Whitelist/verification spam filters
pereira at cis.upenn.edu
Wed Aug 28 15:29:42 CEST 2002
On 8/27/02 9:28 PM, in article
slrnamo9p8.7q8.gerhard.haering at lilith.my-fqdn.de, "Gerhard Häring"
<gerhard.haering at gmx.de> wrote:
> Fernando Pereira wrote in comp.lang.python:
>> Gerhard wrote:
>>> Hoops? Takes me at most one second.
>> I would guess it's a minute or more.
> You get the TMDA answer that you have to hit reply (or enter some key
> combination in a /real/ MUA ;-) and just do that. The person you sent
> mail to will then be able to see the original mail you sent to
> him/her. I don't see how this will take a minute or two of your time.
Looking at *any* message to decide what do do with it takes much more than
one second. I'll have to think "What is this? Is it spam or some kind of
joke? Oh, I see. Do I want to put up with this silliness? Maybe. OK, let me
reply". One second, no way.
>> Add all those minutes over the legitimate new email senders over a
>> period of time, and you are talking about a significant burden shift
>> from the recipient to the community of legitimate senders.
> There is a burden shift, but I'd not call it significant. If you want
> to blame somebody
I don't want to blame anybody. I didn't. I just pointed out that there is a
burden shift. Its magnitude is an empirical question. But let me quote from
another area, traffic, in which similar problems arise (Wolfgang Sachs in
"For the Love of the Automobile", quoted in a good Malcolm Gladwell article
in the Sep 2 02 issue of the "New Yorker"): "Once a certain traffic density
is surpassed, every driver contributes involuntarily to the slowing of
traffic. The time that the individual driver steals from all the others by
slowing them down is greater many times over than the time he or she might
have hoped to gain by taking the car."
More information about the Python-list