Oddbean

it is only his lack of adequate support of outbox model and the very slow response of the client to auth request envelopes from my relay (sometimes it sends a dozen before the client finally sends back a signed auth envelope) that i don't still have #coracle as my daily driver... the outbox model is basically not negotiable for a relay dev testing client, and without that it doesn't even matter if NIP-42 is working, really, outbox is > NIP-42 support this is my experience as a relay dev, and hodlbod recently got into this arena with his own fork of khatru but charging for access is not on his priority list, and it's only one guy, so i don't like to be too dark about the things missing, as it's trivial in comparison to everything else, primal, amethyst, iris in terms of completeness of their implementation of the critical protocol features

this is also why i am constantly berating fiatjaf, pablo, and others about their work on things that don't address or disregard the importance of making the protocol self-funding without self funding this thing will die like every VC funded attempt at social networks, it's really serious and these people just don't get it, comfortable with their fat grants

bitcoin is now bigger than the swiss franc because it is a self funding distributed system i can't say enough towards this and really want people to get hold of this and run with it because the sooner this protocol is its own circular ecosystem between users, devs and operators the sooner we are outside of the gates, right now we are just like a billy goat bashing against a fence that is strong enough to contain him self funding is like a set of heavy duty bolt cutters

We should start a company, issue a token and build a revenue model. 1. A client cannot start charging fees from the get go. It has to be a freemium model. 2. The client will abstract away all the underlying costs. Maybe a $1/year could be the minimum. 3. It cannot be ads based. It'll misalign the incentives again.