[Cryptography] DIME // Pending Questions // Seeking Your Input

Christian Huitema huitema at huitema.net
Mon Mar 2 16:37:11 EST 2015


> p.34 "PART 5: SIGNET DATA FORMAT
> 
> alma mater field
> gender field"
> 
> dick size field? Too damn many fields.

Yeah, that's weird. When I read that, I am reminded of the origins of X.509,
and the use of "natural names" in X.500. The idea at the time was to avoid
creation of extraneous identifiers, and just pile on enough natural
attributes to achieve uniqueness. If surname and first name do not suffice,
add a city. If that does not suffice, add a zip code... 

There is beauty in keeping things simple. For a user signet, why do we care
about more than the email address and the key? OK, I can see why we want to
add some management fields, such as some form of expiry date and creation
date. And maybe facilitate search. But that does not mean that the
searchable attributes shall be structured. A "common name" type of field
ought to be enough.

-- Christian Huitema

 



More information about the cryptography mailing list