Re: a few more thoughts to chew on

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


Eric S. Johansson wrote:
> as I was moving boxes and furniture out of my aunt's apartment this
>  weekend, it occurred to me that if anyone felt like experimenting,
>  twopenny blue has all of the components you need to build a
> client-side stamper that can use either na=EFve or Send once stamping
>  protocols.  The current downfall will be the inability to
> communicate message delivery failures to the user.  It's not say it
>  can't be done, but we will need to do things a little differently.
>
>
> If I had infinite spare time and money, I would start with
> postfix_stamp_feed as the SMTP level interface to the MUA.  If you
> follow the code, it would put messages immediately onto the queue
> (file_queue).  Then the daemon postfix_stamper generates the stamp
> and sends a message to the designated upstream SMTP receiver.  This
>  rough system has the properties of MUA independence, platform
> independence, resilience across power failures and hiding spam
> generation delays from the user.  Problems include manual
> configuration issues and that messages that don't need stamps must
> wait behind any messages in the queue of meeting stamps.  I see
> that as a future feature.  :-)
I already wrote a postfix content filter that does hashcash stamping
on all outgoing emails, if that is what you are aiming for.


Chris

>
> the prototype suggested is not perfect and may not even be
> practical but it puts you a step closer to client-side stamp
> generation that is platform and MUA independent.
>
> Have at it.  As soon as I figure out how to run a bzr public
> repository, I will even support it here.
>
> ---eric
>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFom+iJQIKXnJyDxURAuQRAJ9jAKdd8KEtzRE89LL2O8zpu0mD/gCfTFY0
BDFFvPTi9YaYaCsMoLQwNRc=3D
=3D/q0y
-----END PGP SIGNATURE-----