Phillip Hallam-Baker wrote on 22/01/2021 01:13:
Deregistration cannot take place in my technology model.
this wasn't fully clear either from your previous descriptions. There's plenty of address space to be able to do this sort of thing, but building a registry on this basis is a design choice which commits to ensuring long-term registry inaccuracy due to inability to perform garbage collection.
A traditional registry would normally be designed on the basis of providing an accurate mapping between resource and resource holder. If that mapping is broken, then one of the core components of the registry function is no longer present. The ietf would want some fairly unambiguous and clearly articulated reasons to justify why this was a wise long-term approach to finite resource allocation, and whether the term "registry" was applicable.
Nick
