Re: [DNSOP] [Ext] New Version Notification for draft-fanf-dnsop-sha-ll-not-00.txt (fwd)

[John Levine <[e-mail]>]

In article <[e-mail]> you write:
>If we can determine when something in the realm of "almost all" DNSEC signing with algorithms that use SHA-1 is
>done, then it is reasonable for the WG to propose that software that validates DNSSEC can stop doing so.

FWIW, the 2007 DKIM spec said that RSA keys SHOULD be at least 1024
bits but allowed 512 bits as what we intended as a short transition.
In fact, vast amounts of mail continued to have 512 bit signatures and
ignored all the pleas and warnings until 2012 when Google told the
world that they'd stop validating 512 bit signatures.  At that point
in about a week, everyone fixed their signers to use 1024.

The people who run DNS servers and the ones who run mail servers are often
not the same, but I don't see any reason to think DNS operators are any
less lazy.

What this tells me is that the IETF cannot make credible threats of
this kind, so don't try.  People will stop signing with SHA-1 when
large DNSSEC consumers stop accepting it.  Comcast, perhaps.
-- 
Regards,
John Levine, [e-mail], Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly