NBC Dateline show on Spammers:
http://www.msnbc.msn.com/id/17713446/
What is Spam?
What is phishing?
How serious is the problem? (Consider the Moore's law.)
What would be the consequences if we don't stop Spam?
What is the economic model behind Spam? (If it costs $200 to send 1 million messages, how many responses does the spammer need to break even?)
What is the technological principle behind Spam? How do they get our email addresses?
As a user, how should we respond to Spam?
What solutions exist?
1. Technologically: Filtering software at individual, corporate network, and ISP levels. Would this work? What are the problems associated with this solution?
Consider the positive correlation between filteration rate vs. false-positive rate
Example: Brightmail's database approach (70% vs. .0001%); Spam Assassin and Spam Killer (95% vs. 1%); SpamNet (user voting); Bayesian filter, which learns over time (99.8% vs. .05%).
Black lists vs. white lists
2. Economically: Changing the way email works, by imposing postage to email messages (real cost or computational cost) and making email traceable.
Would this work? What are the problems associated with such solutions?
3. Legally: U.S. CAN-SPAM law (effective Jan. 2004). Would it work? Why or why not? (details of the law on wiki)
Questions to think about:
1. Today's email system was designed 20 years ago for a small number of people who know each other. Unintended consequences of technology!
2. 90% of Spam is sent by fewer than 200 people. Extremely high external cost.
3. Can you solve a social problem by technology alone, law, or what else?
Tuesday, April 7, 2009
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment