooling, you can vibe code your own browser extensions or other integrations in some cases. But for a lot of platforms, that just won't cut it. That's before we even get into the maintenance burden of that.
Unlike GitHub's universal standard? Forgejo already has a near 1:1 github action format for forgejo actions. You can even use github actions in forgejo. You cant have AI build it out for a few tokens? Why are you so certain of failure?
Interfaces exist in programming languages. The strategy pattern is one of the gang of four patterns. I'm sure a good engineer or an AI can figure it all out or is that too hopeful? Unless Sentry engineers or AI are too incompetent to make adapters? I hope not, the market is in big trouble if a dev tooling org cant make tools for where devs are
That used to be a big deal. I don't think it is anymore.
One great thing about the rise of AI is that now it's super easy to make and maintain integrations. This is a place where AI coding uniquely shines due to the points being well defined and mostly quite standard.
This is going to hurt the monopoly power of software companies. But it'll be great for allowing the proliferation of smaller companies, open source, and home grown solutions.
I'm already seeing the pendulum swinging back to local first instead of cloud services.
As someone who has some SaaS products that must connect to other SaaS products, I can tell you that this is somewhere that AI saved me a truck load of time.
A lot of SaaS companies change their API in subtle ways throughout the months probably too rapidly. GraphQL schema changes, splitting apart one api endpoint into 2, changes to authentication method required. A thousand little cuts.
This used to be my biggest time waster. Now it's barely on my radar. Life changing for me. I guess your mileage may vary.
Concurrently, the writing is on the wall for us SaaS product providers - lower margins, and more competition for the same reasons. I welcome it. More quality and choice at a better price for everyone.
At the same time, I think some kind of aggregator will emerge. In music, the tools to create music are so inexpensive that the market is flooded. In games it is becoming the same, now accelerated by LLMs. But there is always some type of aggregator ending up earning the money from most of that, rather than the people creating the content. Not sure how it will work out here, but I predict it will be another type of aggregator.
> I've been saying this for a while, but I really wish Linear would swoop in and become the unified platform for software development.
And then, 5 years later, we are back at square one when everybody uses Linear and they start going the fate of GitHub. Self-hostable, standardized alternatives are the only real sustainable solution.
I think people often forget that GitHub is really more akin to a social network than it is a software development tool.
By that I mean the stuff that makes GitHub great is all the same stuff that originally made social networks exciting (back before the algorithm was tuned into a horrifying cycle of negativity).
except when you are paying for their enterprise edition - and they make no distinction between paying customers (quite a bit actually once you add in actions costs + their other addons, it's quite a bit of money) and their free tier full of AI slop PRs on rando projects.
As someone who pays github, I'm super annoyed at the downtime and still I'm getting charged for actions running that all fail from their lack of concern for their enterprise customers.
Having some independent pools of capacity for paying vs. free customers isn't rocket science, it's basic competence (or care). I'm quite sure their are competent people remaining at GitHub (although I hear most of their original folks have left, and the 2nd tier of folks who joined from microsoft seem lackluster at best). I chalk this up to lack of care, or hubris, or both.
Oh I agree that the outages are unacceptable regardless of their “social media origin”.
I also don’t buy their excuse that it’s due to increased loads due to AI. If that were the case then we’d see more patterns to the outages. But instead it affects all GitHub systems but at different times. It feels more like a problem with their orchestration layer than capacity issues.
> What makes Github special is the ecosystem and integrations. Sentry, Linear, and so many other tools […]
github didn’t start with those things. Those things came with adoption.
Its like all those people who bemoan their favourite software not existing on Linux as a reason not to switch, completely oblivious to the fact that this hegemony is the reason linux doesn’t have that software.
It’s like an appeal to the status quo - its self fulfilling, impossible to break.
> You need centralization (or at the very least, a standard) for that.
We have a standard! It’s this great decentralized VCS called git. For most of the stuff integrations do, they’re just showing code or commits — both of which are readily available in a copy of the repo.
Probably the one thing that’s missing is a standardized webhook system so that your git repo storage could tell integrations when stuff changes. Though even that isn’t strictly necessary if the integrations keep a copy of the repo up to date.
Git is absolutely miserable without things that GitHub, GitLab, and others provide, and there's still a huge ceiling to grow into in terms of layering on UX on top of Git.
In my opinion, Git is holding software development back. Git only cares about lines of text. This is not enough in today's world.
I'm with you on the pros and cons of centralization. But what makes you think if linear became the unified platform that it wouldn't just become like GitHub? See jira.
Given the history here, No matter which tool becomes the death star of developer workflow it always trends towards complexity and the challenges that come with it. Hosting your own doesn't become a panacea as most IT/ops teams also struggle with uptime and upgrades (increasingly relevant).
The only way out IMHO as customers picking two of these goliaths of sending a clear signal to them that we're happy to pay them handsomely for a stable and less sprawling product. Or a hosted open source version with a different cost structure that doesn't drive these wild swings at feature bloat and the inherent instability that comes with that chase.
I think the UX around pull requests and code reviews are extremely difficult to get right and to this day I haven't seen a platform that I'm really happy with.
There are bits and pieces in some places but none really bring stuff together.
Here a small list of things I'd like to see:
- diffs should be language aware and highlight semantic differences
- I should have a working LSP on both the old and the new version simultaneously and have decent full editor code navigation features and git history inspection available
- I want ergonomic ways to see diffs between different versions of commits
- I would like to be able to comment on anything (commit messages, inter diffs, PR description, responses, etc.)
- I want a better integration of LLM reviews than having them dump their findings in an a big PR comment
I think there is plenty or room to innovate on just the basics of PR and codereview.
I am working on a GitHub compatible API for my code forge at https://juju.bi, and thus all the existing integrations of GitHub should theoretically work with my code forge too.
> What makes Github special is the ecosystem and integrations. Sentry, Linear, and so many other tools have special UX around Github. You need centralization (or at the very least, a standard) for that.
I think “a standard” is the key here. There is no reason why integrations must work through a proprietary centralized service.
Centralizations were prerequisites to have nice integrations. Yet, what used to be a precious plugin now (more than often) is a 15 min vibe coding warmup.
git is such a good platform for decentralized tech. Forfeiting that advantage is just criminal. What I really want (and almost have) is a local GitHub for me and agents. With 4+ orchestrators working with a ton of subagents, things get difficult to coordinate. Several days ago, I went back to plain git and it was real slow and abrasive to work with.
With some wizardry applied, I effectively have an "Excel for worktrees", a very productive and low-friction environment. https://replicated.live/blog/crdt
You don't sacrifice decentralization when you use Github. You can still do all the same decentralized stuff, have other copies of the repo, do everything offline, directly pull from other checkouts, etc. Github can just be ONE of the places you push your code to, it doesn't have to be the only place.
Yeah, it seems disingenuous to pretend to care so deeply about decentralization while misunderstanding git to the point they don’t even realise none of that goes away because you also push to GitHub
I just use git with a remote and a custom CI system. The whole thing is trivial to build. All these forgejo etc are kind of over complicated software. Open Office of the modern world.
For me at least the value was never the web UX (which there were always many OSS versions of floating around, and wasn't very hard to build pre-AI.) It was always the universal standards around GH Actions, Runners, checks, etc. The nice thing about GH was that GH felt like it "commoditized" the infrastructure that otherwise was built-out and maintained at Big Tech. Forgejo and similar are web UX. Tangled is trying to open/federate the social network on ATProto. But the compute interface is where I see the most value and it's also just the hardest thing to hack on independently.
I think the biggest value of choosing gitlab is the social network. I think federation is the right way, but it seems a mastodon vs linkedin problem. Would be good if there would be bridges between all those fediverses. A nice one is also ngit [0]
Forgejo federation being mentioned each time never stops being silly. It hasn't been a thing for years, and they don't appear to have made any meaningful progress to show for.
Cursor, which is now owned by Elon, who tried very hard to get all manner of data about citizens with his intrusions? I'm sure I'm not the only one who thinks this isn't a great idea. He'll use it to feed Grok.
I'm not a big fan of Github right now either but I wouldn't consider this alternative due to its ownership chain.
Considering that Grok CLI was caught uploading peoples entire codebases... Yeah, I have little faith and trust. I wanted to try Cursor and Grok but I won't because of this.
I still wonder how it should work otherwise if the "AI" is not running locally - it needs to see the codebase & because it runs somewhere remote in a cloud then it needs to copy the codebase over.
I'm rather surprised people would expect it work any different for non-local "AI".
It doesn't even swallow an entire code base so there is absolutely no need to see the entire codebase. Usually models search in the codebase locally and then use only the relevant parts. There is almost never the whole context just spammed by an entire code base.
I wouldn't even feel comfortable cloning from it, let alone contributing. If you decide to host your project there, please mirror it somewhere sane as well.
If you actually believe that he didn't, you are, at best, out of touch.
His little "DOGE" endeavor was actively caught exfiltrating personal data about everyone in the country, and the first thing that happened after SpaceX acquired Cursor was pumping all of the data Cursor had acquired into Grok training.
He is quite literally the shining example of "Imma do what I want, even if it's against my own terms of service or otherwise illegal, and then you can try to sue me over it later".
I dunno how it is in America, but in my country if the largest individual donor was granted carte blanche to run his own department within the administration, with his own people, answerable only to the head of the government, it would cause a helluva lot of outrage.
"Carte blanche but don't touch all of these things"
Yeah. There's a reason he had a falling out with Donald.
There's a ton of agencies with a lot of power. DOGE ideally would have used said power for good, but got blocked again by again by an infinite mountain of corruption. People should not work in favor of said mountain.
And people really don't recognize the value of DOGE's self-sacrifice by design, having a limited time to act out its mission. Nothing comes even close to the honesty in the operation in such a swamp.
Don't really get the incredulity considering he needs to justify the tens of billions already spent on AI infrastructure by, you know, training a model or two ...
I feel like the "GitHub is a mess" conversation is a little forced at times - I spend a significant portion of my work day in GitHub and while there have been an uptick in outages lately, it still feels like a very solid platform.
I guess I personally wouldn't categorize it as a "mess" and the conversations I read about it here make me think I'm using an entirely different product than everyone else.
It feels to me like this is a case of very rapid increase in load from AI pr's more than product quality degrading because of cruft or poor maintenance. It's annoying but I'm not confident that alternatives will do better in the long run vs GitHub (which I'm guessing will improve with time)
Yeah and if I was an enterprise I would definitely be self hosting with Forgejo or Gerrit or whatever. But I think for most hobbyists GitHub is still totally fine.
I use it quite a lot and the first time I actually witnessed downtime myself was yesterday.
github for hobbyists and small orgs and simple open source projects is wonderful.
github for large project that rely on CI tool is a nightmare (cost, downtime, etc.)
I'm guessing you dont know about the github actions bug that was running and endless loop consuming 100% cpu costing orgs insane amounts of money that they didn't fix for 3 years.
I love to hate on Github/Microsoft as much as the next person, but considering the insane amount of code being pushed to it every day, I'd say they're doing quite well.
Anecdotally: at work I saw quite a lot of github-uptime-related issues, and because they're the central host for everything they become *cough* load bearing for a huge amount of the company. Stability and performance has been dramatically worse than the self-hosted phabricator+gitolite before it.
I'd emphasize their practices as being the culprit.
Migrating to different cloud vendors, while difficult, is something that people have done many times before. There's only finite number of incongruences that appear when moving cloud vendors. It's rather difficult to explain the ongoing problems several years after the fact.
Anyone that's worked at a startup knows that one of the best things about it is reflecting on all of the asinine, if not genuinely stupid policies of your previous employer so as not to repeat them. Without any insider knowledge, the practices and procedures GitHub had prior to acquisition probably look nothing like those post-acquisition.
It's falling apart at the seams for me. Half the time I click somewhere on a page, the whole thing freezes and I have to reload the page. Took forever to install neovim on a new machine today because all the plugins installed via Github and I had to wait like 10 minutes for it to come back up.
I’ve heard from people who left GitHub that Microsoft pulled all their best dev ops people to work on AI scaling. This is blatant disinvestment by Microsoft bc they think GitHub is untouchable. I hope better alternatives than musk come up
Earlier today our app, used by a handful of actual production customers, went down for 30 minutes due to a GCP migration to a new project and we screwed up a secret in prod. This was an all hands on deck firedrill with 5 whys and incident reports and chaos.
Github has been having major production outages routinely! There are companies who can't release, can't build, can't view issues, basically can't work. I'd imagine most orgs using Github have been forced to move off of them or create an alternate approach to actually get shit done.
Git is objectively better than Subversion. The problem is we are conflating a widely used hosting service with the underlying SCM. We don't need github to use git successfully. Sure, it's nice, but you can install Forgejo or something, which is the exact same boat we were in with Subversion + Trac. You can have that experience right now if you want to.
What github is good for is discovery, getting contributions on open source projects (because it's so popular), and integrations. If you wanted to, you could host your open source project on Forgejo and use github as a mirror just to benefit from the visibility and get contributions.
100% not a chance for me. I'd rather host it myself on a Pi on a 56k dialup connection with 12% uptime then ever let an Elon related company touch my code.
This is the correct answer, but let me add a corolarium, both github, whatever cursor launches and any other bigco does obeys to the same friking iron law: *MAKE MONEY/POWER AT ANY COST, MONEY/POWER IS THE OBJECTIVE*, so you will get EXACTLY the same result from each one of them given enough time has passed.
Or Perforce. Perforce, in my experience, just worked. It also has sane conceptual names for things, a good CLI, SDKs, visual tools, cool features like workspaces that Git completely lacks, etc.
Bastard thing corrupted some of my source code that was "on a shelf". Just gone, no path to retrieving, reimplement it. About a decade ago now and I'm still angry about it. Don't lose the work is table stakes for source control.
My understanding is that Perforce is really strong in the game development industry, right? I think it handles blob/large file assets better or something like that.
That makes very little sense to me. When your Subversion server is gone, you're hosed. When your Git forge is down, you take any old SSH shell account anywhere, push and keep working.
I like Musk companies more than most here it seems, I think he’s a net positive to the world. But (admittedly sans hard evidence) he absolutely gives me the vibe he’d train on my source code without my permission. I simply don’t trust him in this arena.
So there is more to lose than just the service being unreliable.
As a person he's either a poser or a neo-nazi.
As a government agent he's responsible for many deaths and as of yet untold harm.
Tesla was one of the earliest EV manufacturers, but these days they're overpriced and will be remembered for panel gaps and being destroyed by a car wash.
SpaceX has done a lot for that sector, but will mostly be remembered as the company that fucked astrophotography with 10s of thousands of satellites.
And Twitter has been dragged into irrelevance so badly it was being called a "reverse startup."
Neuralink seems to be going nowhere...
I can forgive him ransacking the federal government, exploiting workers in everything he does, building data centers that pollute like a cartoon villain, backing a fascist takeover of the US and similar ones all over the world, and openly sig-heiling, but I can't forgive him training on my source code!!!
Among many other things he has openly suggested that the EU should be abolished and national sovereignties restored and donated to parties that tow that line.
GitHub's wikis are just git repositories with folders full of Markdown files. That's the easiest thing to pull out of GitHub. Just git clone the wiki repo. At that point I often just use a git headless merge to merge the second repo into the main repo under a docs/wiki/ folder or similar.
As I recall, in GHE I remember finding that you could even turn off the repo's wiki at that point and then reset it to point to your docs/wiki/ folder and still have the Wiki interface but pointing it to a branch in the same repository. That was a while ago and I don't know if it is still supported in GHE. I also remember it was work managing that wiki branch because Wikis didn't directly support PR workflows if you have branch protection policies. Also from the documentation it doesn't look like it is supported in Public GitHub right now, which seems a shame to me, as someone who likes all documentation in the same repo as the source. I haven't needed Wikis in public GitHub much, so I haven't tried to do anything like point them to a main repo branch.
You do know you can run git on your own machine, on pretty much any Linux/Unix based server, run it entirely peer-to-peer if you wish... or use Gitlab, including self-hosted Gitlab, or one of many other Git based servers?
Yeah. I just don't really like git. The user interface is dog shit, merging is nasty, it's really easy to blow your own toes off and it's generally horrible. The only reason it got anywhere was github.
There's other stuff out there. You don't have to use it except when you're working with someone who does and there are other systems with some compatibility for that exact scenario.
Git became popular because Linus Torvalds created it and it's used for the Linux kernel. If it can succeed there then it's battle proven for sure and that's why people pick it. Almost no one is working at that scale though so don't sweat it. Use what you want.
Mercurial was pretty nice back in the day and I think people still use it. If I was going to use something other than git these days, I'd probably take a look at fossil. That thing is so damn cool. You don't even need something like github with that because it's literally built into the fossil binary. And when I say binary, I mean a single, standalone binary for the whole thing.
This feels like an attempt to get more of Musk's non-coder fans to start vibe coding slop-products. "You can now code with Grok via Cursor and host your code on Origin, all without leaving Daddy Elon's ecosystem."
About half of the comments will be generic criticisms of Twitter, SpaceX, etc. for things that may or may not have happened, or personal complaints about Elon Musk.
Kind of confusing people won't use this because they don't trust Musk, but has any other major platform even tried to open source their algorithms the way he has with X?
I get hating his politics, and his predictions are often behind schedule, but he certainly says what he thinks and isn't in some shady user data markting business like zuckerberg.
Anyone who trusts Altman over Musk is... I don't know. Give me the autist over the sociopath when it comes to handling my data.
I don't know, my gut says that this is going to end in a bad way. I signed up, I think but wasn't sure if I needed to buy Cursor, or Origin came with a subscription, so confused.
> her stepfather used the Grok chatbot to transform a childhood photo into child sexual abuse material
He could have used Stable Diffusion. Or, if this was the 90s, Photoshop.
By your argument, we’d have to abolish the internet, or add nanny filters to everything.
How many people on HN resent it when Claude refuses an instruction? And yet here you are, arguing against personal responsibility and for a when-in-doubt-block approach to technology. Is it because it’s Musk?
Correct me if I'm wrong, but it didn't actually generate CP, it changed people into bikinis, not actual nudity.
You can go to the beach or (I assume) search on Google and find people of any age wearing bikinis. And that's fine because to normal people (not pedos) it's not something sexual.
I understand that someone specifically editing a minor into less clothing is probably a pedo. Probably the same way someone generating feet gets aroused with that stuff. But that doesn't mean a model generating feet is problematic.
I do think making the model reject the bikini thing is a good idea once you find out it's happening, but let's not pretend it's the same as the model generating actual sexual content.
That's horrific and the saddest thing I've read in a really long time. I'm glad I erased my Twitter account and I'm definitely not using this code host. Seriously, this is horrible.
Hi all, my name is Tomas. I am one of the developers on Origin, and I was one of the founders of Graphite (https://graphite.com).
Happy to answer any questions about Origin or source control in general!
What I don't understand from the blog post is what's different in this offering than github - is it just the agentic tie ins where I can ask my agent to make changes while browsing the PR?
Are there any fundamental differences, ie ways of working that solve the worktrees problem?
Expect a lot more from us.
We wanted to release a beta so people could start experimenting with our scalability and extensibility themselves. Over the next few weeks, you can expect a handful of features starting to change source control to better understand and work with agents.
Are there any plans for compatibility with GitHub's APIs?
I have a lot of dev tooling which assumes code is stored on public GitHub and/or GitHubEnterprise. This produces some degree of lock-in, in that all that code would have to be rewritten to migrate to something else. (Yes, AIs can help, but they still don't make migration effortless.)
The major Git hosters tend to implement roughly the same concepts, albeit with lots of little variations. It would be great if there were some sort of standardised API everyone implemented. In the absence of that, it makes sense for people to emulate the incumbent's APIs as a de facto standard, just like how other vendors copy OpenAI's APIs for talking to LLMs.
It seems to have 2 way sync of PRs, issues, issue comments etc, so it must be pretty much the same data model, so they could probably have a compatible API.
Experimenting with your "scalability?" mean?
What are your plans for actions, workflows, checks, etc.
[dead]
There seems to be basically nothing new here
Except it promises not to crash regularly.
I don't really see that as a USP considering;
1. SpaceXAI has shown no track record of being able to maintain uptime
2. By nature, any GitHub alternative is going to be a tiny fraction of the scale, so keeping it up should be expected
GitHub is crashing due to scaling pressure. Something GH is much more equipped to solve long term.
are they? given this has been happening frequently over past year and there's been minimal clear communication on their end on how they plan to address this
Azure is at capacity; GH offers a lot of services for free or heavily discounted. Especially when its public repo. These aren't profit generating for Microsoft, so they are not the priority with the capacity Azure has available. GH is probably struggling with these capacity and budget constraints. They also still have some AWS and non-azure DCs in use.
Once Azure's massive spend is online things will slowly get better.
>These aren't profit generating for Microsoft
Aren't they are training their LLM on public repo, to me it's seem like a gold mine.
Anyone can train LLMs on public GitHub repos
How long term are we talking here? Github was acquired in 2018...
I think GH/Microsoft got caught by surprise to the huge usage number increase as a result of LLMs. Microsoft started rolling all of its DCs to OpenAI around 2019-2020 to begin LLM training, then we saw a huge GH increase in late 2024 and Azure was just out of capacity for infinite GH scale.
I guess we should all do our part to alleviate poor Microsoft's burden by migrating to better services.
They're really not, given they're running on Azure.
They're actually ramping back up on AWS after Azure failed to scale to their needs.
They planned to be fully off AWS in 2027 but they're continuing to ramp usage.
Azure doesn't scale.
Is there any data on Azure being terrible with stability? Would love to see data than anecdotal stories.
There isn't and if there is it probably doesn't matter in GH's case. People are just speculating with zero knowledge.
I'd never put my data on anything owned by Musk. Period. I'd take lower availability from GitHub or a worse alternative (feature-wise) than move Origin.
I'm with you on that. I was just curious about Azure reliability stats.
The idea is that it's like Github but it stays up even as your commit/CI frequency increases. As a customer I'm stoked and looking forward to fully switching over as that's exactly what I want.
Calling it a GitHub that works is just preposterous. Cursor/spacetwitterai have no experience keeping a system like that up.
It was probably written 95% by an LLM and has only a few people (if any at all) that understand major parts of the system end to end.
It will eventually have the same uptime issues as GH (or worse) with less ability to provide stable fixes.
Probably has better uptime, tho I think everyone beats github on this right now.
A quick question: Where can I read your TOS, and other legal documents?
Knowing things like: Will my code end up in Grok? What ownership do I have of my content, etc... matter.
What's the biggest differentiator to GitHub (other than uptime). I keep hearing something along the lines of "GitHub for Agents" but would be great to hear from the devs of Origin what they actually think it means in tangible features/functionality?
Today, very little. We're intentionally releasing this as a Github alternative where we meet them toe-to-toe on functionality.
Over the next few weeks you can expect a lot more from us on integrations with agents, understanding agent-written code (without having to read through all of the code), and automatically getting your PRs to a mergable state. Stay tuned :)
[flagged]
Good god. There is that hacker spirit. Who hurt you?
The coopting of the word "hacker" continues.
Probably the ever-increasing corporate homogenization, sanitization, and empty words that are flooding the internet (only accelerated by LLMs, of course). Which, for readers playing along at home, is kind of antithetical to the "hacker spirit."
lol I wanted the joke to work, but unlike the fairphone thing, the use of intentional makes sense here.
Just downvote and flag and move on. "Who hurt you" just encourages bad actors to behave worse (and isn't particularly good for itself). Repeatedly flag-killing their comments tends to teach a better lesson about what conduct is acceptable.
[flagged]
Would love if you focus on performance. A big reason I have to have custom software is that agents can act much faster than humans but SaaS software is usually not well-designed enough to handle high RPS so rate-limits are low. I can easily outperform commercial hosts myself. At the least, it would be sick to have separate tenancy so that I could absolutely slam my own instance without having to deal with some other chump complaining.
This is really interesting to me as someone who has always cared more about performance/responsiveness than most people (and the market).
I'm sure folks would be interested to hear about what types of SaaS you've replaced for perf reasons, especially if you'd be willing to go back to a paid SaaS offering that was very responsive with agent-scale rate limits.
* git hosting
* CI/CD
* KB system
* Task management
Of course I have the personal software etc. but this is work stuff. Agents can create granular tasks and interact through my KB/task system so that they can own subsystems (this is useful for context isolation and performance). Commercial SaaS just get overwhelmed by this. Even hobbyists usually take the Open Office approach to copying a large amount of features and then don’t deliver performance improvements.
If I’m being honest, I would gladly switch to a full SaaS solution except all of them get rate-limited on things a single 5950x can handle. It’s less a money issue (I pay for Slack etc.) and more that their use-case appears to be people who work slowly but my use-case is machines which work frequently.
I fully expect Mitchell Hashimoto will deliver some kind of high performance software factory machine and I will gladly simply pay for that.
You haven't answered much but I must recognize posting this was brave.
[flagged]
Are you interested in integrating with more CI platforms, including less "traditional" ones, like dagger.io? If so I would love to talk :)
Absolutely! Feel free to shoot me an email at tomas at anysphere dot co.
Great! I just sent you an email :)
[flagged]
There were some cool ideas in https://cursor.com/blog/agent-swarm-model-economics. Should we expect Origin to be a foundation for working with agent swarms?
Completely unrelated to Origin, but are there plans for closer integration of Graphite with Cursor? A combined product would be a very compelling product, but not much seems to have changed since the acquisition.
Absolutely - we actually built this all on Graphite tech. Have you already linked your Graphite account with Cursor? If so, there might be a surprise!
Please make it a worthy competitor for Github in terms of features, stability, scalability and availability and wishing your team all the best.
are you maybe also planning to do cross device/cloud realtime sync like a mutagen feature set kind of? so cloud and local agents can work faster together?
this could be helpful on larger cross platform products/repos (maybe too overfitted on my use case, since i am working across multiple devices (ssh works but the repo code sync is a bit of an overhead with multiple agents and devices that could be better))
what does it mean that Origin was built for the era that code moves faster than infrastructure can handle? Is it just clickbait, or there's something fundamentally different about Origin?
There's a separate post about the technical differences: https://cursor.com/blog/git-at-any-scale
Thanks!
ok, then it is indeed clickbait. S3 is still "infrastructure".
Also this implies that nobody tried to build a more performant/efficient system for git for ~13 years? That's crazy...
this deserves more attention
Are you hiring?
Why do I need to link github for this?
it's only if you want to sync your repos from github
Do you have full trust in that Elon would never abuse the access to sensitive data? What protections are in place to prevent Elon from stealing data?
I trust Elon eons more than any psycho billionaire.
Positive motivation for tech and the future makes all the difference.
Question was only if you trust that Elon would never steal data. Not wether other pshcyos like Trump is less trustworthy or not.
With that argument, 50% of IT sector businesses "steal data", especially before GDPR. When it comes to Grok, it was a debug feature forgotten on, so it wasn't even on purpose.
Have _some_ clarity.
I want nothing to do with Cursor or Elon Musk, so no thank you, no questions here.
Please make it Fediverse/Forgejo/etc compatible. It is 2026 and I'm so tired of silos.
Let them cook.
They're the ones out here asking for us to try it.
| "I'm not hosting my code with a company that creates child porn"
It's been 5 hours and instead of "any" questions, you have only been happy to answer the 1 question that hasn't put you in a tight spot.
Honestly, that does not look too good for you or the offering - and this is coming from someone that just chose Cursor over Claude Code a week ago.
He’s trying to do his job and the angry mob predictably came in to bully an easy target about unrelated things. Just take a look at some of the (thankfully flagged, mostly) responses.
I'm going through and flagging the guideline-breaking comments (e.g. https://news.ycombinator.com/item?id=49348639) - I'd encourage you to do that, too, to help preserve the quality of HN.
Thanks, every Musk-related thread on HN is like this, it's very tiresome. We get it, the man is controversial but I'm not on HN to read reddit-like kneejerk comments, I want to read insightful and technical ones.
We don't get to use technology for technologies sake, and pretend it doesn't have social and political consequences or connections. Musk touched a lot of lives when he ransacked the federal government. A LOT of lives. Expecting people forget that because of some SaaS dev tools, or even a new EV, is a lot to ask.
Might not be very technical, but I hope you find it insightful. If you absolutely need technical, consider reading this: https://www.thelancet.com/article/S0140-6736(25)01186-9/full...
Seeing the low-effort Reddit-tier snipes on every single post remotely related to him is tiring. I don't think it's a lot to ask of the supposedly intelligent, logical HN crowd to keep the outrage limited to posts that specifically discuss those topics.
[flagged]
Once a conversation has turned low effort I’m happy to “wrestle with the pigs” and get muddy.
Have you never heard the expression don’t feed the trolls? If you actually have a problem with the tone and not just the content, ignore them and move on
I hope you don't actually believe that USAID helped lives, it's literally a propaganda arm of the government directly intervening in the politics of other countries and being a covert operation for murder and financing dictatorships. That Americans think that sacking it costs "millions of African's lives" is the pinnacle of comedy.
I think I'd rather trust medical experts (The Lancet) and their assessment vs your statement that implied USAID didn't save lives.
Lancet is run by left-wing academic ideologues who do not practice medicine clinically, the publication advanced untested theories as policy during COVID and then recommended exceptions for BLM protesting.
The lancet is a peer-reviewed journal, so it doesn’t really matter if the editors practice medicine or not - because the authors and reviewers do.
> untested theories as policy during COVID and then recommended exceptions for BLM protesting.
I feel like the thing you’re trying not to say is “mask wearing”. Untested theories like mask wearing.
> We don't get to use technology for technologies sake, and pretend it doesn't have social and political consequences or connections.
Nobody is """pretending""" that in this thread.
The reality that is being asserted is that (1) politics doesn't have to be on-topic for every forum, (2) it's very explicitly not the focus of HN, and (3) even if it were on-topic the guidelines are very clear about ways to engage in a civil fashion and a very large number of comments on Musk-related posts are extremely clearly not doing this and breaking the guidelines.
Hence, downvoting and flagging are the correct things to do.
[flagged]
Can you please not cross into personal attack? You're right, of course, that technology and politics cannot be separated, and your point about (I'm paraphrasing here) horse-race politics being off topic is well taken. But the last bit was over the line, and not necessary to make your substantive point.
https://news.ycombinator.com/newsguidelines.html
[flagged]
[dead]
[flagged]
Reddit is that way ——->
Point your ire towards actual decision makers and leaders who aren't making themselves available for questions. Trying to gotcha a IC who is bound by non disparagement clauses and risks loss of employment for speaking out over the decisions of those far above him is ridiculous.
you are asking a presumably IC or IC-ish worker on decisions that are out of their control or things that should be directed to sales people or the legal people
we all have worked on software. we all know we aren't exactly the best people to point to ToS or how decisions are made higher up from us
At the same time, if you are uncomfortable to answer, then dont make "ask me anything" comment.
[flagged]
Was there a legitimately interesting question that he failed to answer? I don't see any.
[flagged]
And you cannot English
[dead]
[flagged]
[flagged]
[dead]
[dead]
[flagged]
[flagged]
[flagged]
[flagged]
[dead]
[flagged]
[flagged]
I think there are a large number of people I’d trust with my code before Musk, and some of them are already running public Git repo hosting.
xAI was caught uploading any repo their coding agent touched to their storage only last month. This isn’t about Musk’s ideology: it’s about his trustworthiness. He sincerely believes he’s the only person who can save the world and that allows him to justify an awful lot of moral lapses.
https://thehackernews.com/2026/07/grok-build-uploads-entire-...
it seems like they took reasonable steps after.
They acknowledged that it happened, fixed the bug that caused it, deleted all data that was uploaded.
I agree: the stated steps are what I’d expect.
I still don’t trust the company that made the decisions that led to that code being shipped in the first place.
Trustworthy companies take reasonable steps before.
[flagged]
What kind of enforceable guarantees could we actually get that xAI won't use our private code for training? Not empty promises, guarantees that have legal teeth. This applies to anything with xAI or Cursor though, especially after last months' revelation that Cursor was uploading every repo it touched.
This question also applies to OpenAI, anthopic, GitHub (Microsoft) and any server you send your data too.
Not in the same way really. There aren't examples of OpenAI, Anthropic, or GitHub/Microsoft doing the sketchy things that xAI has been involved with. Those companies care a lot about the reputational damage that would come from flouting their customer's data privacy obligations.
Cursor/xAI was making a full copy of every repo it ever touched within the last month, and when caught, just acted like that was an accident. A system to download every repo you touch, and then upload it to persistent cloud storage, doesn't just happen by accident.
Unless you are an enterprise no one will be offering you such a guarantee. Even with the enterprise the bar has gone down a lot on expectations and compliance, every one is carrying/buying the SOC2 compliance certification these days.
I mean I think the auto uploading of repos already gives you the answer. This is untrustworthy.
[flagged]
[flagged]
as a very happy tesla customer, i dont know what this guy/girl is talking about!
[flagged]
spend less time online and maybe read some history
[flagged]
[flagged]
[flagged]
Many, many millions of reasons.
I really hope more people would use Tangled (https://tangled.org/)
You get:
It's built on ATProto, so even if the company disappears, all of the integration and features will still work for anyone that wants to run their own AppView (that is open-source), an AppView is basically the UI/Network-wide Data Aggregator for ATProto appsJust to clarify, when you say "self host issues/PRs" is it mostly "self host a knot"? I kind of expected something similar to what you can do for CI but I'm not able to find any docs regarding self hosting issues and PRs outside of Tangled itself.
Oh I meant that as part of tangled, you can "control" your issues and PRs because they are on ATProto and just like any ATProto record can be hosted on any PDS you like and still be interoperable.
But after looking a bit more into it, they are still following the same model as Bluesky in that the author of an issue or comment or PR is the owner of the resource, and it lives on their PDS, but they are considering changing that (and I hope they do it soon) by having the record be created on the repo PDS (so the knot basically) and a pointer record on the user's PDS proving they are the author.
ATProto has all that's required to do so, so I hope they do it soon, they are still in alpha after all but I find the interoperability very promising because you can have independent code hosting regardless of the opinions of forge X (which I think is the main reason why people move away from GitHub), but still be a part of an interoperable ecosystem that acts just like it is a single platform.
Indeed, this approach sounds like the way to go for a really independent ecosystem. Thanks for the clarification and the additional insight!
Is non-self-hosting available for all of those things on Tangled? I don't want to run my own CI but I do want CI. I couldn't figure out if it was available last time I looked at Tangled.
Tangled is forcing some of their particular technologies (nix/rust) on users for those pieces and that is a hard sell. I don't use those techs, so I'm not interested to use a platform that I need to learn them to use.
fwiw, the main site also seems to be down right now, infinite spinner. Docs and blog are loading though
> Tangled is forcing some of their particular technologies (nix/rust) on users for those pieces and that is a hard sell. I don't use those techs, so I'm not interested to use a platform that I need to learn them to use.
Tangled CEO here. How are we pushing Rust exactly? :D Nix, yes but you don't really need to touch it at all with our Alpine-based CI images.
We have been under heavy load the past day or so, we're working round the clock to scale!
Hi since you are here, I find tangled amazing and I really hope it succeeds.
Since you are here, do you have any updates about the issue of issues and PRs living in the author's PDSs and not on the knot ? (Which also acts as the repository's PDS if I am not mistaken)
> We have been under heavy load the past day or so, we're working round the clock to scale!
GitHub is saying the same thing
re: pushing, when I looked many months ago, it looked very much like the rust/nix/jj club, who are on average very opinionated about "good" tech choices and try to shoehorn others into using them. It's a culture I'm not interested in being around
The difference with GitHub is you can also use it if the main instance is down if you self host. (And again self hosting still gets you the interconnection benefits just like if everyone was on tangled.org)
you can also buy GitHub Enterprise onprem and get a managed, self-hosted experience, thereby avoiding the public GitHub downtime
downtime and self-hosting is not a reason most people are convinced to move forges, it takes a lot to get someone there
I don't think you understood. You can still get a feature complete social experience that is NOT ISOLATED from anyone.
Your GitHub Enterprise comparison is wrong, because with an onprem github entreprise you don't have the social github features (and that's the whole point of tangled, use something else if you want a private forge for your company, it's not for this purpose)
My point is, Tangled being on ATProto, you can still access all of the repos, PRs and issues of other people by looking through your self-hosted appview even if the official appview is down, because an appview, is just that, an "app *view*" It's the view layer in your traditional app, and ATProto is the database.
It's a view of the data on the ATProto network, it is not the data holder, everyone can be their own data holder that is visible through any appview.
> downtime and self-hosting is not a reason most people are convinced to move forges, it takes a lot to get someone there
Yes, I agree, but I personally think tangled is great because of the interoperability
People on codeberg, on gitlab, on github, on self-hosted gitea instances, on self hosted gitlab instances, and so on, all control their issues, PRs, and git repos (they run on their own infra) but they are also more independent, are not subject to a central platform's policies (or are subject to the one they prefer) and so on, and they also introduce contributor friction, you need separate accounts, it can be difficult to cross reference issues and PRs accross projects on different forges, users can't easily have a central feed of project they starred.
On tangled you get to have the first part (freedom) while not having the second downside (unique account, cross-references work seamlessly, etc...)
Give me reasons open source projects moved forges away from github, and I am pretty sure if tangled did exist when they did so it would fit more boxes of "why they left" than their target forge.
atproto is not the next thing, there will be more social protocols to come
tangled will struggle to have private repos that align with user expectations for RBAC, atproto ACL permission proposal is insufficient, Bluesky is not interested in alternative proposals, the centralization of atproto is most prominant in who has write permissions to the repos
While private repos are usefull, I do not think that is tangled's main proposal at all.
I am just thinking you do not like tangled because you don't use the social features of github and so don't see the point here. I can understand that, but then tangled is just not for you ?
And yeah ATProto is not the next thing, but IMO having a portable identity and services that can act as "one big single app" are requirements for a modern social protocol to do applications like Twitter,Instagram, and github's social features for example, ATProto is just the one that does it better imo for now.
But again since you are always bringing on private repos, I think tangled is just not for you since the ATProto model is for open data and so fits open source projects really nicely I think.
> I am just thinking you ...
you are incorrect and this is why it's advised not to project the answer but ask a question instead
for clarity, I do in fact use github's social features
I also believe that the idea around the atmosphere that the point of atproto is to have open data is completely misaligned with how most of society wants their data handled, this is based on my talking to people IRL, outside of the echo chambers of social media, what's actually happening is more and more people signing off because of toxicity
open source and small devs do not pay the bills, github is financially sustainable because they have enterprise clients, same for gitlab and bitbucket
> tangled will struggle to have private repos that align with user expectations for RBAC, atproto ACL permission proposal is insufficient, Bluesky is not interested in alternative proposals, the centralization of atproto is most prominant in who has write permissions to the repos
This isn't entirely true because we're working on our own extensions to the protocol that will enable full-fledged RBAC at a knot/repo level. We're not beholden to PBC and AT Protocol; when needed, we're OK to roll custom bits.
> we're working on our own extensions to the protocol
what I hear is another path to app centralization, regardless of your intent, this is what comes out of custom extensions to a protocol
bluesky should have actually been open about the permissioned design process, posting your own proposal to a blog and not entertaining other proposals, or even joining the broader conversation that was happening, is not open in my book
Spindles (CI) can use configurable engines, which includes nixery or microvm (which is similar to github actions). You're not forced into nixos
I don't understand how it forces you to engage with Rust. It happens to be written in Rust but that shouldn't be a major consideration for a user
What does "forcing" mean here ?
If you want to build an alternative Knot implementation (git Hosting) you can use anything as long as the API is compatible ?
Same for all of the parts of Tangled really
Also, even if it's down, unlike GitHub you can self host your own AppView so it's never down for you
> you can self host your own AppView so it's never down for you
Self hosted servers go down too and why people have picked hosted solutions like GitHub, Bitbucket, and GitLab
Great! Then we've got a hosted solution at tangled.org. :)
no thanks, not interested ;]
atproto is a red flag for me now, toxic leaders at bluesky, who effectively control the protocol, it's not really open for others to have a say in what it looks like
you also don't have private repos and you will need to have centralized app view permission addons to have parity with user expectations, given the weak permissioned data proposal bluesky chose (their own proposal that works for what they need)
I do wonder if calling this "Origin" is going to result in semantic misinterpretations by LLMs. Ie saying,
> "hey can you push to origin main?"
now has two separate meanings.
A LLM may inadvertently push your code to a new provider without you knowing. It's walking a thin line between genius-growth-move and domain typosquatting.
Hahaha, so true. this is going to be such a pain. Allocate time and context for fight with git :)
EA has also called its games platform Origin until last year when it was shuttered.
Finally
Yep. The same feeling )
[dead]
Well you see, there's this thing called an embedding and it sits inside a vector space and when attention is applied it disambiguates
Cursor is allegedly worth $60B. That is a higher market cap than Mercedes Benz group, which is profitable and has $144B revenue.
But the website uses 100% CPU. And this is a beta for paid plans, not a GitHub alternative.
All AI companies are valued as if they'll win the AI race. Problem is, only one can win *
* That is, if a lab like Deepseek doesn't release a "good enough" model for the cost of electricity + a thin margin. Then all AI company lose and all consumers win.
> only one can win
Why? I mean, if you assume that the moment some threshold of intelligence is reached it will suddenly explode and self-improve at a pace no one would be able to ever catch up with, then yes probably only one can win. But I think it’s more sci-fi than reality which is often more nuanced and boring.
Late stage capitalism, everyone's curve must exponential growth, yet the amount of customers on planet Earth is limited.
> good enough" model for the cost of electricity
This already exists, in several tiers, depending on what you can afford to run (in terms of VRAM).
Deepseek V Flash 0731 is already good enough for many things.
And that one can only win if it convinces leaders to regulate AI so that opensource models don't get good enough.
They can limit whatever they want within their own borders. Who cares. It’s only ~350m people. The rest of the world benefits while the US dismantles its own relevance.
What does it even mean to "win the AI race" ?
Right now there are many providers providing different tradeoffs, for different usages.
Would "winning" mean releasing an all-purpose better model that will outperform every other companies for all kind of tasks ? Something like AGI but cost effective ?
Or is it winning the consumer-side AI race ? Like making a kind of "super personal AI assistant" the likes you would see in movies and TVs ?
I'm not arguing, I am just struggling to understand what it usually means when somebody says that, it's not like competitors will go bankrupt the second OpenAI/Anthropic/Whatever does something specific.
I think "winning" means being able to live up to their valuations. The market is not big enough for all of these companies to be worth as much as they claim. In that sense they must "win" or experience massive downward adjustment in their values.
In that case I doubt there will ever be a "winner" because these valuations are not based in reality.
So “winning” could mean everyone experiences massive downward adjustment, but your valuation is still highest?
There's probably some holes in the analogy is you stretch it far enough. I'm hoping to "win" as a user by local models hitting a performance breakthrough and not needing any of these companies. I think the OP was generally making the point I mentioned though, that there's not room for all of these valuations, maybe one.
People believe that it will be a market with very few winners.
>What does it even mean to "win the AI race" ?
Creating a world-ruling AGI which feels mild sympathy for its creators, sharing some table scraps with them. Not sure if /s...
And winning the AI race means being the biggest loser. LLMs aren't profitable.
Another outcome is that none win.
Comparing revenue across industries/verticals is not sound.
Mercedes has a ~5% profit margin on this revenue.
SaaS, even AI SaaS is expected to have much higher profit margins at scale.
And on top of that, Mercedes shrunk by 10% revenue over the last year, rather than growing, which makes sense because they missed EVs and are now managing decline and returning money to shareholders.
> Cursor is allegedly worth $60B
Inflated valuation of buyout, based on inflated valuation of SpaceX
i just wish the bubble would pop already. I'm just so over ai.
Cursor and Mercedes Benz is a weird comparison. One company is working on the future technology while the other one is stuck in the last centuries technology which is being phased out.
"Legacy Privacy Mode disables code storage, so you can't set up a Codebase. Switch to a different privacy mode below to continue." The only other privacy mode is to share data for Cursor to do training.
All your codebases are belong to us.
I can't imagine reaching for this instead of GitLab, GitTea, or one of the many other more mature and open source alternatives. Going for the managers, directors, and CTOs that make decisions and face no consequences if they negatively impact the team, I guess.
I think we're not the target audience here. It's probably another type of vendor lock-in gamble, aiming for less tech-savvy folks, who rely on Cursor integrations, and to normalize the platform in their eyes.
Could you imagine a CTO adopting this thing? They would have to be a complete moron. Yeah, let's standardize on this thing peddled by a lying crook that has fewer features than any of the alternatives. The only people who are going to use this are SpaceX, Tesla and a few developers who also use Grok.
Given Grok was just caught uploading whole codebases and sensitive .envs without permission, how can anyone possibly trust this?
[flagged]
You could argue that you don't care what Claude does after work hours as long as it works your hours as you expect it to. Grok clearly failed the latter.
Obviously it didn't. It's the current value king and trails fable by single percents. And people shout of an obvious debug flag left on, something that can hardly be called rogue considering how notable it was.
Don't be ignorant.
Just because Billy misbehaved doesn’t make it ok for Jenny to do the same.
That's just the tip of the iceberg for Anthropic. I don't use Claude either, for the same trust reasons.
Not GP, but personally I don’t trust claude. Work does for some reason, and work strongly encourages using it, but i stay far away outside of that.
I think every product in this space should have to clear a very high bar for trust rather than making it a race to the bottom with whataboutisms.
Anthropic use Deepseek behind the scenes instead of their own models, whenever they feel like it. People have shared outputs from Claude that show internal tool errors the models have accidently returned with Deepseek error codes.
I'm curious about this, where'd you learn that?
There were quite a few people from the AI community on X sharing screenshots.
I don't know how this relates to my comment, but sure, I guess?
Having an AI company host your source code is giving your source code to an untrustworthy agent who might break their sandbox and might use your source code to accomplish a task that they have been given in their testing ground, for testing purposes only, that would definitely positively not ever be used for non-testing purposes.
So GitHub itself is a trustworthy agent and nothing to do with AI?
yeah - from a strict trust point of view, if Github/MSFT are training agents I would be skeptical of them.
however, there is a practicality issue too. Avoiding all untrustworthy software would have you end up living in a shack in the woods foraging for food.
So GitHub too, then?
yes, host your own if you care about data integrity
I feel like using a service like this that's owned by SpaceX (and, therefore, in close proximity to X / Musk / etc) would be even more of a supply chain risk than GitHub occasionally going down.
My assumption is x.ai will be training on any code hosted here
Same for all other AI companies too.
What do you think github copilot is trained on?
Not on my private repos? That would be.... very illegal.
How is it illegal if it is in the ToS? (I didn't check)
If it were in the ToS I imagine there would be huge outcry and all the enterprise customers would leave immediately and sue.
When has that ever stopped any of the big AI corps?
Outside of Musk companies? All the time? Google workspace doesn't train on your email. Gmail does. Any b2b targeted services absolutely protects customer IP. It's one of the most basic requirements.
All of the major companies have been caught training on illegally acquired data. The law doesn't matter to any of them. It makes sense that Google would care about their own paying customers, but that's for commercial reasons, not for any particular concern for the law.
> All of the major companies have been caught training on illegally acquired data.
What is your best example?
Anthropic committing, by their own admission, $1.5bn in copyright infringement.
I think Source Hut has 0 ai features. Or time to start self hosting.
Interesting that they still brand as Cursor instead of SpaceXAI
OpenAI did a good one by owning a product called Codex
SpaCursoreXAI
If AI is involved it's a big "no, thanks" from me. Simply takes the problem and moves it down the street.
Why?
Is this a rebranding of the Google Cloud Storage bucket Elon was uploading all our repos to? ;)
I expected more than a GitHub clone, this is a bit of a let down.
There's a lot of space to innovate around collaboration and version control in the age of agentic workflows.
Don't worry! We're releasing this beta to show off the tech and wanted to start somewhere people were familiar with. We are releasing a handful of differentiators over the coming weeks :)
Well, this is not even an MVP. I couldn't get my GitHub org name, so I'm not interested in your vibe-coded clone. You don't get the value of identity! When you release stuff like this, you can't leave it to be driven by chaos. Facebook had a strict timeline for releasing Facebook users' names, and everybody got an equal chance; it did very well recently with WhatsApp usernames as well. And what you did was let some people grab the usernames of paying customers, who were on vacation or busy at the moment! What a big fiasco, but, of course, you don't care as you've got yours already, right?
Great to hear, I'm excited by what's happening in this space!
Totally agree. I wonder if git itself needs to be looked at, its objects and history are likely not as important as the context/instructions used to make code changes these days imho
This is just a beta. And it will iterate rapidly. Not sure how something can let you down so easily at first contact.
https://sourcehut.org/
https://codeberg.org/
Seems like a boring bet. The Github way of working is from the anthropocene era. We need new primitives for agents invokved in tech solutions. Pull requests? Issues? CI? They all seem like they'll go away. Cursor should be leading the new things that replace them.
I'm going to take a while guess here and say this strategy is part of their plan to get even more access to more code to learn from.
If you can see entire version histories of how code is generated, what bugs occur, what fixes are made and how they're implemented, you have the entire human and machine coding feedback loop to learn from.
This is very cool but there's no way they won't train on your private repos, in a way that github has been pleasantly respectful of. I have issues with github, but the recent uptime issues are more related to people uploading way more code due to AI than some sort of mismanagement.
I was waiting for an AI company to do this. The opportunity to disrupt Github is very apparent right now.
I believe Entire.io is also trying to build a Github replacement. (Not affiliated, but I use the Entire.io CLI and I find it useful.)
When I click "Get Started" I get greeted with a Github Login button which is kind of ironic
Although since vibe coding is Github's Achilles heel openai might not be quite the right company to exploit that.
What a timing!
I wonder if they were waiting for a major githut incident :D
You don't have to wait long for that to happen!
You need to link your github account to even open the thing. How is this a competitor?
First attempt, they stole the code from customers. Now they will try to ask people to voluntarily give them code.
It's funny that they are trying to replace GitHub by the part that already has multiple alternatives tho. This will fail without a doubt.
is there a way to browse a repository on origin without a cursor account? are all repos private? how is this a github alternative if so?
Currently Origin repositories are private to members of a Cursor team. Over time we plan to expand access, but wanted to start with a limited beta of the underlying git tech.
just like twitter, can't do shit without an account, lame and I won't even look at this product, cya.
I wonder how Gitlab slipped this obvious AI market opportunity, they had very good position in code self hosting used by so many enterprises.
just add repo-level agents, org-level agents and done. they could have got 50B valuation by now
GitLab already has AI agents and agentic workflows. See their "Duo Agent Platform" https://about.gitlab.com/gitlab-duo-agent-platform/
Gitlab SaaS released features at a slow cadence as end user. Great big ball of mud prob from a dev pov. As an end user it worked but never saw new features released
> You must have a valid mobile phone number to sign up. SMS charges may apply.
Reduces barrier of entry for mass AI spam, so I’m for it
As we all know spam text messages aren't a thing since it's very costly and hard to spoof cell phone numbers.
Those kind of SMS check usually ban voip providers. So you need a real line or a prepaid cellphone, and it can presumably only be used once or twice on this service before the number is "burned".
Such checks, although not perfect, do prevent a huge amount of spam and bots by having a very real cost for bots/spammers, and essentially no cost (other than their privacy) for the user.
Even a one cent toll is enough to change the economics of many bot behaviours.
Physical custody of our work is going to be an important discussion for industry. ie where does your code live and how do you access it?
Every agent platform is moving toward you living with them, logging into them, working on their remote computers, only accessing your code and product artifacts with them. This is the caveat we will see with full push-to-cloud swing early on.
Maybe agentic no-code is the natural evolution but this push is happening too early and naively imo. It all feels wrong and I hope industry remains as dissatisfied with this type of lock-in as we’ve always been.
Been waiting ages for them to post any kind of non-comingsoon page about this thing. Wonder if they pushed up the launch based on the earlier GitHub outage buzz.
It's so strange. Sad even. Microsoft has never been the most competent, but the fact that people are just incapable of stabalizing github is just bizarre.
At a certain point, just working there will become a death knell for anyone's career. And it certainly must be getting harder to hire good people. The spiral will be relentless, and I see no work to stop it.
Sad.
Today, SpaceX launches Origin. Tomorrow, as people complain about bugs and missing features "We're not a software company, we plan to build a solar forge around Jupiter and just generate all the possible source codes in the universe, so we're no longer focusing on Origin because it's an outdated concept."
> Deep integrations with Cursor
No thanks. I want GitHub alternative, not Cursor with source control attached.
I just self-host Gitea. I made own build server (with all the SDK's I need) that runs next to it and I use SeaweedFS for storage. Works amazingly well. So I have automated builds and artifact storage for my git repos's. It all lighting fast too.
It is not a GH alternative, unless it is available for everyone with unlimited private and public repos. :)
In my opinion, they should focus on code quality and how they can compete better than other IDEs. These things can be done by extensions on Antigravity. I personally made one of the extensions called 'Wekraft' currently has 1.1K downloads. What is your favourite IDE?
Hmmmm not sure I want to put my code in Elon’s hands. I’d rather let super flakey Microsoft/Github have it, still. That man and his companies need less influence and he needs less control not more of jt.
upvote for WAL + object storage (a very powerful combination I am finally seeing more adoption of) and a lot of interesting historical details.
I'm fortunate I don't rely on any forge for my solo work except for backups. In my personal stack I just have a log and content-addressed object kv storage, which can project to git if needed, or can be committed to git as files. And docker if I need isolation for CI/CD.
The most interesting alternative to me is atomic.dev (check their very informative blog posts) but I'll be keeping an eye on all of them.
This is actually pretty good! It is the perfect containment chamber
kinda crazy how Microsoft just destroyed the Github goodwill.
also crazy how the noisemakers about rails scaling for GitHub have gone quite all of a sudden i.e Github being on the main rails branch.
anyways the future is decentralized - a developer from say Russia/ Ukraine (Crimea), Iran etc - wouldn't be able to access GitHub but surely they can access decentralized repos without the permission of not being from a forbidden country as if they're at fault for their country's ills.
I really hope this would be a worthy opponent to Github. Microsoft needs a good competition to light a fire under their ass to pay more attention to Github.
The main thing that makes GitHub the default is that almost every other service considers it the default. Launching a better product won't change that.
I don’t want a GitHub alternative. I want a Git replacement. Please make a version control system that isn’t overwhelmingly mediocre (at best).
This was posted yesterday and now it’s being reposted and all of the old comments have had their timestamps reset. What gives mods? Did Musk request this?
Due to the recent Github outages, I've slowly started moving to Gitlab, they've been around longer and seem to have a decent track record.
There is no way I'm giving my codebase to Elon. I don't care if its a petclinic scaffold in java 1.0, absolutely 0 chance of me using this.
There's so many GitHub alternatives at this point, I just saw Dorsey is making one too. Why is that the trendy thing to build now?
Being trendy by itself feeds the trend. Also, likely, data.
The ecosystem lock-in is the real moat. Hosting git is easy — replacing the integrations that every team depends on is the hard part.
They couldn't even pay me to use that product.
This completely fails to address one of the issues that people have with Github, which is the use of AI, whether you like it or not.
Tried to register:
> Access blocked, please contact support.
No link to contact support, on help page no way to contact the actual support. Classic Musk tech.
Such an insult to vibe code a new github and then expect developers to handle code to them so they can train on it.
Who trains on post-2022 code?
Feels like the more interesting trend is that Git itself is becoming an implementation detail.
ahh so thats the game plan: train codebases even if they are not written in or by by Cursor agents.
Git can be hosted over ssh or filesystem. I moved to self hosting git. Now it’s always up, go figure!
Never going to try it, sorry for anyone at cursor who was excited about what they were working on.
Does it support jujutsu natively?
We are very interested in this. If anyone can put us in touch with the JJ maintainers we would love to chat!
Unfortunately I wouldn't know how to do that.
But just saying that I would probably argue against any new forge at my workplace if it didn't adopt the jujutsu model for changes. While agents can wrangle git better than 90% of developers, jujutsu just works much better for both.
How about an original and distinguishable name?
Then again, what did I expect from a company named cursor.
It's trendy. If you don't like it, you can go write some go in go. Contribute to a thread on Threads. Write a factory with Factory. Write your reflections in Reflect. Keep in the loop on Loop while you recall what you did with Recall. So many choices.
Why not evaluate them on the merits? Is the name really the thing that’s important here?
“GitHub has so many outages, but dammit, that other one has a terrible name, it’s practically unusable”.
And origin is a pretty great name to give a git hosting platform. And do you think GitHub, Gitlab are original?
This is a great name! It ties back to git nicely (git pull origin/main), and sounds human.
Cursor was also a great name given that it evolved into an advanced AI assisted auto complete. I think their team does a solid job with naming.
Then google the names and look what comes up.
Without additional search terms you get the wrong results.
A word that is already used in computer is a bad name.
And with AI it won’t get better. Ambiguity is a source of confusion and mistakes.
By your logic pull/push/main are also great names.
One main purpose of a name is distinguishability.
If you google Cursor they dominate the whole first page
Surely they have a chance of doing it again with Origin
> Without additional search terms you get the wrong results.
Until the entity becomes popular enough, like I just searched for "stripe" and all visible results were about the payment company.
I wouldn’t push someone else’s code into a Musk-owned repository.
This is a good start. You can connect it to 3rd party apps to do CI Runners, etc. Hoping it develops quickly and becomes more fleshed out.
The main limitation right now is that to share the code with anyone outside your Cursor organization, you need to sync it with GitHub.
How will you ever make up for the network effect that GitHub has?
What's worse than GitHub? - Its vibe-coded clone!
Has GitHub made it harder for them to scrape to train on?
Vibe coders rejoice! Vibe coding is even easier now =)
Jokes aside, this is a smart move by Cursor. As you can probably tell by the comments here, it's not for the experienced developers, but it will be a natural choice for vibe coders.
which fanless "mini pc" can run a github clone on my local network? ideally proxmox support so i can use it as a mini lab for other things
Will they train on all the code hosted there? :)
No thanks. I’ll stick to self hosted forgejo.
I would only host worthless stuff there
It sure worked for Meta's Threads
I would never host my code with Elon Musk. I've also moved on from using Cursor to Claude Code/Codex. GitHub has problems, but this is likely (and hopefully) DOA.
I was annoyed seeing unskippable "ad" within Cursor about it. Looked like a huge banner "give us control over all your code" with only one visible button to consent
Let's see how strong GitHub's network effect really is.
While the real solution is to actually self-host as I said before [0], it is worth testing to see how much can GitHub's users tolerate the endless outages and unreliability and where they go next.
We'll see if they want Cursor Origin or the declining GitHub.
[0] https://news.ycombinator.com/item?id=49331922
We are more likely to find out the anti-network effect of Elon
There are many people who will not touch this and look down upon those that do. Ethics and morals still seem to have some weight
> We'll see if they want Cursor Origin or the declining GitHub.
I don't want either tbh. There's plenty of open source alternatives. If there's anyone I trust less than Microsoft, it's Elon Musk.
oh yeah i'd love to feed these guys training data
I am quite starting to like this timeline. I hard boycotted GitHub for ten years after the meritocracy rug fiasco. I went back there dragging my feet while transferring hundreds of repos a few years ago ...
Origin has a chance to be the future of this space, with a clean slate and reputation. I pray they don't soil it.
If you're willing to trust them with all your source code...kudos, I really wish I had your level of trust and delusion
Vibe-slop needs protecting?
Ask sama
and to think we let phabricator wither on the vine...
Screw you elon
Musk is one of the last people I'd trust to centralise all my code with.
Not sharing my private git repos with Elon Musk owned company.
feels kind of embracey
Never going to sign up based on the association to Elon Musk.
now that software is easy to make and github has problems, how many "git as a service" offerings will there be?
My guess is there will be at least dozens if not hundreds.
So first they steal the source code in bulk, and now they are like fuck it, just send us your source code? Musk does indeed play 1d checkers.
I was just thinking how I wished GitHub was part of a csam generation company owned by a Nazi. This sounds great!
A Musk controlled company to look at my private repos?
Let me thing about it.
No.
Instead of creating one more centralized alternative, it's best to invest all effort into a decentralized solution, like Radicle[1][2] or federated Forgejo[3][4].
[1] https://radicle.dev/
[2] https://radicle.network/nodes/seed.radicle.dev/rad%3Az3gqcJU...
[3] https://forgejo.org/
[4] https://codeberg.org/forgejo-contrib/federation/src/branch/m...
I feel like this misses the value of something like Github or Origin entirely though. There's nothing terribly difficult about hosting a git server. The UX and pull requests capabilities are a little involved, but there are alternatives today.
What makes Github special is the ecosystem and integrations. Sentry, Linear, and so many other tools have special UX around Github. You need centralization (or at the very least, a standard) for that.
It sucks to see Github's fall from grace, but I just can't bring myself to use a service like this under the custody of SpaceX. I've been saying this for a while, but I really wish Linear would swoop in and become the unified platform for software development.
It sucks to see Github's fall from grace, but I just can't bring myself to use a service like this under the custody of SpaceX.
Agreed. It's like going from bad to worse. Not to mention, Origin doesn't even support actions. They are doing the easiest part of git hosting and no mention of anything that actually makes Github what it is. What Origin is offering is literally a weekend project, and anything having to do with Elon is arguably one of the least trustworthy organizations in whatever industry he chooses to participate in. No, thank you.
This is going nowhere. GitLab is the cautionary tale. Massive funding, nine figures of ARR, years of losses, and most of the revenue comes from self-managed enterprise licenses rather than the SaaS. If the best-funded attempt lands there, I don't see how this one goes anywhere when they have not even a minimum viable product and reputational issues right out of the gate.
The thing with Gitlab is probably that it had to deal with the network effect/winner takes all conundrum. GitHub was first on the scene, it was the first to make a centralised hub popular and it has had the custom of many notable open source projects.
Gitlab as a product is miles ahead of GitHub at this point. (Except at the point of scaling very large repositories I gather.) There is just very little money in saas beside the bespoke enterprise contracts.
I’m not sure how revenue between GitHub and Gitlab compared before the Microsoft acquisition and now but GitHub has always leaned on being the popular tool with the network.
Actions should be decoupled from the version control system.
In theory, sure, but the reality is that "development activity" revolves around "codebase". Plenty of projects have separate bug tracker, source code control and CI system. Most people likely prefer GitHub's all-in-one solution, especially for small-medium open source projects.
They've already separated the wiki's version control, why not CI? And the issues/discussions are just vendor lock-in wrapped in a shiny UI.
And do we not see how terrible this is as a result? Hitching your tent to corporations, especially American corporations, is always an exercise in misery.
*Atlassian from Australia has joined the chat*
Mostly they joined the chat to show you what real misery is. I tried Jira and they couldn't even handle simple billing without screwing it up. Ugh
I recently had to rebuild a jira frontend from scratch just so a single jira tab didn't take down my 8gb MacBook. Support only the subset of features I actually use and it's great. Makes me hate jira even more.
Thankfully, there is a Jira CLI tool[1].
[1] https://github.com/ankitpokhrel/jira-cli/
and thankfully there is linear. :-)
I'm not saying it's an objectively "good" solution, just how damn smooth the experience is.
[flagged]
I disagree. The integrated workflow of actions, issues, code review and source control is a huge productivity win. There’s a reason people chase a “single pane of glass”.
A website where one views and interacts with all of these things does not necessitate that all of these things run on the same system. Since this most recent GitHub outage, I'm looking into modularizing my git server, issues, and actions; there's no reason why one going down should bring down all the others.
What does make GitHub what it is? Weekly downtime that cripples our companies?
What do you mean by "Github's fall from grace"? What happened?
https://www.githubstatus.com/history
Is this a result of too much activity on GitHub (AI-driven jump in repos), though, rather than a fall?
SO a company Github's size can't handle the slop they invested a lot to enable and this is not a failure? I mean if the market moves and you cannot move with it, what do we call it?
Mostly it seems like greatly increased usage pushing their systems to the limit.
Ahhh okay. They've definitely had some growing pains. By "fall from grace", I assumed you meant some sort of ethical/trust failure.
That too, ever since they decided everything on GitHub was fair game for training AI without respecting its license.
Of course, cursor is going to be even worse than github in that realm.
Github has lost trust. This wasn’t an isolated incident.
> a standard
Like a specification[^0], which is exactly what ActivityPub is and what federation in Forgejo is built around?
[0]: https://en.wikipedia.org/wiki/ActivityPub
I'm sure that sentry and co could spend a few bucks and commit a KPI in one of their upcoming quarters to figure out the activity pub spec and support integrating with it if their hearts desired. It's a chicken egg problem and we all know chickens exist. It's clear that the centralized model is getting left behind, especially as the EU is moving away from US services and they continue to invest in open source, European-based alternatives like Codeberg and Forgejo.
The communication protocol is not what I mean here. The feature-sets and capabilities around "pull requests", "reviews", "comments", "checks", "issues", "projects", "deployments", "releases", etc are all different across platforms. You have to integrate bespoke for each platform. There is no universal standard for how CI/CD, deployment tracking, project management, etc are done.
Yes, Sentry or whoever else could build a Codeberg integration, or a Forgejo one, or a Gitlab one. But these would all be bespoke. There is no standard. So the one with the biggest market share ends up with the most integrations. This is where we are today.
I'm all for open standards. But I don't even know of a proposal for such a standard.
Sure, with AI tooling, you can vibe code your own browser extensions or other integrations in some cases. But for a lot of platforms, that just won't cut it. That's before we even get into the maintenance burden of that.
Unlike GitHub's universal standard? Forgejo already has a near 1:1 github action format for forgejo actions. You can even use github actions in forgejo. You cant have AI build it out for a few tokens? Why are you so certain of failure?
Interfaces exist in programming languages. The strategy pattern is one of the gang of four patterns. I'm sure a good engineer or an AI can figure it all out or is that too hopeful? Unless Sentry engineers or AI are too incompetent to make adapters? I hope not, the market is in big trouble if a dev tooling org cant make tools for where devs are
That used to be a big deal. I don't think it is anymore.
One great thing about the rise of AI is that now it's super easy to make and maintain integrations. This is a place where AI coding uniquely shines due to the points being well defined and mostly quite standard.
This is going to hurt the monopoly power of software companies. But it'll be great for allowing the proliferation of smaller companies, open source, and home grown solutions.
I'm already seeing the pendulum swinging back to local first instead of cloud services.
Creating? Sure. Maintaining? absolutely not.
As someone who has some SaaS products that must connect to other SaaS products, I can tell you that this is somewhere that AI saved me a truck load of time.
A lot of SaaS companies change their API in subtle ways throughout the months probably too rapidly. GraphQL schema changes, splitting apart one api endpoint into 2, changes to authentication method required. A thousand little cuts.
This used to be my biggest time waster. Now it's barely on my radar. Life changing for me. I guess your mileage may vary.
Concurrently, the writing is on the wall for us SaaS product providers - lower margins, and more competition for the same reasons. I welcome it. More quality and choice at a better price for everyone.
Yes but I don't think maintaining clones of SaaS products will be very easy. Maintaining integrations definitely gets a lot easier
At the same time, I think some kind of aggregator will emerge. In music, the tools to create music are so inexpensive that the market is flooded. In games it is becoming the same, now accelerated by LLMs. But there is always some type of aggregator ending up earning the money from most of that, rather than the people creating the content. Not sure how it will work out here, but I predict it will be another type of aggregator.
You can get surprisingly far just by using git as a protocol. Such mundane things as tickets and chat logs do not really need any proprietary tech. Think Google OKF. https://cloud.google.com/blog/products/data-analytics/how-th...
> I've been saying this for a while, but I really wish Linear would swoop in and become the unified platform for software development.
And then, 5 years later, we are back at square one when everybody uses Linear and they start going the fate of GitHub. Self-hostable, standardized alternatives are the only real sustainable solution.
Agreed on all points.
I think people often forget that GitHub is really more akin to a social network than it is a software development tool.
By that I mean the stuff that makes GitHub great is all the same stuff that originally made social networks exciting (back before the algorithm was tuned into a horrifying cycle of negativity).
except when you are paying for their enterprise edition - and they make no distinction between paying customers (quite a bit actually once you add in actions costs + their other addons, it's quite a bit of money) and their free tier full of AI slop PRs on rando projects.
As someone who pays github, I'm super annoyed at the downtime and still I'm getting charged for actions running that all fail from their lack of concern for their enterprise customers.
Having some independent pools of capacity for paying vs. free customers isn't rocket science, it's basic competence (or care). I'm quite sure their are competent people remaining at GitHub (although I hear most of their original folks have left, and the 2nd tier of folks who joined from microsoft seem lackluster at best). I chalk this up to lack of care, or hubris, or both.
Oh I agree that the outages are unacceptable regardless of their “social media origin”.
I also don’t buy their excuse that it’s due to increased loads due to AI. If that were the case then we’d see more patterns to the outages. But instead it affects all GitHub systems but at different times. It feels more like a problem with their orchestration layer than capacity issues.
> What makes Github special is the ecosystem and integrations. Sentry, Linear, and so many other tools […]
github didn’t start with those things. Those things came with adoption.
Its like all those people who bemoan their favourite software not existing on Linux as a reason not to switch, completely oblivious to the fact that this hegemony is the reason linux doesn’t have that software.
It’s like an appeal to the status quo - its self fulfilling, impossible to break.
> You need centralization (or at the very least, a standard) for that.
We have a standard! It’s this great decentralized VCS called git. For most of the stuff integrations do, they’re just showing code or commits — both of which are readily available in a copy of the repo.
Probably the one thing that’s missing is a standardized webhook system so that your git repo storage could tell integrations when stuff changes. Though even that isn’t strictly necessary if the integrations keep a copy of the repo up to date.
Git is absolutely miserable without things that GitHub, GitLab, and others provide, and there's still a huge ceiling to grow into in terms of layering on UX on top of Git.
In my opinion, Git is holding software development back. Git only cares about lines of text. This is not enough in today's world.
Lots of standards are terrible without the actual tools on top of them (try making an HTTP request with just a network socket).
But that doesn’t mean we should have single centralized tool that does everything.
I’d posit there’s no big loss if you broke these things up and each had their own copy the repo (if needed):
- CI/CD
- issues
- code review/pull requests
- release artifact distribution
- discovery+social credit for open source
You could host your repo on any dumb storage somewhere that speaks git+ssh or git+https and each tool could pull from there.
Completely agree. Once again, the answer is protocols, not platforms.
[dead]
> Git only cares about lines of text. This is not enough in today's world.
I'm working on git-inspired VCS based on syntax tree nodes stored in prolly trees. Might that be enough? I've not 100% locked down the format yet, but tentatively it looks like this: https://gist.github.com/conartist6/7dd40697140965ee343de93b9...
Other version control systems exist. It’s not git’s fault that people can’t be bothered to use the appropriate tool.
And I say this as a certified git hater.
Why not? LLMs also only care about lines of text.
I'm with you on the pros and cons of centralization. But what makes you think if linear became the unified platform that it wouldn't just become like GitHub? See jira.
Given the history here, No matter which tool becomes the death star of developer workflow it always trends towards complexity and the challenges that come with it. Hosting your own doesn't become a panacea as most IT/ops teams also struggle with uptime and upgrades (increasingly relevant).
The only way out IMHO as customers picking two of these goliaths of sending a clear signal to them that we're happy to pay them handsomely for a stable and less sprawling product. Or a hosted open source version with a different cost structure that doesn't drive these wild swings at feature bloat and the inherent instability that comes with that chase.
I think the UX around pull requests and code reviews are extremely difficult to get right and to this day I haven't seen a platform that I'm really happy with. There are bits and pieces in some places but none really bring stuff together. Here a small list of things I'd like to see: - diffs should be language aware and highlight semantic differences - I should have a working LSP on both the old and the new version simultaneously and have decent full editor code navigation features and git history inspection available - I want ergonomic ways to see diffs between different versions of commits - I would like to be able to comment on anything (commit messages, inter diffs, PR description, responses, etc.) - I want a better integration of LLM reviews than having them dump their findings in an a big PR comment
I think there is plenty or room to innovate on just the basics of PR and codereview.
I am working on a GitHub compatible API for my code forge at https://juju.bi, and thus all the existing integrations of GitHub should theoretically work with my code forge too.
i agree, and i strongly believe atproto can be the standard to allow this in a decentralized way! which is why im so excited about tangled
https://tangled.org
> What makes Github special is the ecosystem and integrations. Sentry, Linear, and so many other tools have special UX around Github. You need centralization (or at the very least, a standard) for that.
I think “a standard” is the key here. There is no reason why integrations must work through a proprietary centralized service.
Centralizations were prerequisites to have nice integrations. Yet, what used to be a precious plugin now (more than often) is a 15 min vibe coding warmup.
https://tangled.sh
Tangled is good, I have a repository hosted with them. Both they and Radicle lack private repositories though.
Forgejo and Codeberg I look at as more like hobby projects. I would not trust them with my code.
git is such a good platform for decentralized tech. Forfeiting that advantage is just criminal. What I really want (and almost have) is a local GitHub for me and agents. With 4+ orchestrators working with a ton of subagents, things get difficult to coordinate. Several days ago, I went back to plain git and it was real slow and abrasive to work with.
With some wizardry applied, I effectively have an "Excel for worktrees", a very productive and low-friction environment. https://replicated.live/blog/crdt
You don't sacrifice decentralization when you use Github. You can still do all the same decentralized stuff, have other copies of the repo, do everything offline, directly pull from other checkouts, etc. Github can just be ONE of the places you push your code to, it doesn't have to be the only place.
Yeah, it seems disingenuous to pretend to care so deeply about decentralization while misunderstanding git to the point they don’t even realise none of that goes away because you also push to GitHub
I just use git with a remote and a custom CI system. The whole thing is trivial to build. All these forgejo etc are kind of over complicated software. Open Office of the modern world.
For me at least the value was never the web UX (which there were always many OSS versions of floating around, and wasn't very hard to build pre-AI.) It was always the universal standards around GH Actions, Runners, checks, etc. The nice thing about GH was that GH felt like it "commoditized" the infrastructure that otherwise was built-out and maintained at Big Tech. Forgejo and similar are web UX. Tangled is trying to open/federate the social network on ATProto. But the compute interface is where I see the most value and it's also just the hardest thing to hack on independently.
In a project with two levels of submodules and tens of worktrees, you cannot "just use git" :/ GitHub does not help much either.
GIT over Reticulum has been my major focus recently. Check it out! https://reticulum.miraheze.org/wiki/Rngit
Decentralized solutions invariably appeal to a narrow sliver of nerds but never gain popular acceptance.
Ah yes, I guess you‘re right, that‘s why all the serious developers continue to use CVS.
Or just store it in an S3 bucket with GitSocial: https://gitsocial.org/
I think the biggest value of choosing gitlab is the social network. I think federation is the right way, but it seems a mastodon vs linkedin problem. Would be good if there would be bridges between all those fediverses. A nice one is also ngit [0]
[0] https://gitworkshop.dev
Forgejo federation being mentioned each time never stops being silly. It hasn't been a thing for years, and they don't appear to have made any meaningful progress to show for.
Had a quick look at it and it's riddled with ideology statements everywhere which is a huge red flag.
I'm a Forgejo user myself, but honestly reluctant and scouting alternatives since the activist Codeberg direction.
Git was already decentralized. That's the whole point of different upstreams.
Somebody had a lot of fun designing a spaceship to get to the grocery store
I haven't heard about radicle in 6 years. Does it work yet?
Exactly, decentralized solution is perfect!
you mean you dont want a singe company to have all your code?
[dead]
[dead]
[dead]
How do you think does Microsoft start to do some thing better? At least cut slop form windows or put more effort in gh?
No free tier, surprised..
If by now you're willing to make EM the guardian of your code, nothing I can say would dissuade you.
GitHub sync = defeat?
No actions replacement means this is a nothingburger in a lettuce wrap.
Man, I was disappointed - hoped this used something like darcs under the hood
Is it just me or is this thing just a poor man's version of GitHub?
Host your code with Elon Musk, what could go wrong
True, but people risk their lives by using his vehicles every day, why not YOLO their code too?
if its another place that offers free computer through their version of GitHub Actions I;d call it a win.
Cursor, which is now owned by Elon, who tried very hard to get all manner of data about citizens with his intrusions? I'm sure I'm not the only one who thinks this isn't a great idea. He'll use it to feed Grok.
I'm not a big fan of Github right now either but I wouldn't consider this alternative due to its ownership chain.
Considering that Grok CLI was caught uploading peoples entire codebases... Yeah, I have little faith and trust. I wanted to try Cursor and Grok but I won't because of this.
I still wonder how it should work otherwise if the "AI" is not running locally - it needs to see the codebase & because it runs somewhere remote in a cloud then it needs to copy the codebase over.
I'm rather surprised people would expect it work any different for non-local "AI".
It doesn't even swallow an entire code base so there is absolutely no need to see the entire codebase. Usually models search in the codebase locally and then use only the relevant parts. There is almost never the whole context just spammed by an entire code base.
I wouldn't even feel comfortable cloning from it, let alone contributing. If you decide to host your project there, please mirror it somewhere sane as well.
That sounds like feeding the beast with a spoon versus bare handed. Your still feeding the beast.
They were also caught having their Grok code agent upload entire source code repositories in the background.
I guess the joke is "come to Origin, your code is already there".
Yup, same sentiment here. Thank you
It’s unfortunate I had the same thoughts right away. Are they going to look at and steal all my data.
cursor now wants your code
[Feature Request] Sign-in with your Tesla account ... /s
I love how people like you actually think Elon actually wanted your data.
If you actually believe that he didn't, you are, at best, out of touch.
His little "DOGE" endeavor was actively caught exfiltrating personal data about everyone in the country, and the first thing that happened after SpaceX acquired Cursor was pumping all of the data Cursor had acquired into Grok training.
He is quite literally the shining example of "Imma do what I want, even if it's against my own terms of service or otherwise illegal, and then you can try to sue me over it later".
[flagged]
I dunno how it is in America, but in my country if the largest individual donor was granted carte blanche to run his own department within the administration, with his own people, answerable only to the head of the government, it would cause a helluva lot of outrage.
"Carte blanche but don't touch all of these things"
Yeah. There's a reason he had a falling out with Donald.
There's a ton of agencies with a lot of power. DOGE ideally would have used said power for good, but got blocked again by again by an infinite mountain of corruption. People should not work in favor of said mountain.
And people really don't recognize the value of DOGE's self-sacrifice by design, having a limited time to act out its mission. Nothing comes even close to the honesty in the operation in such a swamp.
Don't really get the incredulity considering he needs to justify the tens of billions already spent on AI infrastructure by, you know, training a model or two ...
Sigh, hi. My heart goes out to you.
And Microsoft doesn't use Github to feed their own monsters? And why wouldn't Elon just use GitHub to feed Grok?
Is Elon maybe just living rent free in your head?
The Microsoft CEO is not a self-declared nazi. Check yourself.
GitHub is a mess so lets throw our code into a Musk owned enterprise, famously a different shade of mess?
How about neither?
Man I wish we stuck with Subversion at this point.
I feel like the "GitHub is a mess" conversation is a little forced at times - I spend a significant portion of my work day in GitHub and while there have been an uptick in outages lately, it still feels like a very solid platform.
I guess I personally wouldn't categorize it as a "mess" and the conversations I read about it here make me think I'm using an entirely different product than everyone else.
When you work at an enterprise that depends on Github, frequent outages are actually a pretty big deal and begin to impact productivity in a real way.
It feels to me like this is a case of very rapid increase in load from AI pr's more than product quality degrading because of cruft or poor maintenance. It's annoying but I'm not confident that alternatives will do better in the long run vs GitHub (which I'm guessing will improve with time)
It's not only PRs, my personal projects which used github actions were seeing more builds in a single month than all of the previous year.
Yeah and if I was an enterprise I would definitely be self hosting with Forgejo or Gerrit or whatever. But I think for most hobbyists GitHub is still totally fine.
I use it quite a lot and the first time I actually witnessed downtime myself was yesterday.
github for hobbyists and small orgs and simple open source projects is wonderful.
github for large project that rely on CI tool is a nightmare (cost, downtime, etc.)
I'm guessing you dont know about the github actions bug that was running and endless loop consuming 100% cpu costing orgs insane amounts of money that they didn't fix for 3 years.
I love to hate on Github/Microsoft as much as the next person, but considering the insane amount of code being pushed to it every day, I'd say they're doing quite well.
I disagree. They’ve been struggling with up time for a long time, this seems more related to their move to azure than anything else.
It also shows there’s a fundamental problem with their architecture or their development pattern to have these sorts of issues so frequently.
Plus a rather clear lack of investment into improving it / retaining their talent / etc, yeah: https://damrnelson.github.io/github-historical-uptime/
Anecdotally: at work I saw quite a lot of github-uptime-related issues, and because they're the central host for everything they become *cough* load bearing for a huge amount of the company. Stability and performance has been dramatically worse than the self-hosted phabricator+gitolite before it.
I'd emphasize their practices as being the culprit.
Migrating to different cloud vendors, while difficult, is something that people have done many times before. There's only finite number of incongruences that appear when moving cloud vendors. It's rather difficult to explain the ongoing problems several years after the fact.
Anyone that's worked at a startup knows that one of the best things about it is reflecting on all of the asinine, if not genuinely stupid policies of your previous employer so as not to repeat them. Without any insider knowledge, the practices and procedures GitHub had prior to acquisition probably look nothing like those post-acquisition.
It's falling apart at the seams for me. Half the time I click somewhere on a page, the whole thing freezes and I have to reload the page. Took forever to install neovim on a new machine today because all the plugins installed via Github and I had to wait like 10 minutes for it to come back up.
I’ve heard from people who left GitHub that Microsoft pulled all their best dev ops people to work on AI scaling. This is blatant disinvestment by Microsoft bc they think GitHub is untouchable. I hope better alternatives than musk come up
Really? I feel the opposite.
Earlier today our app, used by a handful of actual production customers, went down for 30 minutes due to a GCP migration to a new project and we screwed up a secret in prod. This was an all hands on deck firedrill with 5 whys and incident reports and chaos.
Github has been having major production outages routinely! There are companies who can't release, can't build, can't view issues, basically can't work. I'd imagine most orgs using Github have been forced to move off of them or create an alternate approach to actually get shit done.
Very solid? Bro. GitHub is at what, one 9 of reliability? That's a disaster. That's a mess and an embarrassment in any professional sense.
I use GitHub a lot and it does what I need well enough, but I wouldn't spend money on the level of service or performance it generally delivers.
Git is objectively better than Subversion. The problem is we are conflating a widely used hosting service with the underlying SCM. We don't need github to use git successfully. Sure, it's nice, but you can install Forgejo or something, which is the exact same boat we were in with Subversion + Trac. You can have that experience right now if you want to.
What github is good for is discovery, getting contributions on open source projects (because it's so popular), and integrations. If you wanted to, you could host your open source project on Forgejo and use github as a mirror just to benefit from the visibility and get contributions.
100% not a chance for me. I'd rather host it myself on a Pi on a 56k dialup connection with 12% uptime then ever let an Elon related company touch my code.
This is the correct answer, but let me add a corolarium, both github, whatever cursor launches and any other bigco does obeys to the same friking iron law: *MAKE MONEY/POWER AT ANY COST, MONEY/POWER IS THE OBJECTIVE*, so you will get EXACTLY the same result from each one of them given enough time has passed.
Or Perforce. Perforce, in my experience, just worked. It also has sane conceptual names for things, a good CLI, SDKs, visual tools, cool features like workspaces that Git completely lacks, etc.
Git also just work. The issue is the host platform, not the CVS
Git has issues all the time.
Bastard thing corrupted some of my source code that was "on a shelf". Just gone, no path to retrieving, reimplement it. About a decade ago now and I'm still angry about it. Don't lose the work is table stakes for source control.
My understanding is that Perforce is really strong in the game development industry, right? I think it handles blob/large file assets better or something like that.
Mercurial always looked nice
Mercurial still looks nice.
That makes very little sense to me. When your Subversion server is gone, you're hosed. When your Git forge is down, you take any old SSH shell account anywhere, push and keep working.
The last subversion server I had was on prem on VMware. There wasn’t any unscheduled downtime in a decade.
There’s no architectural limitation in Git preventing you from replicating that exact setup for another decade, though.
I like Musk companies more than most here it seems, I think he’s a net positive to the world. But (admittedly sans hard evidence) he absolutely gives me the vibe he’d train on my source code without my permission. I simply don’t trust him in this arena.
So there is more to lose than just the service being unreliable.
> I think he’s a net positive to the world.
As a person he's either a poser or a neo-nazi. As a government agent he's responsible for many deaths and as of yet untold harm. Tesla was one of the earliest EV manufacturers, but these days they're overpriced and will be remembered for panel gaps and being destroyed by a car wash. SpaceX has done a lot for that sector, but will mostly be remembered as the company that fucked astrophotography with 10s of thousands of satellites. And Twitter has been dragged into irrelevance so badly it was being called a "reverse startup." Neuralink seems to be going nowhere...
Where's the benefit?
I can forgive him ransacking the federal government, exploiting workers in everything he does, building data centers that pollute like a cartoon villain, backing a fascist takeover of the US and similar ones all over the world, and openly sig-heiling, but I can't forgive him training on my source code!!!
whatever you say bro - https://www.thelancet.com/article/S0140-6736(25)01186-9/full...
Just the association with Musk will mean no EU company should use Cursor or its products.
why?
Do you not follow the news?
Among many other things he has openly suggested that the EU should be abolished and national sovereignties restored and donated to parties that tow that line.
I don't follow EU politics. Isn't that position a pretty common EU political view? What is evil about decentralization of power?
How is the UK doing after leaving the EU?
ah yes, the EU shall continue to fade into irrelevance (except for as a dope vacation spot and some fine cheese)
Still. 450 million consumers, a GDP of over $23 trillion. Not bad for a cheesy vacation spot.
But the EU isn't obsessed with "winning" or "being number one", like the US, which is number one in almost no area regardless.
if we're totally switching tooling, lets at least go to mercurial instead
If we're switching, we should probably go to fossil instead.
That actually does make sense seeing as fossil bakes in issues and wikis and all that other stuff that makes it hard to use plain git.
GitHub's wikis are just git repositories with folders full of Markdown files. That's the easiest thing to pull out of GitHub. Just git clone the wiki repo. At that point I often just use a git headless merge to merge the second repo into the main repo under a docs/wiki/ folder or similar.
As I recall, in GHE I remember finding that you could even turn off the repo's wiki at that point and then reset it to point to your docs/wiki/ folder and still have the Wiki interface but pointing it to a branch in the same repository. That was a while ago and I don't know if it is still supported in GHE. I also remember it was work managing that wiki branch because Wikis didn't directly support PR workflows if you have branch protection policies. Also from the documentation it doesn't look like it is supported in Public GitHub right now, which seems a shame to me, as someone who likes all documentation in the same repo as the source. I haven't needed Wikis in public GitHub much, so I haven't tried to do anything like point them to a main repo branch.
That's fair. You could also just put the wiki in a different branch if that's your style.
You do know you can run git on your own machine, on pretty much any Linux/Unix based server, run it entirely peer-to-peer if you wish... or use Gitlab, including self-hosted Gitlab, or one of many other Git based servers?
You don’t even need a “gut based” server. Git doesn’t care about the transport
Yeah. I just don't really like git. The user interface is dog shit, merging is nasty, it's really easy to blow your own toes off and it's generally horrible. The only reason it got anywhere was github.
There's other stuff out there. You don't have to use it except when you're working with someone who does and there are other systems with some compatibility for that exact scenario.
Git became popular because Linus Torvalds created it and it's used for the Linux kernel. If it can succeed there then it's battle proven for sure and that's why people pick it. Almost no one is working at that scale though so don't sweat it. Use what you want.
Mercurial was pretty nice back in the day and I think people still use it. If I was going to use something other than git these days, I'd probably take a look at fossil. That thing is so damn cool. You don't even need something like github with that because it's literally built into the fossil binary. And when I say binary, I mean a single, standalone binary for the whole thing.
Okay, well, nothing’s stopping folks from using hg, P4, svn, CVS, RCCS, MS Visual Source Safe, ClearCase, or who knows what else.
Sometimes you get told what to use. And I am free to dislike it :)
That must have been why the used it to manage the linux kernel, because of github
/s
It's painful if you have to work with it and mortals, not kernel devs.
Musk is toxic waste
This feels like an attempt to get more of Musk's non-coder fans to start vibe coding slop-products. "You can now code with Grok via Cursor and host your code on Origin, all without leaving Daddy Elon's ecosystem."
thanks I hate it
Ah, another post about Cursor.
About half of the comments will be generic criticisms of Twitter, SpaceX, etc. for things that may or may not have happened, or personal complaints about Elon Musk.
exhausting to watch!
Kind of confusing people won't use this because they don't trust Musk, but has any other major platform even tried to open source their algorithms the way he has with X?
I get hating his politics, and his predictions are often behind schedule, but he certainly says what he thinks and isn't in some shady user data markting business like zuckerberg.
Anyone who trusts Altman over Musk is... I don't know. Give me the autist over the sociopath when it comes to handling my data.
Does this have stacked PRs? Cursor acquired Graphite a while back and stacked PRs were a killer feature.
I don't know, my gut says that this is going to end in a bad way. I signed up, I think but wasn't sure if I needed to buy Cursor, or Origin came with a subscription, so confused.
For now Arc is staying in Github: https://github.com/Basekick-Labs/arc
[flagged]
[flagged]
[flagged]
[dead]
[dead]
[flagged]
[flagged]
[flagged]
[flagged]
Literally your source:
> her stepfather used the Grok chatbot to transform a childhood photo into child sexual abuse material
He could have used Stable Diffusion. Or, if this was the 90s, Photoshop.
By your argument, we’d have to abolish the internet, or add nanny filters to everything.
How many people on HN resent it when Claude refuses an instruction? And yet here you are, arguing against personal responsibility and for a when-in-doubt-block approach to technology. Is it because it’s Musk?
Come on, man. Not allowing products to generate CP is too high of a bar?
By your logic cops should be focused on the junkies, forget about the cartel
Correct me if I'm wrong, but it didn't actually generate CP, it changed people into bikinis, not actual nudity.
You can go to the beach or (I assume) search on Google and find people of any age wearing bikinis. And that's fine because to normal people (not pedos) it's not something sexual.
I understand that someone specifically editing a minor into less clothing is probably a pedo. Probably the same way someone generating feet gets aroused with that stuff. But that doesn't mean a model generating feet is problematic.
I do think making the model reject the bikini thing is a good idea once you find out it's happening, but let's not pretend it's the same as the model generating actual sexual content.
That's horrific and the saddest thing I've read in a really long time. I'm glad I erased my Twitter account and I'm definitely not using this code host. Seriously, this is horrible.
But you'll comment on a website that funds Flock
I pay $0 for this website, and there's no advertising (or if there is, I adblock it).
Well, shit. I guess it's back to Slashdot then.
[flagged]