@doesnm As far as I know, these are just an implementation quirk.But repeated parameters are allowed by URL (WHATWHG) standard: https://url.spec.whatwg.org/#urlencoded-parsing

@silverpill Yes, I think there are reasons to want to do this even in a pure P2P setup - such as having one copy on a mobile device for portability, and another on a home device as a backup in case you lose your phone.Agreed that it's worth looking at Iroh - even if it's not useful in the long term, it'll be worth trying to work out what the dependencies on HTTP actually are, and what it'll take to separate ActivityPub from its transport more generally.

The NomadPub intro page now includes a "Research" section:https://codeberg.org/ap-next/ap-next/src/branch/main/nomadpub.md#researchI listed the most important open research problems there: key management (alt-DIDs), alternative transports, E2EE and generic servers.#NomadicIdentity

I managed to send an encrypted message from one Mitra Mini account to another.The gateway should run the latest development version of Mitra.Spec: https://codeberg.org/silverpill/feps/src/branch/main/0806/fep-0806.md

Regarding the validity of at:// URIs in atproto:https://bnewbold.leaflet.pub/3mph4hzvbdc2vWe have a similar problem with FEP-ef61 'ap' URIs. In their canonical form, they are not valid RFC-3986 URIs.I think if atproto devs decide to move away from ://did:.. syntax, we'll have to do that as well.#fep_ef61

FEP-ef61: Portable Objects has been updated: https://codeberg.org/fediverse/fep/pulls/872The ap+ef61 URI scheme is now allowed, while ap remains the recommended one. This is to ensure compatibility with @fedify whose maintainers decided to use the ap+ef61 scheme until the specification is finalized.#fep_ef61 #nomadicidentity