I would like to clarify what current (Lima) AENS is capable in aspect of pointers.
Pointers are things that map name to data and they live in blockchain state.
We allow the data the be encoded as: account, oracle, contract or channel types.
What’s great about AENS, is, in given context, say Spend Transaction, we check if recipient is actual address or a name. If that’s a name, we look for well known key in
account_pubkey key and corresponding
account_pubkey type hash. You can have multiple pointer keys with values limited to 32 bytes.
To sum up, thanks to AENS, you can use labels/pseudonyms in signed transactions and they are still valid and executing correctly.
However, there is an API to fetch data that is attached to a name.
This opens up opportunity to use AENS outside of blockchain.
Current limitation is that you need to encode arbitrary data as
account/channel/oracle/contract hash types. You can store it under arbirary key. Also it is limited to 32 bytes.
What would be good extension of the AENS protocol to facilitate even more general use-cases?
Our proposal is to allow
byte type, so the encoding is more elegant and we don’t need to pretend it’s something else.
Please share your ideas!