Back

@JamesGleick @molly0xfff @pluralistic

Shouting out @aral and the Small Web as well: https://ar.al/2020/08/07/what-is-the-small-web/

@currentbias @JamesGleick @molly0xfff @pluralistic @aral After reading the article, I think that P2P technology like Bittorrent, IPFS, Tribler, Zeronet, Freenet, Webtorrent, Cabal, Holepunch and other technologies are the logical conclusion of the small web. With P2P technology, if your file goes mega viral, you don't get a giant web hosting bill because everyone else helped share the burden with their own connection.

@JoeBecomeTheSun @currentbias @JamesGleick @molly0xfff @pluralistic The goal isn’t to go mega viral, it’s to encourage human-scale networks and communities. (1/5)

I initially started with the same assumption and prototyped the early versions using the hypercore protocol (e.g., see https://github.com/indie-mirror/hypha-spike-multiwriter-2/tree/master). I thought at the time that a replicated directed acyclic graph could be core of such a system but it has a huge downside: the data structure must be replicated on all nodes. (2/5)

Add to that the universal issues with peer to peer (findability and availability) and you can see how my design evolved to take advantage of the inherent strengths of the web (an always on node at a simple address – a domain name). (With any highly available peer to peer system you still need an always on fallback node for findability and availability/relay so why not make that a node you own and make that the core of the system?) (3/5)

The issue, of course, then becomes: how do we make it as simple as possible to setup and use such a place, without any technical knowledge. How do we make it as easy to own your own place on the web as it is to sign up for a Facebook account? How do we make it ten seconds and zero technical knowledge? How do we make it so these places can communicate with each directly. In other words, how do we build a peer to peer web (the Small Web)? (4/5)