in fact, now that im thinking of it, the UX can imply sync across a set of git urls. that way you can clone any url in the set, make a fork, push it anywhere, submit a PR, reference a branch, include a comment and title, and that branch should be "assumed sync"

i think that's key to keeping it simple.

here i wrote the specification for you of all the features that the command line needs:

https://cli.github.com/manual/

there's obviously no need for repo hosting, git is already decentralized enough, and part of the NIP could be to specify one or more repo URLS to use

yeah, its more necessary with nostr. although i wish there was a "mute" feature. follows also dictate who can dm you, and also the reach of you cli as it discovers relays in your contact list, etc.

this is why i think its critical to standardize on an "algo server"

the standard can be VERY simple:

--------------

NIP-99: Relay Aggregator and Filtering

Clients MAY choose to use a Relay Aggregator and Filtering service

Events can be ranked with an optional ["r": ] tag appended to the events by the relay.

This functions as a "single inbound relay" with a feed of events.

Clients SHOULD display this feed in a separate area, tab, window, or other location.

Clients MUST allow the user to enable/disable these Aggregators

Clients MAY use the information from these Aggregators in any way, including modifying follow feeds, ranking comments, etc.

--------------

Done.

humans aren't a hairy as we were 1000 years ago, because we invented clothing

we aren't as strong as animals with similar mass, because we use weapons

we cook our food, so we have lower gastric ph

tools have created our biology

it's obvious

same. this is ridiculous. good luck finding out whether or not i sent someone in iran sats. i'll wait.

you can't be reasonably held accountable for knowing that based on an lnurl

no thanks. i'd rather just host my lnurl anon and make sure it cant be traced to me

algos can work fine. just build a relay that "does an algo", it's already baked into the protocol

what's even more critical is developing credible, standardized, permissionless reputation system. there have been other distributed protocols, like email. what destroyed them was spam.

https://cfenollosa.com/blog/after-self-hosting-my-email-for-twenty-three-years-i-have-thrown-in-the-towel-the-oligopoly-has-won.html

if we don't have a *distributed reputation system*, NOSTR will eventually go the way of SMTP

big relays will start whitelisting other big relays, users will be wary of any relay not in the top 10, because they're filled with crap DM's that will overwhelm them. Some relays will get banned for hosting illegal content and not taking it down. Etc.

If we tackle this head first, and assume adversarial environment and begin to develop distributed WOT systems now, we can prevent centralization

i need this on my tv:

https://shopify.github.io/spatial-commerce-projects/WonkaVision/

just need to add private channels

then you can invite a bunch of people to a private channel

and all the posts will live there

I

some clients just put them one on top of each other in a long list

I would much prefer swiping or collageing

you know they still do

everybody's busy building fun new things on top of nostr

so few people are worried about the censorship resistance and spam prevention and web of trust and all the things that are needed to keep the protocol from falling apart

tragedy of the commons

without a distribued protection system, all commons are either destroyed or become rules by tyrants

building something on nostr is probably the faster way to make bank

http is stateless

nostr is stateful

anything you can build with http that requires state can be built better with nostr

i expect the entire internet could transition over time

there's a wip nip for this now