Hi Vincent,

> Date: 2026-08-25 01:15:32+0200
> From: Vincent Lefevre <[email protected]>
>
> On 2026-08-24 10:26:30 +0100, Ian Collier via Mutt-dev wrote:
> > On Sun, Aug 23, 2026 at 10:19:11PM -0400, Kurt Hackenberg wrote:
> > > I think Mutt should not do that, because a forwarded message is not a 
> > > reply.
> > > Those two message types are used for different purposes. Conflating them
> > > would cause confusion and frustration.
> > 
> > I agree that a forwarded message is not a reply and should not be
> > recorded in the In-Reply-To header.  However, I think that if you
> > forward a message then you are explicitly referring to it, and so there
> > is no clear reason why the References header should not be used.  On the
> > other hand, this doesn't benefit the recipient, because they don't have
> > the original message (or you wouldn't have needed to forward it).
> 
> If the forwarded message has
> 
>   References: <A> <B> <C>
>   Message-ID: <D>
> 
> it is debatable whether to put <D> in the "References:" of the new
> message. But it really makes sense to still have <A> <B> <C> in the
> "References:" of the new message.

Hmmm, agree.

And as a consequence, I think I'm tending more towards the approach of
thunderbird(1), which is to behave exactly as if it were a reply.
Otherwise, it'd be hard to know exactly where to thread the message.

I think it might be good to have:

        References: <A> <B> <C> <D>
        In-Reply-To: <D>

But at the very least it should have:

        References: <A> <B> <C> <D>
        In-Reply-To:


Have a lovely day!
Alex

-- 
<https://www.alejandro-colomar.es>

Attachment: signature.asc
Description: PGP signature

Reply via email to