Re: [kernel-hardening] [RFCv2 PATCH 00/18] refcount_t API + usage

[Solar Designer <solar@openwall.com>]

I dropped all CC's since this sub-thread is only relevant to those
receiving this via the kernel-hardening mailing list.

On Fri, Jan 20, 2017 at 09:20:08AM +0000, Reshetova, Elena wrote:
> Really wonder how it happened that Peter was dropped out of that thread and all other mails were correct.
> Anyway, here was the mail:
> 
> http://www.openwall.com/lists/kernel-hardening/2017/01/05/19

I'm afraid the kernel-hardening mailing list setup may be at fault.
On this list, as well as on most others at Openwall, we introduce the
Reply-To header pointing to the list.  This is meant to keep discussions
on the list, avoiding inadvertent off-list responses.  However, this has
a well-known unfortunate side-effect: many MUAs will then only use this
Reply-To address and not the previous message's From address when doing
a reply-to-all.  This will keep everyone from the previous message's CC,
but the From person may be missing from the reply.  In Mutt, after
pressing "g" for group-reply, to keep the From person on the reply one
has to answer "n" to the "Reply to kernel-hardening@lists.openwall.com?
([yes]/no)" question, in which case the reply is also sent to the list,
but via a CC.  This is counter-intuitive.

With this list's practice (inherited from LKML) to CC lots of people and
other relevant lists on almost every posting, and with many people
actively participating in discussions not being subscriber to this very
list, perhaps we should reconsider and drop the forced Reply-To header.

In fact, I've just made this change.  Please let me know if you think I
shouldn't have - that is, that we should reintroduce the forced Reply-To.

Alexander