Here's a real world example of this. I wanted to read the reports you referenced in this thread. My preference is to remain as private as I can on the Internet, I'd rather not reveal to your organization that I visited and read these reports. Yet, I filled out this form so I could read them. 😅

Is Dan right?

This privacy analogy is spot on and more plainly reveals the dissonance between behavior and beliefs.

Most people do care about privacy. Yet within the confines of a product experience, the choices being made don't reveal this, because most people optimize for expedience.

I didn't take it that way. The gist of that thread was to ask Dan to stop making sweeping statements about people not wanting decentralization based solely on Farcaster experiments. I took it as "Dan, stop pretending you have the magic eight ball on all of society".

If anything it was pro research.

I meant, as a dev of that onboarding flow, can we add a service/key to the DID before it is created?

I don't want to deal with the extra friction of updating it after the fact, because it requires getting a code via email.

When I say PDS-as-a-service, I mean, we use it behind the scenes, but the user doesn't know or care. What's important is that we offer the same backup/exit features as BSky.

That onboarding is smooth! In that flow, how much control do you have over the DID that is created (we need to add to it)?

Asking app developers to to host the PDS may make sense for some apps, but not all. If I am building an app that only uses the AT Protocol identity and never store or access PDS data, asking me to host the PDS for newly onboarded users is a huge lift.

Who's build PDS-as-a-service? We'd pay for it.

It seems to me that this whole thread can be boiled down to this:

Services that only want to use AT Protocol for account identity need a way to to onboard new-to-atproto users without taking on the burden of hosting the user’s PDS.

crosspost from:
discourse.atprotocol.community/t/the-case-f...

It was a pretty good game for a neutral. Fun to watch Messi do Messi things. The third assist was lights out.

Whitecaps still had a terrific season even tho they didn’t win the final.

I refuse to pay attention until I learn whether or not Italy 🇮🇹 will be playing.

thank you!

I think account abstraction took time to emerge out of pains people had, so I understand why it took time.

But not supporting r1 was a carry over from the "NSA has a backdoor" concerns that started with BTC.

Fast forward 10 years and no one cares if there is backdoor.

Wonder how much sooner we would have had mobile friendly apps on Ethereum had secp256r1 been available on day one?

What do you mean by “auth app provider”?

I understand this perspective!

are you ok running ON crypto rails vs. sending/receiving crypto?

There is a project to send/received $USD on the ZCash network, but I don't think it is ready for public use yet.
zechub.wiki/zcash-tech/z...

When I hear words like "Chaumian ecash" and private payments, my first thought is that you should chat with @zooko2.bsky.social

Not saying you should use ZCash, but I'm not saying you shouldn't. 😬

If you do, I'd love to incorporate it into the ATProto native-messaging protocol we're working on!

Great article! I suspect there are many developers coding with no pants.