2019-03-07 19:28:49
A plan that we have been kicking around for simple payments and auth: https://docs.google.com/document/d/17Xx9Y_lxk5XdYsuXJA0IYhxcoI15OxCvTZgu876JmwA/edit?usp=sharing
Fireduck
2019-03-07 19:45:03
Concept looks simple enough to work great. Nice.
Tilian
2019-03-07 19:45:20
Plus that authentication mechanism allows for a _lot_ of possibilities.
Tilian
2019-03-07 19:46:37
yeah, no one wants to have usernames and passwords
Fireduck
2019-03-07 19:46:44
and site operators don't want to deal with them either
Fireduck
2019-03-07 19:46:53
It's stupidly redundant for any crypto project, since we're using pubkeys already.
Tilian
2019-03-07 19:46:54
but not everyone wants to use google or facebook oauth either
Fireduck
2019-03-07 19:47:01
right
Fireduck
2019-03-07 19:47:10
I would 100% use that. Not using Google and Facebook is a pain sometimes.
Tilian
2019-03-07 19:47:18
"A single client/seed would use different addresses for identity with various verification gateways.", so it just does keypool+1?
Tilian
2019-03-07 19:47:37
I imagine that opens up some sort of DoS vector, though probably easily throttleable.
Tilian
2019-03-07 19:47:47
It would be more like HD m/2338/hostname/0 or something
Fireduck
2019-03-07 19:48:04
where we do some magic to the verification gateway hostname to turn it into some integers
Fireduck
2019-03-07 19:48:24
ah, of course, sounds good
Tilian
2019-03-07 19:48:37
that way restore from seed works as expected
Fireduck
2019-03-07 19:48:53
and that address uses for identification with the gateway doesn't actually appear on the blockchain anywhere
Fireduck
2019-03-07 19:54:45
`Payments could go from the user directly to addresses controlled by the content provider (preferred, less trust)`
Tilian
2019-03-07 19:54:53
so the gateways would simply "relay" the payment request from the content provider?
Tilian
2019-03-07 19:55:03
or how would that work w.r.t. address privacy?
Tilian
2019-03-07 19:55:37
So the verification gateway would have an xpub for the content provider and generate new addresses as needed
Fireduck
2019-03-07 19:56:07
So the gateway would request the payment, validate it, it just wouldn't have the keys to spend it
Fireduck
2019-03-07 19:56:08
right, of course
Tilian
2019-03-07 19:56:15
that makes a lot of sense
Tilian
2019-03-07 19:56:28
and the gateway could just do this on set intervals
Tilian
2019-03-07 19:56:30
solid.
Tilian
2019-03-07 19:59:05
so I could just run a very stupid payment gateway and use it as a "pay to DNS" kind of system, nice
Tilian
2019-03-07 21:59:31
also, would this be "roadmapped" (broadly speaking) before/after channels?
Tilian
2019-03-07 22:00:40
Not sure. So many things to do and limited time.
Fireduck
2019-03-07 22:01:18
Kinda depends on interest
Fireduck
2019-03-07 22:13:50
Not always easy to gauge
Tilian
2019-03-07 22:13:59
yeah
Fireduck
2019-03-07 22:14:05
although, minimum viable product on simple payments/auth seems easier
Tilian
2019-03-07 22:14:28
payments will generate interest
Rotonen
2019-03-07 22:14:51
channels will bring comparison to ’mastodon on the blockchain’
Rotonen
2019-03-07 22:16:14
having a easy-to-use method for recurring payments in crypto would be a first, as far as I'm aware
Tilian
2019-03-07 22:16:21
emphasis on easy-to-use
Tilian
2019-03-07 22:16:21
also i suppose ’pay to dnssec’?
Rotonen
2019-03-07 22:16:49
reminds me of some ongoing discussions with BCH developers, where they want a "friendly handle" system
Tilian
2019-03-07 22:17:01
I never understood why they come up with complex on-chain mechanisms if you can just build on DNS (+DNSSEC)
Tilian
2019-03-07 22:17:07
An important note is that while each channel would be its own blockchain, they are not in general on the main blockchain at all
Fireduck
2019-03-07 22:17:43
exactly, ’mastodon on the blockchain’
Rotonen
2019-03-07 22:19:21
@Tilian NIH is a primal human condition, no reason crypto would be immune to it
Rotonen
2019-03-07 22:20:49
I know very few communities where it's more prevalent than crypto
Tilian
2019-03-07 22:22:01
but hey, Satoshi's release had pay to IP
Tilian
2019-03-07 22:22:05
not sure if it resolved DNS though
Tilian
2019-03-07 22:30:30
incentives for TCP spoofing, now that’s a blast from the past
Rotonen
2019-03-07 22:33:51
ha, yeah
Fireduck