__/ [ Peter Köhlmann ] on Tuesday 28 March 2006 21:05 \__
> Roy Schestowitz wrote:
>> __/ [ Peter Köhlmann ] on Tuesday 28 March 2006 20:10 \__
>>> Roy Schestowitz wrote:
>>> < snip >
>>>> I currently have Boxtrapper enabled for 5 mail accounts and I still
>>>> check the moderation queues at the end of each month, which makes it
>>>> seem like a rather benign solution. The big pitfall is people who refuse
>>>> to verify using BoxTrapper or do not comprehend the challenge. They get
>>>> a reply up a a month overdue. At least they have the cause/evidence in
>>>> their box.
>>> I have made it a policy to *always* respond to challenges *if* they come
>>> with my mail address, but are in reality spam.
>>> That way I make sure that the user with the C/R system gets at least the
>>> spam which seems to be coming from me (and naturally isn't)
>>> On the other hand, I *never* respond to challenges for genuine mail
>>> Either the user with C/R can correspond normally with me, or he isn't
>>> worth corresponding with
>> As I said, I only have C/R enabled for accounts that should not
>> (conventionally) be addressed. The UseNet E-mail is one such account as it
>> attracts nothing but around 20-30 spam per day. C/R delivers real messages
>> very, very rarely, maybe once per 400 messages. As I said, I am somewhat
>> apologetic and ashamed, but as Brad Templeton said, why should one be
>> penalised for attempting to end spam? The mechanisms of E-mails were
>> poorly built and there are no ideal solutions.
> The problem with C/R is that innocent users are penalized for the spam.
> They get to respond to challenges for mail they never sent in the first
> place, making the spam-problem even worse for them, because they get
> challenges in addition to the normal spam.
> That is the very reason I try to make sure that at least the spam seemingly
> coming from my address will get to those C/R users until they notice that
> somethings wrong with their white-list.
> And they have to put my address into their black-list so that I don't get
> further challenges. That way they have at least some work to do by hand
> The only way to deal with spam is making it a severely punished crime, and
> to let those assholes pay hefty fines who do their advertising by spam
> C/R systems are by far the worst way to deal with it
As I said, I feel very apologetic about this and I still seek to reduce the
number of challenges delivered, if not drop that filtering system
altogether. For all BoxTrappers I have put _SpamAssassin in front_ (or "on
top", depending how you view this... vertically or horizontally), which
means a challenege will be delivered _only_ if the message 'bubbled' past
SpamAssassin. I would say this reduces the volume of challenges by about
Roy S. Schestowitz | Linux: most popular O/S, yet not most widespread
http://Schestowitz.com | SuSE Linux ¦ PGP-Key: 0x74572E8E
7:10am up 20 days 20:55, 9 users, load average: 0.87, 0.75, 0.76
http://iuron.com - next generation of search paradigms