libtool
[Top][All Lists]
Advanced

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

Re: [libtool lists]: do not send held-for-approval messages


From: Ralf Wildenhues
Subject: Re: [libtool lists]: do not send held-for-approval messages
Date: Thu, 27 Apr 2006 08:58:59 +0200
User-agent: Mutt/1.5.11

Hi Bruce,

* Bruce Korb wrote on Wed, Apr 26, 2006 at 10:53:44PM CEST:
> On 4/26/06, Ralf Wildenhues <address@hidden> wrote:
> >
> > Currently the libtool mailing lists employ first-post moderation for
> > nonmembers.  When such a moderation request occurs, the list software
> > sends a message to the poster that her message is being held back for
> > moderation.

> > I have disabled the held-for-approval messages for now.  Any objections
> > against this step?  (For list moderators, the setting is found here:
> > [General options] [respond_to_post_requests]).

> Can the "backscatter" be suppressed only for forged headers?
> i.e. email coming from an IP address different from that of the
> apparent sender?

To tell you the truth, I don't know.  The online help describing the
option is as follows:

| respond_to_post_requests (general): Send mail to poster when their
| posting is held for approval?
| 
| Approval notices are sent when mail triggers certain of the limits
| except routine list moderation and spam filters, for which notices
| are not sent. This option overrides ever sending the notice.

So it seems if we would beef up the spam rules, that could help generate
less messages to victims ending up in spam From:'s.  But I have no idea
whether mailman can even find out whether the header is forged.  I'd
guess this would be very difficult to find out reliably anyway.
(I, for example, frequently send with a different IP.)

I haven't updated the spam rules in a while; it's been more
time-efficient for me to just drop all spam messages, and we've been
burnt once or twice by wrong regexes in the mailman spam filtering.
So not only are those spam regexes error-prone, they are also hardly
useful, with spam changing all the time.

> Otherwise, silence for the neophytes is better than
> suppression for the larger audiences.  This is awful.  :(  - Bruce

Well, if the moderation can be done in a timely manner, I don't think
there is a big problem: the gnu.org list archives take up to a full day
anyway to show your message (if you just posted after one of the
12-hourly updates).  Only when moderation isn't done for a while, there
may be an issue.

Things may be different for subscribed posters that may be moderated
because of a message that exceeds the size limit.  They may try to send
off several messages.  Oh well.  If the moderator sees it, it's still
possible to avoid multiple messages to end up being delivered, but I
guess this could be annoying for users.

I don't know, and wouldn't mind to be convinced to some better solution.

Cheers,
Ralf




reply via email to

[Prev in Thread] Current Thread [Next in Thread]