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 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”.
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.
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.
> 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'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.
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.
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.
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.
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.
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'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.
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.
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.
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.
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.
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.
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."
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.
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.
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.
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.
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.
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 :)
So, a complex, vibe-coded platform that was prematurely pushed out to opportunistically take advantage of GitHub's egg-on-face moment? But it doesn't do anything better? And it's owned by the same guy who routinely takes shortcuts that causes both EVs and rockets to crash? Sign me up!
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."
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))
Will code uploaded to Origin be used to train LLMs to generate more accurate child porn via Grok? Seriously how do you stomach working for such a company?
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.
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.
- 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.
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!
> 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 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)
> 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.
That's a good point. I hate Flock and YC should be ashamed of funding it. Still, Musk has a wall to wall record of horrific actions going back years. The Grok CSAM is the worst I've seen yet though. Absolute deal breaker.
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 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.
"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.
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.
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.
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.
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 :)
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
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.
I actually think the cursor acquisition will destroy value, unless completely left alone by Elon. Cursor's strength is its brand, and they should add more services, but should not integrate with Grok whatsoever.
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.
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
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.
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.
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?
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.
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.
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 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
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.
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.
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.
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.
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.
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?
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.
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.
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 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.
[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...
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.
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.
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.
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.
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.
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.
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.
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.
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 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...
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.
With some wizardry applied, I effectively have an "Excel for worktrees", a very productive and low-friction environment. https://replicated.live/blog/crdt
How about neither?
Man I wish we stuck with Subversion at this point.
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.
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.
It also shows there’s a fundamental problem with their architecture or their development pattern to have these sorts of issues so frequently.
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.
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.
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.
/s
So there is more to lose than just the service being unreliable.
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.
I'm not a big fan of Github right now either but I wouldn't consider this alternative due to its ownership chain.
Happy to answer any questions about Origin or source control in general!
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.
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.
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.
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.
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
Once Azure's massive spend is online things will slowly get better.
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.
It will eventually have the same uptime issues as GH (or worse) with less ability to provide stable fixes.
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 :)
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))
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.
They acknowledged that it happened, fixed the bug that caused it, deleted all data that was uploaded.
Hahahahaha...
Well, if you believe that, I've got a bridge for sale...
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.
> "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.
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 appsfwiw, the main site also seems to be down right now, infinite spinner. Docs and blog are loading though
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!
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
downtime and self-hosting is not a reason most people are convinced to move forges, it takes a lot to get someone there
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.
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
Self hosted servers go down too and why people have picked hosted solutions like GitHub, Bitbucket, and GitLab
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)
https://www.washingtonpost.com/technology/2026/08/15/woman-a...
https://katv.com/news/local/prominent-arkansas-photographer-...
> 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?
By your logic cops should be focused on the junkies, forget about the cartel
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.
All your codebases are belong to us.
But the website uses 100% CPU. And this is a beta for paid plans, not a GitHub alternative.
* 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.
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.
Creating a world-ruling AGI which feels mild sympathy for its creators, sharing some table scraps with them. Not sure if /s...
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.
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.
Inflated valuation of buyout, based on inflated valuation of SpaceX
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.
There's a lot of space to innovate around collaboration and version control in the age of agentic workflows.
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.
https://s-1.vercel.app/
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.
just add repo-level agents, org-level agents and done. they could have got 50B valuation by now
https://codeberg.org/
I wonder if they were waiting for a major githut incident :D
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.
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.
No thanks. I want GitHub alternative, not Cursor with source control attached.
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 funny that they are trying to replace GitHub by the part that already has multiple alternatives tho. This will fail without a doubt.
The main limitation right now is that to share the code with anyone outside your Cursor organization, you need to sync it with GitHub.
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.
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.
> Access blocked, please contact support.
No link to contact support, on help page no way to contact the actual support. Classic Musk tech.
My guess is there will be at least dozens if not hundreds.
Then again, what did I expect from a company named cursor.
“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?
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.
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.
Surely they have a chance of doing it again with Origin
Until the entity becomes popular enough, like I just searched for "stripe" and all visible results were about the payment company.
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.
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
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.
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
For now Arc is staying in Github: https://github.com/Basekick-Labs/arc