Next.js Isn't OSS
twitter.com
twitter.com
I’m not sure what the solution is. It’s driven us to consider going back to basics by building our own internal SSG React framework to avoid the intermittent grief of running Next.js on Netlify. It’s sad because there are things we like about all parties involved here.
Meanwhile, Next.js is becoming increasingly opinionated in ways that make it difficult to justify ongoing investment. Turbopack was announced without any commitment for plugins, for example. Hats off to your team for being able to attract and retain talent, but I (and I suspect many others) have a growing sense of anxiety centered around the fact that you can pull the rug out from under the rest of the ecosystem at any time. In exchange for that we're being given incremental build performance improvements and more configuration requirements that we could live without on all sides.
It's untenable to try to secure commitment to migrate a plethora of sites implemented in a given framework to its parent platform when that platform effectively appears hostile towards alternatives.
Vercel's goal is to make it easy to run any of these frameworks. If you prefer an alternative to Next.js and want to switch, from Vercel's platform perspective, that's fine. I'd certainly love to make Next.js better in that instance, but it's not trying to lock you into the box. There's too many good options for frontend frameworks today to do that. And if you're dissatisfied with Vercel, you can also always self-host Next.js.