The Early History of Usenet
cs.columbia.edu
cs.columbia.edu
Can decentralized services work on the modern Internet? Even setting aside how spam and illegal content destroyed Usenet as a discussion medium, in its late stages it became a playground for kooks and trolls at the expense of legitimate posting. Killfiles didn't work as the bad actors could easily change names and email addresses; get them thrown off one server and they'd just sign up on another one. The only way to ban someone from a newsgroup and make it stick is with a court order - and that happened at least once.
Another post from the same blog details why the troll problem was technologically impossible to solve: https://www.cs.columbia.edu/~smb/blog/2018-02/2018-02-23.htm...
The first time I can recall anyone tried widespread cancels in Usenet was Clarinet, and it didn't work very well.
For access to the textual groups, you can use Eternal September (https://www.eternal-september.org/) or AIOE (https://news.aioe.org/) with a newsreader (https://en.wikipedia.org/wiki/List_of_Usenet_newsreaders) to see what discussions are continuing to happen.
There are still some people there, and talking about stuff. It's very much like hacker news, but with much less people :)
A big reason why Usenet is still alive and kicking is thanks to piracy. The same applies to IRC (XDCC), but probably to a lesser extent.
Mutually reinforcing cycles that led to the current state of Usenet as a seedy backwater where it was once an essential part of the Internet (and the UUCP network before it).