On Thu, 23 Mar 2006, Mark Delany wrote:
On Tue, 21 Mar 2006, william(at)elan.net allegedly wrote:
Its nice to see you finally come around and adapting further META
signing structure
...
of course you will probably never admit where this is all coming
from ...
Of course we will. The idea is merely a variant of Content-MD5.
And domainkeys is variant of PEM with difference that PEM specified use of
several PKIs, one of them based on specialized binary record for retrieving
public key from DNS, where as domainkeys came up with reusing TXT record for
it and based it entirely on DNS. As Tony mentioned - ideas don't die, I
guess they just get patented and only then reused in another protocol :)
(BTW - I'm not saying you knew about PEM when you came up with domainkeys,
reinvention of same idea 10 years after is not unusual)
What Meta-whatever-it-is has to do with that admission I do not
know. Are you suggesting that Content-MD5 derived from Meta-whatever?
One of the META's primarily features is separation of header and body
canonicalization and hash creation for email signature - so the signature
then is truly header-only that incorporates one or more body hash digests
with that hash data available as separate item in email header.
As far as I know what is at metasignatures.org is the first published design
that used this concept, although as Phil said and I believe him it was also
used by others in their private non-published efforts.
As far as Content-MD5 it is well mentioned in content-digest draft as direct
origin for Content-Digest header field with Content-Digest being done partly
because Content-MD5 needs to be replaced by something more advanced and not
tied to specific hash algorithm in field name.
--
William Leibzon
Elan Networks
[EMAIL PROTECTED]
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html
