I've been doing IT for decades. Many times I've come across something that seems stupid yet also has a non-trivial number of supporters. When this happens (and it will many times), don't just continue to think it's stupid. Question if you are stuck on some paradigm or mindset that makes the thing seem stupid to you. That might call for a good head shake, a walk in the park, or a weekend-long bender to open your mind to new ideas. You don't have to jump on their bandwagon, but at least you'll increase your understanding and strengthen your position.

Maybe it’s for a use case I don’t understand… but I don’t see why anyone would use this for a modern, complex, richly interactive web-app. I also don’t really see any examples of a high quality major production site using htmx.

idk sometimes it turns out to be a pretty good approach for some problems:

https://htmx.org/essays/a-real-world-react-to-htmx-port/

Just reacting to the executive summary: where are the tradeoffs? The port decreased memory usage on the client significantly, by how much did the server's burden increase? How have their hosting costs changed now that less work is being offloaded to the clients?

idk they didn’t say, but the server side delta of JSON generation vs HTML generation is small in the overall scheme of things and HTML-style apps tend to be less chatty because they push you towards the one-request-per-view ideal

I'm already a fan of htmx, you don't have to sell me anymore.

Right? Someone show this kid how we used to make web apps back in the day, using hardware 1/100th as powerful. Turns out you don't need to send megabytes of JS to the client to have it render a table