[tor-talk] [Bitcoin-development] Tor hidden service support

[grarpamp]

>> 256 bits - A tough interop problem for sure. It seems maybe
>> a worthy long term goal for the projects to sort out.
>
> It can be probably solved on the I2P side. The 256-bit address is a
> SHA-256 of the full 3072-bit address (2048-bit public key + 1024-bit
> signing public key) + a few extra bytes. I guess that I2P could
> support a shorter hash for unique local IPv6 mappings — say, 95 bits.

What is use case of 95 bit choice?

Anyway...
https://en.wikipedia.org/wiki/Cryptographic_hash_function

Short hashes unavailable, so it would be truncation, which made
from a secure hash probably better than junk short hash anyway.

I had also thought I2P could generate keys, but reject putting
into operation those that fall outside of some desired mask
range, say only use first 80 bit of 256 space. About the same idea.

Could be problem that 80 bit is not necessarily strong future days.
But then is problem of IPv6 128 width as private RFC/48+80 size.

Tough choices. Or just accept <=80 strong in human weak link :)