I’m happy to see permissioned data being rebranded to atproto spaces, I think it more accurately reflects the design of the proposal, which is that a record that exists in a atproto space has two authorities and the space has more authority that the repo. At least to me, this is subtly different than what permissioned data implies; where for a given record the repo is granting a space (potentially multiple) access to it’s data. This tension most clearly shows up in discussion about whether atproto spaces should use a separate addressing scheme.
One implication of giving the space more authority than the repo is that records are bound to the space they were created in. There are many cases I can think of when someone (e.g. creators) wants the same record flowing through multiple spaces, both with long-form writing or, as @iame.li notes, with streaming/VOD. There is an object permanence of the record being shared and atproto spaces lacks this affordance, I guess requiring a new of record in every space.
@chris.pardy.family elaborates on this when talking about the difficulty of adversarial space migration. The governance of space authority seems implicitly feudal and it’s unclear if there is a technical way for communities to migrate. As @chris.pardy.family concludes:
Conceptually a record is in a space, but in practice it’s more accurate to say that a record is authorized by a space. This gives us the opportunity to break the conceptual relationship without breaking the practical one. Spaces still authorize access to records but a record can be part of 0 - n spaces. Moving a record between spaces would be the process of adding an authorizer and then removing the old authorizer. “Deleting” all the records from a space would be removing that authorizer from all records.
Personally I think this points at the direction of a single address scheme where the repo is the authority and spaces are parameter that help determine access to records. That said, I’d love to get a better grasp on what the plan is for atproto spaces to handle these cases or the tradeoffs that have been examined not to support them!