Crazy Stats on Spam

gtaylor writes "An article in the Korea Times says that market research firm Emforce has established that South Korean internet users average about 1600 pieces of spam annually, summing to around 39 billion pieces of spam per year. According to the same story, Americans receive about 2500 pieces of spam per year." I figured that I get somewhere around 30-40,000 pieces of spam annually. Lucky me... I get *this* statistic to be on the other side of the bell curve :)

Comments

Re:It must work on someone.

[rgmoore]

Not necessarily. Arguing that spam must work because people keep trying it is like arguing that "Make money fast" must work because people keep sending it (or variants on it). All that's required for people to keep spamming it is that they think that it works, not that it actually does work. My guess is that the only spam that actually gets a big enough response rate to justify sending it is the kind that advertizes spamming services. Unfortunately, we'll only know for sure in 10 or 20 years when everyone who's unscrupulous enough to try spamming has done so. If they all give up because it doesn't work, we'll know that it was a failure and people were just trying it because they didn't know any better. If it keeps up indefinitely, we'll know that it does work and we'll have to start revoking net access of anyone dumb enough to reply.

The Missing Stat - SNR!

[nyquist_theorem]

Unsolicited mass emails are never going to go away 100%. It frustrates me that so much time and energy and print/webspace is given to studies and articles that don't include what I would think to be the most important indicator of spam's level of infiltration - Signal to Noise Ratio. Sure, the "average" user gets xxx Spam per day/year/minute, but on what amount of traffic? If the "average" user gets 1600 spam out of 1700 emails, that's obviously very bad, but 1600 on 170,000 emails a year is a lot better. The poster's comment about being on the wrong side of the bell curve doesn't neccesarily mean he's getting more spam than most people as a ratio of spam-to-legit-emails. I would be most interested in studies that analyze the SNR, for in doing so I think we'll see (even more clearly!!) that there is indeed a spam problem that must be dealt with through enforceable legislation and/or international agreements.

As a side note, I have taken to giving out different email addresses for every place I'm asked for one, and using a "catch-all" from my domain, for example my email address here is slashdot@theoretica.net, but it might be goatpornmailinglist@theoretica.net or vic20overclockerslist@theoretica.net for other places. That way not only can I see what spammers got my email address from where, but I can also block a given address once its been overcome with spam - you know those places where you are asked for an email address and you just *know* you are going to get spammed senseless for providing it, but you must to get a login or pwd or whatever?

I also have OE move everything that's been BCC'd to me into a spam folder, mark it as read, and review it once a week.