[ietf-dkim] Broken signatures, was Why mailing lists should strip them

[Alessandro Vesely]

On 30/Apr/10 08:50, Murray S. Kucherawy wrote:
>>  bounces at mipassoc.org] On Behalf Of Alessandro Vesely Sent: Thursday, April 29, 2010 10:55 PM
>>
>>  Yet, it would seem that by, say, hashing just invariants of binary representations of the first entity, e.g. discarding its white space and punctuation, one may reach very high percentages of unbroken retransmission.
>
> This sounds like what DomainKeys (RFC4870) called "nofws" canonicalization, which was discarded in favour of what is now "relaxed" in DKIM.

Not exactly, removing punctuation would also take lines beginning with 
">from". For the body, we could peek any suitable baseline 
tokenization and hash its results.

> I don't specifically recall the reasons now but I'm sure they're in the archives if someone else cares to dig that far back.

The reason is meticulous security, which makes mailing lists' contents 
sleazy and illegitimate.

One is http://mipassoc.org/pipermail/ietf-dkim/2005q3/000002.html
(the previous part of the discussion is in some other archive or lost, 
but much text can be read in the quoted part of the message.) It 
exemplifies

   "Amoeba yeast" to Amo ebay east

Another good summary of the driving thoughts is given in
http://mipassoc.org/pipermail/ietf-dkim/2006q3/004416.html
(while discussing whether to keep body- "relaxed".) It exemplifies

   --boundary
   Content-Type: image/jpeg
   Content-Transfer-Encoding: base64

to

   --boundary
   Content-Type: image/jpegContent-Transfer-Encoding: base64