Re: [Dcrup] Hashed Key Records

[Scott Kitterman <[e-mail]>]

On Thursday, June 22, 2017 08:01:28 PM Phillip Hallam-Baker wrote:
> On Thu, Jun 22, 2017 at 7:51 PM, Steve Atkins <[e-mail]> wrote:
> > (and maybe be specific about what embedded whitespace means)
> > 
> > These all seem good features..
> > 
> > I'm reminded of the $1$... style password hashes. Immediately recognizable
> > in
> > general, but including enough embedded information that you know which
> > spec to follow.
> 
> ​The other thing I would like is the ability to truncate fingerprints to
> arbitrary lengths by truncating the string specifying the fingerprint. In
> all but a few applications, a fingerprint with a workfactor of 2^118 is
> sufficient to prevent most attacks.​ That is 25 characters in my proposal
> (since there is a one byte algorithm identifier prefix).
> 
> There are also some hacks that can be used for compressing the fingerprint
> but I don't think these would be relevant here.

I'm probably just grumpy because it's been a long day in a long week, but if 
you are going to not answer my question and go off in a completely different 
direction (in case you missed it, my question was about what's in the draft 
now, not an invitation to imagine what should be in the draft), would you 
please at least start your own thread so others are less likely to assume I 
already got my answer somewhere in what seems likely to be a long thread 
spiraling off into some new ocean that wants boiling?

Thanks,

Scott K