So, yeah, I like what you’re saying ’cause I pretty much agree on every front. This guy Nelson’s thought something up that isn’t yet achievable, and may not ever be achievable. See, on the one hand things like math are pretty easily explained, and a systematic method of doing so like Xanadu could be useful in organizing and presenting that knowledge. But what about the scholarship surrounding Moby Dick? Could we effectively collate and organize the scholarly discourse in a satisfactory manner? I find it highly unlikely; at the very least, we’d be unable to remove an inherent bias towards one or another position.
Regarding things that deserve a place within human knowledge, but are not as quickly defined and categorized as things like country capitols, types of clouds, and the planets, I think any project would run into the curse of Wikipedia: too little research and hasty assumptions or assertions. As a philosophy student, I know that any discussion on a philosophical concept on Wikipedia immediately runs into these issues: soundbites from famous people make the cut, but the scholarly discourse on something like Nietzsche’s theory of Dionysian faith is poorly represented. And I don’t think that’s because Wikipedia sucks; it’s great! But it’s facing a problem that any “Xanadu-like” project will face, which is inherent user bias, an inherent sense of “correct” knowledge versus “incorrect”, and what you called vision being imposed from the outside, or perhaps the top-down.
Personally, I think it’s an ideal that can’t be achieved because it wants to create an essentially static system; it wants all the beings without all the becomings, at least in my perspective. Or, rather, it wants to turn all the becomings into beings as well, and turn human discourse into a digital monument.