It's a very interesting concept in the current climate. The idea of sharing content as atomic and self-sufficient units of data, much like you would share the raw idea itself, but with the ability to have layout, rich content and possibly even interactivity. A web of ideas rather than ideas on pages with servers. The disintermediation of the web.
That's why I let that part of the project be the focus!
I also hope the interest it has generated will inspire others with more knowledge and experience to make better versions or variations that are more practical.
A deeper implementation would be a browser designed for this.
Some hard problems. E.g. URLs provide a consistent place to go to find varying content. If the URL is the content, then it cant vary. How does one solve content discovery without resorting to hosting?
PS: The first person who says blockchain buys the beers.
Blockchain would be silly, there's no need for a globally-consistent view just to have search.
Distributed search has been around for some time, e.g. eMule uses Kad[1], Tribler[2] has a solution based on Tor, but the most appropriate here is probably YaCy[3]. Of course, all of those are technically "hosted", insofar as p2p involves peers acting as servers. But I bet someone has made a f2f[4] version, meaning you'd only serve to a few, hopefully trusted peers.
I have an example that I haven't yet uploaded of an entire (small) site in one URL. If the links are all to be URL Pages, the site has to have a strict tree structure, otherwise there is a circular dependence/chicken-egg problem.
I could see this issue becoming problematic with a browser implemented based on this idea, unless I am misunderstanding something.
I suspect that's why this item is so popular.