search

multiformats / multicodec

Compact self-describing codecs. Save space by using predefined multicodec tables.
MIT License
336 stars 201 forks source link

Add EIP3770 chain-specific Ethereum address in table.csv #302

Closed qizhou closed 1 year ago

rvagg commented 1 year ago

How would this end up getting used in practice?

Let me pull in a few folks who I suspect may have some overlap between here and CAIP to figure out if this is the right approach; tbh I don't quite have the background to make much of a judgement on this. @oed, @ukstv, @clehner, @bumblefudge.

qizhou commented 1 year ago

How would this end up getting used in practice?

Let me pull in a few folks who I suspect may have some overlap between here and CAIP to figure out if this is the right approach; tbh I don't quite have the background to make much of a judgement on this. @oed, @ukstv, @clehner, @bumblefudge.

We plan to use it in web3:// access protocol (https://eips.ethereum.org/EIPS/eip-4804), where a name from ENS can be mapped to a chain-specific address of a contract that hosts a decentralized website.

ukstv commented 1 year ago

@rvagg Thanks for mentioning!

@qizhou Why would you need multi codec to use with EIP-4804?

ukstv commented 1 year ago

Maybe CAIP-50 could work for your purposes?

qizhou commented 1 year ago

Maybe CAIP-50 could work for your purposes?

Thanks! Let me take a look.

bumblefudge commented 1 year ago

TBH I'd prefer a binary namespace for the general pattern of VM:chainID:address, whether it's CAIP-50 or something else on the same pattern, rather than a separate one for each VM of the pattern chainID:address. in fact, half of CASA might throw tomatos at me for suggesting it, but even VM:chainID:publickey might be the more multiformats appropriate low-level expression, although then you get into the feature and/or bug of needing a signature from/connection to the wallet to derive said in EVM systems...

In any case, I think CAIP-50 has been languishing around looking for EXACTLY THIS use-case and a champion building on it, so @qizhou if you'd like to discuss this in a public CASA meeting let's get you calendared and invite the right interlocutors...

oed commented 1 year ago

I would love to see CAIP-50 pushed forward!

qizhou commented 1 year ago

TBH I'd prefer a binary namespace for the general pattern of VM:chainID:address, whether it's CAIP-50 or something else on the same pattern, rather than a separate one for each VM of the pattern chainID:address. in fact, half of CASA might throw tomatos at me for suggesting it, but even VM:chainID:publickey might be the more multiformats appropriate low-level expression, although then you get into the feature and/or bug of needing a signature from/connection to the wallet to derive said in EVM systems...

In any case, I think CAIP-50 has been languishing around looking for EXACTLY THIS use-case and a champion building on it, so @qizhou if you'd like to discuss this in a public CASA meeting let's get you calendared and invite the right interlocutors...

Sure. Could you please send the invitation to me? My email is qizhou@ethstorage.io.

BigLep commented 1 year ago

2022-11-22 IPLD triage: check in on this in a future week or two to see how conversations progressed between interested parties. We may not need to make this specific change if it fits in CAIP-50.

rvagg commented 1 year ago

@qizhou are we OK to close this out or have we not reached a satisfactory position with regard to the other CAIP work?