That's true for all the features mentioned in the Next.js 10 blog post: Image optimization, Internationalization, but also Analytics.

Since you mentioned the latter: Did you notice that we wrote a detailed documentation page about how to report those metrics to any service of your choice?

https://nextjs.org/docs/advanced-features/measuring-performa...

Sure, there's obviously a whole UI piece to this as well – but since you interact with Next.js programmatically and there's a quite sophisticated data storage that needs to back all of this as well, Vercel optionally provides you with a pre-built solution.

But again, to be clear: If you want to use Analytics + Next.js elsewhere, Next.js supports reporting them natively – even if not hosted on Vercel.

Perhaps you should link that on the Analytics[0] page instead of directing users to contact the sales team for more information on non-Vercel deployments.

[0] https://nextjs.org/analytics

I'm slightly "worried" about Sapper not being at v1.0 yet but... it's been working great for my MVPs

https://www.youtube.com/watch?v=qSfdtmcZ4d0&t=3s

If someone doesn't beat me to it, I'll come back this afternoon and TL;DR the 22 minute video linked here.

https://github.com/sveltejs/sapper/issues/1429

Not to say Svelte doesn't have great innovative features, but I myself don't still feel that tempted to jump into Svelte. What some of us don't like, is the infamous JS churn where libraries are changed every year because "this new one is better".

[0] https://sveltestrap.js.org/

My team recently evaluated 30-40 CSS frameworks to use with a custom component library, and we were surprised when Bootstrap was the clear winner. Other CSS frameworks have gotten buzz over the years, and we assumed something new and shiny would come out on top, but when we evaluated all of them, we saw sporadic commits, many unresolved issues (e.g., see Bulma issue 3074), small bus factors, inadequate support for keyboard navigation, and other issues.

So using Bootstrap with a newer alternative to React is valid. (Though, we're using Bootstrap with Next.js.)

"We don't have a good answer for that yet, but it is a priority."

I was interested in Svelte until I read this. When they have a good answer, I'll be interested again.

https://svelte.dev/faq#how-do-i-test-svelte-apps

The only thing on the list that's related to Vercel is the analytics platform. Even that is clearly listed as something you can leverage via your own provider.

There is only 1 thing that would make me switch from Elm to Svelte is template as a function. Make it compose like functions! So that you can have as many as functions you want. With current Svelte, these functions are bunch of files .. unfortunately.

The DX is the developer experience. It follows a similar idea. How do you feel about using git? What about Visual Studio Code versus neovim? How do you like using Elm? What don't you like about.

Think about all the libraries or frameworks you use. Some of them clearly offer a better experience than others, these are the types of things that affect the DX.

Take documentation for example, a piece of tech (whether it be an IDE, library, or CLT) with proper documentation full of examples, usage, and recipes is extremely nice. I would favor tech that has better documentation over others, documentation is just one part of the DX. Not the only part, but hopefully you can start to see what can increase or decrease the DX.

To me, how I feel about using (working with) git is git's UX.

How I would feel working on git (were I to do so) would be git's DX.

Now, for a project that has git as part of its expected dev tooling, how I feel about using git in that particular context would be part of that particular project's DX.

You may think that, but it's a standardized term already.