Re: Spam in the lists out of control
On Tue, May 11, 2004 at 08:53:34PM +0100, Henning Makholm wrote:
> Scripsit Neil McGovern <email@example.com>
> > however, the question is if we should make the list postable to only
> > those who are subscribed, or those on a whitelist. As it stands at
> > the moment, the list policy, whilst very admirable for it's open
> > access mentality, is prone to spam.
> However, nobody have proposed any solution that does not entail a
> greater problem than the one it is supposed to fix.
> Rejecting posts with a From address that is not also a subscriber
> address is Not An Option:
> 1) It will prevent me from posting to the lists unless I change my
> From address to the one that I use for receiving mail from that
> list, which
Erm, not with the use of a whitelist.
I'm subscribed to receive the lists to firstname.lastname@example.org
I'll post to the lists using email@example.com - which I'll put in
> 2) It will, by definition, make it impossible to participate in a
> single discussion on a list without having *all* the unrelated
> discussions and flames on the list sent to one's personal mailbox.
> This possibility is valuable not only for the individual "from
> outside", but also for the list itself. There are many situations
> where it can be extremely useful to involve an upstream author in
> a single thread on a list, but where it would be discourtious to
> require upstream to be buried in the entire list traffic.
See above. Posting address doesn't have to be the same as receiving
> 3) Many Debian lists are not only discussion fora, but *contact
> points* for specific functions in the project. The ease with with
> people can contact these lists will be greatly reduced if we have
> to say
> If you want advice with DFSG-freedom of a license, please
> subscibe to firstname.lastname@example.org and post your question there.
> instead of just "post to d-l and ask for a summary after the
This could be a problem, indeed, but they woudn't need to subscribe to
the actual list, just the whitelist.
> Or how about:
> If you want to report a bug against any orphaned package, please
> first subscribe to debian-qa-packages and then post the bug to
> the BTS.
> If you want to request a rebuild on your package on a particular
> architecture, please first subscribe to email@example.com, and
> then ...
I can submit bugs without being subscribed to debian-qa-packages.
Changing list policy won't, and I agree _must not_ affect the ability
for people to submit bugs.
> 4) It will be impossible to read list mail through a NNTP gateway
> shared with others and still post directly by email.
Impossible, no. Harder, yes.
They would also subscribe to the whitelist.
> 5) The number of morons who manage to subscribe to a list only do be
> hit by severe amnesia and become unable to follow the clear and
> easy unsubscription instructions at the bottom of every list mail
> will increase dramatically, because more morons will be forced to
> subscribe in the first place. The number of "plaese unsucbirbe me"
> emails may become comparable with the number of spams a
> subscribers-only policy will keep at bay.
No, people won't have to subscribe to the list. Simply the whitelist.
Apologies if this sounds like I'm annoyed, it's not intened to be, and
I've had a bad morning :)
A. Because it breaks the logical sequence of discussion
Q. Why is top posting bad?
gpg key - http://www.halon.org.uk/pubkey.txt ; the.earth.li B345BDD3