Understanding the Three Fundamental Principles of How IPFS Works | IPFS Blog

"InterPlantery File System - IPFS - is a peer-to-peer distributed system for storing and accessing files, websites, applications, and data... to power the Distributed Web - DWeb... by using IPFS to download files from another system, your computer also becomes a distributor... a part of a decentralized network, helping ... distribute information"

• files added to IPFS are split into smaller chunks, cryptographically hashed, and given unique content identifier - CID: a single identifier, an address based on the underlying content - doesn’t indicate where the data is stored
• Content is linked via Directed (connections between nodes only go one direction) Acyclic (won't loop back to the start) Graphs (DAGs)
• Content discovery via distributed hash tables (DHTs)... a distributed system for mapping keys to values. In IPFS, the DHT is... like a cross between a catalog and a navigation system, maps a CID to the peer storing the matching content.
• when you grab a file, you cache a copy and become a distributor until you clear your cache or discard the file, unless you pin the file
• new version = new crypto hash = new CID = new file: changes to a file do not overwrite original (version control hell?). Unchanged chunks are reused to minimize storage costs.
• IPNS decentralized naming system, and DNSLink map CIDs to human-readable DNS names.