There's a whole spectrum of options between super-fat React-in-Electron and expert-optimized platform-specific (perhaps GPU-rendered) apps. VS Code and Zed are close to opposite ends of this spectrum.
- slim down to a more efficient, non-virtual-DOM web stack (e. g hypermedia)
- move to a wrapper around native webview (Tauri)
- use one of the excellent cross-platform frameworks (Flutter, or that new one just open sourced by Snap)
For WhatsApp specifically, I don't understand why a company with the size and resources of Meta can't support native versions for Windows, Mac _and_ Linux. I think many people would accept non-feature-parity with the web/mobile versions, in exchange for a tight, reliable desktop messaging app...
There's a world of difference between saying "our stuff sucks" vs "here are the specific ways our stuff isn't ready for launch". The former is just whining, the latter is what a good PM does.
If you're interested in this space it's worth looking at data-star.dev. who takes hypermedia in the opposite direction - one endpoint per page, then push updates to components over SSE.
My worry with MESH is that many endpoints might become a (sorry) mess...
I remember seeing this on the HN front page a while back! I subsequently forgot all about it, of course. Helps to see it again a second time, especially with the new context - I might give it a try next, see how I go. Thanks for the reminder!
It does require a mind shift as generally you have one connection morphing in all your updates and do CQRS.
The connection per component model that mesh uses is fine until you have concerns that cross across components (this was an issue hotwire also ran into before they introduced morph/refresh).
Instead you have one endpoint per page re-render the whole page via morph on every change. You still have backend components they just send requests up and get their updates via the sse connection for that page. Kinda like view = f (state) just over the network.
It’s easiest to start with Pico (especially classless) and the make the decision on where to go, when you decide to add classes whether it’s continue with Pico, easily head towards Tailwind, or something else.
Chesterfield's Crooked Spire in Chesterfield | Atlas Obscura https://share.google/ABVhVgtzQ8KxnO27L
reply