Comment by devalexwells

I resonate with much of what you've written in here. In particular, the cultural costs. I work on a "frontend platform" team at a mid-sized company and spend most of my day trying to get the React/Next ecosystem to play nice for our engineers. If I rearchitected it from scratch, it would not use Next, but I digress.

Some critiques:

1. As others have mentioned, lots of abstract in here, not so much data. That's a hard sell to engineers. You're going to get a lot of theory responses, which maybe is what you want at this stage, but it might help to make that clearer if so.

To avoid this, you might consider refocusing the narrative. I have personally found that while appealing to Developer Experience (DX) is the fad, debating DX is an uphill battle. IMO we should appeal to something more concrete: User Experience. The discussion has to start and end with users. DX is part of it, but often over-represented. If I can sit down with someone and show them X technology is failing our users because Y fundamental flaw, the conversation is no longer on frameworkism. It's instead on how you solve problems for your users effectively. See Alex Russell's discussions on this, as divisive as they might be [1].

2. The why is clear. The what and how are not. Clarify what you're doing that is different. For example, Astro is also standards-based. Astro also approaches the problem from progressive enhancement and fundamentals over frameworkism. It seems like your focus is on separation of concerns and design principles? While I disagree on the dogmatic approach re: separation (I personally like components and composition), it would help to highlight your diff.

* I mention Astro, but this is true of other newer approaches, e.g. 11ty, Qwik, Enhance.

3. In another comment you claim that [in Nue] 90% of your codebase becomes CSS. I get that this is coming from the reduction of JS to do what HTML and CSS should, but it feels... disingenuous? At scale, no frontend or fullstack engineer I work with is spending even close to the majority of their time on CSS, regardless of stack. The focus IME is on building features, parsing/integrating data, and everything tangential to the feature (architecture review, testing, docs, etc).

Just some thoughts! I'd be interested in hearing more in the future.

[1] https://infrequently.org/2024/11/if-not-react-then-what/