I think familiarity is a big part of why things catch on. If something is too different to what people know there will be only a few people who want spend their time learning it. And it would have to be revolutionary for these people to be able to convince others to also learn it.
It would have been helpful if in the video they would have discussed how an alternative could have even looked like and why it would be better. This is a demo of Project Xanadu, the system Ted Nelson envisions where he shows how it could work. He seems to propose that it would be hyper interconnected for every user of the system and every piece of media in it (another interview where he describes it). I'm not sure something like this could reliably work at a scale similar to the internet (he claims his system could have been the internet had they delivered it earlier) and also I'm not sure how it would work for what people actually want to do with the internet in addition to reading documents. Companies also want a certain control over the work they publish so I don't think they would like a system that connects their work to everything else. And you also have to keep in mind that there are people who want to actively do bad things so I am not sure how a hyper interconnected system could protect its users from bad actors.
Thanks for the links! Odd he included that video but didn't include the points you/he made beyond the criticism.
Edit: Watching the video you linked that is included in the posted video, not sure how big a deal Xanadu is. I wonder how this would compare to git history and tag maps such as Obsidian.