On 14 Jun 2004, Bernd Eckenfels wrote:
> In article <[EMAIL PROTECTED]> you wrote:
>> This sort of thing is why I would rather use any RBL within
>> SpamAssassin, rather than at SMTP delivery time. Even if one of these
>> services goes completely belly up and blacklists the world, I don't
>> automatically lose mail from it.
>
> Please dont do this.
Eh? You seem to have made an incorrect assumption about what I "do" to
the mail with SpamAssassin.
> You MUST reject mails (by spam scanners, malware scanners or
> blacklists) on the SMTP level, otherwise you become a pretty big
> annoyance to the internet (if you bounce) or will siletnly lose mails
> (if you drop them).
...or, options 3, I deliver them to the end user tagged as "likely spam"
when they look like spam. Then the end user can filter them out as they
please.
I certainly agree that bouncing SPAM messages, just like "reporting"
virus infections, is an anti-social behaviour.
If I chose to silently drop mail after "accepting" it, though, that is a
legitimate and reasonable disposition of the content, as far as I can
see.
Claims that this is anti-social seem spurious to me; can you expand on
your reasoning there?
Anyway, as I said, I don't take either of the options you suggests.
I use RBL tests at the SpamAssassin level because I *don't* trust them
to be one hundred percent accurate.
If I didn't care more about real mail getting through than the
occasional missed spam, then sure, using RBL blocking at the initial
SMTP stage would be ideal...
Daniel
--
... Far down the vault a man was screaming. His fists were tightly clenched
and he was screaming out imprecations against the humming computers. There
was a hopeless rage in his eyes - rage and bitter, savage defiance.
-- Frank Bellknap, _It Was The Day Of The Robot_ (1963)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
