rich1.0fastr wrotefastr (npub1z3…mcayu)https://yabu.me/npub1z3gdcd5zp0z0fudczjf4grlrkga6ls3yhgr54ph78ufm5lkteywsfmcayunjumphttps://yabu.me#[0] #21e8 protocol by kiwi Mark Wilcox
some transcript for you below the video 🤙🏽
https://youtu.be/RTToM3JeeyY?t=1675
28:00
demo to introduce the idea of ranking content
21e8 was in blockchain ecosystem
until we built up our own content distribution
time stamping, not needing blockchain consensus
28:25
no wallet integrations
we're creating a specific kind of content
proof of work indexable analyzable
so when you get raw bitcoin blocks
those objects fit in our wider data networking tech
without any need to store state
29:02
the internet is a datagram network
it doesn't hold state, NO STATE
it just shunts packets around quickly
routing functions are simple and robust
29:22
tools for better
* ranking
* ordering
* sorting
* caching
29:30
we don't need server network
to store those copies
29:38
extending what internet routing does
to get messages to you in the first place
like all BTC transcations go over IP
internet protocol
29:57
we extend that to the rest of the stack
so there's no arbitrary cutoff point,
no need to go from tcpip into http server
no authenticating with a dns look up
no authenticating into the database
less keys keys keys sign sign sign cruft
in modern software platforms
in modern business models
in all above like cloudfare aws etc
30:36
simplify how
* data
* routing
* communication
* storage
* scheduling
fit into one "1" single thing
similar to internet packet flow
not just physically routing server to server
but also within one node, whether
| cpu
| gpu
for whatever graph i/o processing what you wanna do
within the same structure
31:13
once you scale to huge communications constraints
(sending a peer-to-peer messages is not efficient)
that becomes the real bottleneck for data movement
31:29
if you have to be moving the data,
you've already incurred too much of a cost
to pay for the proof of work essentially
so proof of work is there
🤙🏽