Issue #51: Like A Rock


🎵 Didn’t have a care, working for peanuts, not a dime to spare, but I was lean and solid everywhere… 🎵

While many people wonder whether we need metaframeworks at all and whether they bring joy or contamination into the webdev world, others work hard and thoughtfully to create efficient and robust solutions for full-stack web development challenges. We will look at this eternal party from different sides today, and everyone will (again) be able to make up their own mind.

The Good

The SolidStart team has stabilized version 2 of the metaframework and published the official migration guide. The announcement says the update mostly moves to the Vite ecosystem and a Rolldown-powered toolchain (and, most importantly, Vite’s Environment API, enabling versatile options for targets like client, server, server functions, and whatnot, as well as more straightforward deployment setups), discarding unnecessary additional abstractions.

Solid v2 is moving forward quickly. It represents the next generation of Solid, with an exciting new reactive foundation and new possibilities for the framework and its ecosystem. We are incredibly excited about where it is headed, and we will have much more to share about it soon. With the modernization work completed in SolidStart v2, we can carry this foundation forward as we move towards Solid v2 in future releases.

Interestingly, Angular also just recently officially switched to using Rolldown by default for production and server-chunk optimization starting in Angular v22.1.0. This step is a huge shift for Angular’s conservative toolset toward a Rust-powered build pipeline, moving further away from the conventional, familiar Webpack.

The Bad

Meanwhile, people are getting more and more intimidated by the complexity of the modern webdev ecosystem (not everyone likes even neatly engineered but overly bloated solutions). David Poblador has beautifully described that in his piece on the descent of frontend.

Somebody hit a genuine problem, built a fix, and the fix created the next problem, which got its own fix, and the sum of two decades of reasonable steps is the magnificent, exhausting cathedral of madness you’re staring up at today.

Ollie Williams seconded some parts of that (perhaps unintentionally, but no less persuasively) by focusing specifically on frontend build tools and their role (or, rather, lack thereof) in modern web development.

Frontend build tools have improved over time, yet developer satisfaction with them hasn’t changed since 2016, remaining at 3.6/5. Configuration, excessive complexity, and performance rank as the top build-tool pain points. When asked “What aspects of JavaScript do you struggle with the most?”, dealing with build tools ranked higher than async code, security and error handling.

And even from a practical standpoint, there’s a consistent tendency towards simplifying the stack of technologies for building websites, with a recent example being the TanStack team getting rid of React Server Components on their official website.

Maybe there are many more use cases where that benefit justifies the machinery and we haven’t run into them yet. I’m happy to be proven wrong, but our strongest RSC use case disappeared as soon as the dependency got small, and I don’t think that’s nothing.

The Noteworthy

Another interesting and long-anticipated website rebuild story is the news that AnalogJS rebuilt its official website in… AnalogJS, marking an important dogfooding milestone that is a good signal for people who want to use the framework (I personally consider it a must-have for any metaframework and the minimum viable validation, in fact). The new version 2.7.0 of the metaframework brings a lot of awesomeness to support that.

The Next.js team didn’t rebuild anything, but they kinda called all the stars aboard for the new minor version announcement (which is probably the most interesting part of this announcement, to be honest). There are lots of technical improvements worth checking out anyways and multiple DX and AX enhancements that will make many consumers happier and reduce resource costs.

The Astro team just did the same (with less AI fanciness, thanks JavaScript gods!) and also dropped some news from the past July, which is a great read, as usual.

The Quasar team delivered an important technical release with performance improvements in version 2.24.0, as did the Hono team with version 4.13.0. Releases 1.6.0 and 1.7.0 of RedwoodSDK brought multiple security fixes for the metaframework along with a bunch of technical improvements.

The Svelte community is actively working on a new major version of SvelteKit and, at the same time, preparing for the November Svelte Summit in Ljubljana, Slovenia. So, even though the ticket price might bite, you should definitely check out this beautiful place and such a nice occasion.

As you can see, there’s still a lot to learn and a lot to do by the end of this summer, and even more to prepare for before the fall begins. So take your time, choose your weapon, and let’s get rolling on through this awesome metaframework web development journey.

đź‘‹

Found it useful? Consider subscribing to the newsletter.No hidden catch, no strings attached.