So while I agree that the business case for web apps incentivizes mediocrity, there is still plenty of hope for everyone else. If your startup has a really great web app with awesome UX, you will have a pretty big advantage.
I have 2 examples of web apps working really fucking well:
Missive (https://missiveapp.com/) is an email client web app with features that should be standard in every other email client. I run it via Firefox and it's the best damn email client I've ever used. It uses JS extensively but they use it very well.
Migadu (https://migadu.com/) is the absolute golden standard of email inbox services. It's not fancy but oh my days, does it work and it's so damn easy to use. Every time I have to set up emails I get happy because these guys spent exactly 0 seconds on trying to appeal to people who equates parallax-scrolling with competence. Migadu does not use JS as far as I can tell.
As a counterpoint to both, Office 365/Outlook is a gigantic, floppy donkey dick of a web app that makes all sorts of assumptions, has 3 different interfaces and is generally just a pain in the ass to use. Somthing as simple as logging in/out is a muddled, technical mess that makes it a god damned chore to change accounts.
Web apps get the user 80% of what they want with 50% of the cost, so the business case is clear. That's why these apps are proliferating. They are not proliferating because they're good, as good, or better. They're just not.