Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I don't disagree with the article, but I feel like the author almost landed on an interesting counterpoint. Author points out they didn't do this in ClojureScript, but writing apps in Reagent (the leading ClojureScript React wrapper) has looked almost identical across many years and many versions of React. Many of the state management epochs have also been avoided, because "manage state" is a core idea in Clojure, and so the stuff we had almost a decade ago is still perfectly fine today.

So, I posit that the churn, while definitely real, is not actually intrinsic.

Right now, at Latacora, we're writing a bunch of Clojure. That includes Clerk notebooks, some of which incorporate React components. That's an advantage I think we shouldn't ignore: not needing to write my own, say, Gantt chart component, is a blessing. So, specifically: not only do I think the churn is incidental to the problem, I don't even believe you need to give up compatibility to get it.

Fun fact: despite all of this, a lot of what we're writing is in Clerk, and while that's still fundamentally an SPA-style combination of frontend and backend if you were to look at the implementation, it absolutely _feels_ like an htmx app does, in that it's a visualization of your backend first and foremost (React components notwithstanding).



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: