[Cryptography] On the Impending Crypto Monoculture

Ron Garret ron at flownet.com
Mon Mar 28 01:57:42 EDT 2016


On Mar 27, 2016, at 6:51 PM, ianG <iang at iang.org> wrote:

> On 27/03/2016 20:35 pm, Ron Garret wrote:
>> 
>> On Mar 27, 2016, at 9:26 AM, ianG <iang at iang.org> wrote:
>> 
>>> But this set - any set - has a shelf life.
>> 
>> The right way to deal with this is by versioning the entire crypto suite.  So NaCl is Comprehensive Tool Chain version 1.  When some problem is identified with NaCl, the community can work on CTC version 2.  Maybe there could be two versions active at any one time, a standard suite, and a backup that is “in the bullpen” in case something goes wrong with the primary.  A backup for NaCl would probably look just like NaCl but with Curve25519 replaced with Curve448-Goldilocks.
>> 
>> It’s very rare that a weakness is discovered in a core crypto algorithm without a lot (like several years) of warning so a single backup should cover even the most catastrophic of circumstances.
> 
> That.  Is exactly what we need.

Thanks!

I’m working on designing some minimalist protocols based on TweetNaCl.  I could use help.

rg



More information about the cryptography mailing list