PGP/MIME [was Re: Balsa Encrypts messages with GnuPG!]

[comments below]

On Wed, 3 Jan 2001, Oliver Oberdorf wrote:
> Date: Wed, 3 Jan 2001 20:29:29 +0900
> To:
> From: Oliver Oberdorf <>
> Reply-To:
> Sender:
> Subject: Re: Balsa Encrypts messages with GnuPG!
> With all respect,
> I don't think balsa will be free of mutt for some time.  People have
> wanted
> PGP support in balsa for *ages*.  Others have repeatedly shot down the
> idea of using lubmutt's PGP with this argument, but /nobody/ has time to
> do
> it this way and libmutt seems to still very much a part of balsa.  Now
> that
> someone is 90% done with the obvious approach, I see no reason to stop 
> him.

He's still a long way off, he only has signatures working (and encryption?)
and only for GnuPG. (btw Werner Koch is working on a library called GpgMe
which will replace his efforts for GnuPG support).

> Plus, mutt is time-tested and I see no reason to think a from scratch
> implementation will work better.  More flexible for developers; maybe.
> Better integrated with gnome PGP; obviously.  Ready in 3 months or
> less; not likely.

It can be ready easily in 3 months as I've already written a full MIME
implementation (*with* PGP/MIME support even). All that needs to be done is
to integrate it into Balsa.

IMHO, PGP/MIME is the correct PGP implementation to use - inline pgp hacks
are a waste of time...sure it'd be nice to support receiving them but not
sending them. Receiving them using my library would be a sinch as well

multipart/encrypted and multipart/signed are the accepted way of doing this
and it works a lot better too.

The other concern is that there is no standard way of encrypting
attachments inline (inline in this context means non-PGP/MIME). PGP/MIME is
the only acceptable way of encrypting/signing attachments.

In the case of signing parts, why sign each attachment? that's a waste -
it's better to sign the entire email message. And *that* can only be done
with PGP/MIME.

Anyways, take a look at my API before you toss this idea aside, I believe
that once you see the API you'll be convinced it is THE WAY.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]