I wonder what the technical hurdles would be to building a USENET client in a browser these days.
If not directly implementable, a USENET-to-HTTP proxy running in the cloud (to address the issue the author identifies of "didn't need to be installed") could obviously be done (and has been done, or near to it, a couple times).
My understanding is that the client isn't the technical challenge, it's the (federated) servers, and it's more of a business challenge. In particular, "who pays for them." Usenet servers used to be something ISPs maintained & your ISP subscription would include (usually) a certain number of hours of internet access per month, an email account, access to USENET and perhaps a couple other things.
As it fell out of favor in the mainstream ISPs stopped supporting it/paying for it & it became a niche service to pay for separately, if I understand correctly, and the only people willing to pay (by and large) are people sharing pirated software, media, etc.
For a while, I used a commercial NNTP service which was free for text-only newsgroup access. I think I stopped once I started university, as there were better distractions.
With a quick look on a partial NNTP server (requires registration), the only groups I used to look at that are still active is the old/retro computer one. Most of the posts are people still using these computers day-to-day, and finding problems with Javascript-heavy websites or outdated encryption.
I think the biggest problem is who is building it, and why. Open source is certainly a boon, but when insidiously magnanimous corporations start contributing and start spreading their influence, that's when rot creeps into the project.
If not directly implementable, a USENET-to-HTTP proxy running in the cloud (to address the issue the author identifies of "didn't need to be installed") could obviously be done (and has been done, or near to it, a couple times).