On 4 May 2002, Robert E. Seastrom wrote:
Scott Granados <scott@graphidelix.net> writes:
No I think your message illustrates things pretty well. I guess the fundimental differenc here is not only does it cost usually very little to receive these messages it costs even less infact dramatically to send spam. It seems there is no real reason for the spammer to be concerned with whether the mail is properly targeted or not so a full on flood is possible and the leads generated by this flood percentage wise have to be many factors less than the percentage of success in snailmail.
It does not cost "very little" to recieve spam. At my real job (ie, not seastrom.com), we're running a very nice (but expensive) commercial product to filter this stuff, and in a given time quantum during which we processed 1.9 million messages, spam and virii accounted for about 600k (32% was the last number I saw from our stats script). It's reasonable to assume, since some unwanted messages slip through, that we're over a third of all email being UCE.
<trollishly> I'd like the costs quantified.. Servers and disks are expensive, but if they handle a ten million messages during their lifetime, the amortized cost PER MESSAGE is cheap. How cheap is it? I bullshitted about $.00022/message with Emails's are 10kb. $1/gig (bandwidth) and $10/gig (disk capacity, falsly assuming email is never deleted.) $0 (for the server, cause I can't guess within an order of magnitude.) I bullshitted about $.00022/spam with Spams's are 10kb. $1/gig (bandwidth) and $10/gig (disk capacity, falsly assuming spam is never deleted.) $0 (for the server, cause I can't guess within an order of magnitude.) What do you guess for the amortized cost/spam? ``A modern email infrastructure costing $XXX/day (amortized over 2 years) can handle YYY messages, thus the average cost/message is $XXX/YYY.'' </trollishly> I've not seen quantified numbers bandied about in the past NANOG spam-flamewars, so maybe this isn't beating a dead horse. I do find it amusing that nobody responded to my more relevant and intended thrust, about how putting a 'sender pays receiver for email' could cause a variety of new abuses of the email system. Scott