Yeah, Vercel has already done this before like with Nuxt, Svelte and other. But I also do want to have a better story for auth with all those frameworks and Nextjs as well.
the team at Vercel has been my biggest inspiration and always reflected many of the reasons we started working on Better Auth. This would allow us to focus more on what made better-auth great in the first place It hasn't even been 2 years since we started but thank you everyone from the open-source community for helping us make an impact in short amount of time. There is a lot to do to improve on open source auth and im really excited to be back focusing full time on building
I think auth is complicated outside of browsers too. But browsers do make some things uniquely confusing, especially cookies and general security primitives are full of footguns
Hey, Bereket from Better Auth here. I started Better Auth to solve this exact issue for myself, and it later turned into a company. It always give me joy to just see others getting the same value from it :) There is a lot to work on, would love to know what we can improve
A couple of weeks ago, I started hacking on a project to improve GitHub, mostly because I was tired of the current experience (like many others) and no one seemed to be doing much about it.
It’s still very early and pretty ambitious, but I hope you like it. It’s open source as well.
Fair concern but I don’t think Auth.js was ever “truly free,” considering it was supported by many companies (big or small) including someone like Clerk even running ads on the docs site.
We started Better Auth with the vision of making high-quality auth (with simple abstractions, great docs, extensive set of features...) and make it accessible to everyone . It didn’t start as a commercial venture, at first it was a purely oss project I created. The reason it evolved into a commercial venture is that we saw new ways to make owning your auth even more accessible and scalable for companies.
The reason we’re bringing Auth.js under Better Auth is that the Auth.js team is moving on, and we don’t want the project to be abandoned, that would hurt trust in open-source auth as a whole. We’ve already seen that happen at smaller scaller with Lucia. If that weren’t the case, we’d actually benefit from Auth.js being deprecated, since we’re effectively the next most people would go for and we wouldn't have to take this risk and responsibilities.
> Does every app using an adapter for Better Auth need to implement every plugin’s many thousands of operations, even if they’re only using basic functionality and a handful of operations?
No, and actually if you really really wanna override the core database calls, we have a way to do so. You just need to write a hook or custom plugin to override the `internalAdapter`.
> Auth.js did everything we need regarding authentication
I don’t think this is true. Any sufficiently complex project has had to add a lot of customization and logic on top of NextAuth to make it even somewhat complete. I was one of those people, which is exactly why I started Better Auth.
> auth in JS is in a sorry state
That’s been the case long before we started Better Auth, it’s the reason we built it in the first place. I hope we’ll be able to change that narrative. But I think what we already have is something other ecosystems can only wish for. Some references:
NextAuth has supported delegating your db for years, companies like cal.com, deel.com and many others use that directly (not just for stateless jwt). I don’t really see the difference here, except that we handle more for you. And of course, If you don’t want to delegate your database, you can keep using NextAuth with stateless auth and we plan to add support for that as well.
There are already many companies with lots of users and revenue using Better Auth from simple auth setups to organizations, billing and what not.
If your question is more about whether we should allow database adapters to be written directly by developers (some people ask that) that’s just not realistic at the scale of what we handle. No one is realistically going to write hundreds of queries manually
1. We won’t sunset Auth.js unless we’re confident that anyone currently using it can migrate to Better Auth without any issues, which is quite difficult right now. So we don’t expect to do that anytime soon and chances are we will never require everyone to migrate.
2. The features we offer through plugins don’t exist in NextAuth, so that shouldn’t be a problem. You can use the core library for almost all of NextAuth’s features, and we provide most plugins first-party. Of course, you can choose not to use a plugin, write your own, copy and modify one, or only use the first-party ones we provide. We handle the database so you can own your auth without writing the logic yourself.
3. Auth.js hasn’t been actively maintained for a while. Our main reason for bringing it under Better Auth was to avoid a sudden deprecation, as that would directly harm the open-source auth ecosystem by eroding trust. Something we’ve already seen happen on a smaller scale with Lucia Auth.
We’ve had early access to it for a while now, we’re already running a lot of performance critical workloads on it and it’s been working wonderfully. Congrats sam and the team on setting a new standard for what highly performant managed Postgres should look like :)