| 30 May 2021 |
 owocean | and the json contains the filenames and the corresponding cid | 23:44:19 |
| owocean | ah | 23:44:20 |
| owocean | wow | 23:44:23 |
 kallisti5 | right :-) | 23:44:24 |
| kallisti5 | this is where the deduplication comes from | 23:44:37 |
| kallisti5 | so.. file is 23512345314t6y2345ty4526245564 | 23:44:44 |
| kallisti5 | if you have file.jpg with that hash, and it exists in 100 different directorys... you only store it once. | 23:45:02 |
| kallisti5 | this is all why the webui folks have trouble lol | 23:45:24 |
| owocean | because you can only get one hash from a file's content | 23:46:16 |
| owocean | right? | 23:46:24 |
| kallisti5 | right. | 23:46:31 |
| owocean | i read somewhere that the CID isnt just the file hash | 23:46:33 |
| owocean | theres other stuff to | 23:46:36 |
| owocean | * theres other stuff too | 23:46:38 |
| kallisti5 | yeah.. lots of structure | 23:46:44 |
| owocean | which explains the Q at the beginning of every cid | 23:46:49 |
| kallisti5 | yeah. it's called a "multihash" | 23:46:58 |
| owocean | whats that | 23:47:08 |
| owocean | i read about it before but i didnt understand anything | 23:47:30 |
| kallisti5 | https://github.com/multiformats/multihash | 23:47:33 |
| owocean | :< | 23:47:32 |
| kallisti5 | it's a self-describing hash | 23:47:48 |
| kallisti5 | essentially the IPFS folks can adjust the hash to be other formats at any point (SHA256, blake, whatever) | 23:48:22 |
| kallisti5 | and have backwards compatibility | 23:48:30 |
| owocean | superb. thinking 200 steps ahead | 23:48:52 |
| kallisti5 | ipfs itself is NEAT. it just takes quite a while to get your head around | 23:49:13 |
| owocean | uh, so i ran ipfs pin ls and got nothing. the file is still on my node though | 23:49:24 |
| kallisti5 | try ipfs files ls | 23:49:38 |
| owocean | yeah that listed it | 23:49:48 |
| owocean | is that the cache? | 23:49:51 |
