Based on a recent, small experiment with HTMX + the Plone API, I think there's some scope to do something useful.
The original idea was based on a thread started by @gotcha `plone.htmx` embryo - #17 by pigeonflight
I think I can afford a day over the Christmas break. Anyone else interested in exploring this?
Adding (some) HTMX to Plone Classic could be a really very interesting alternative to other heavyweight frontends.
we (at work) are planning our redesign strategy on using HTMX, so count us in 
As mentioned on discord before: I could not find a reason why using HTMX because patternslib with pat-inject gave us all we needed ... and its there for free in Classic-UI.
One of the interesting features of HTMX is that it doesn't has any dependencies at all (other than the JavaScript runtime). And thus it runs in every browser. Even in IE11!
Comparing HTMX to Patternslib or other frameworks is like comparing apples and oranges. All they have its use cases. Some of them might even coexist.
But when it comes to versioning HTMX has its advantages due to the lack of a build step.
HTMX feels like it could be a building block for the pat-inject if we wanted to save some code from patternslib.
Using and marketing that Plone plays nicely with HTMX might be a good selling point as well.
Unfortunately patternslib nor mockup are well known outside of Plone devs, but HTMX on the other hand is getting quite popular, so I would expect that plenty of Plone devs would be keen on having it available.
Hey
I would be very interested in some more HTMX experiments.
For my company, I'm currently investigating options for how to build an optimized UI for content managers. We have a completely decoupled backend (classic UI) and frontend (next.js, nuxt, whatever). We have other companies building the frontend. We mostly only provide the backend.
HTMX is quite up heigh on the list of technologies I would like to try.