Quoting from Curtis Frye's response: What exactly is Netcom's liability in this case? If a service provider is treated as a phone company, abuse of the service could be grounds for discontinuing the service. As Netcom did not discontinue tmp's service (see below), perhaps there is room for suit there. By your phrase "treated as a phone company", are you implying Common Carrier status? If so, I wonder under what grounds a common carrier is allowed to terminate service? Certainly, non-payment of bills is one. If person A alleges person B has made threatening phone calls, can the phone company terminate person B's service unilaterally based solely on the allegations? Must it wait until civil suits or criminal charges are brought? My understanding of common carriage was that the carrier was obligated to provide service to everyone within its service area, whether or not it liked their politics, or bedmates, or the content of their communications. What protection should Netcom, and other access providers, have against the abuses of their subscribers? Damn if I know, but it's an important question to stay awake at night worrying about... What was in the contract signed by the subscribers? You can bet that if I ever start an on-line service, there will be an "indemnify and hold harmless" clause, and provisions for demanding proof of umbrella liability coverage; lack of coverage upon demand would be grounds for termination. In other words, if tmp@netcom were a subscriber to my service, and someone alleged wrongdoing to this degree, my action would be to demand of tmp proof of umbrella liability coverage of $1,000,000; without such proof, account revoked. The only way to put teeth in an indemnify-and-hold-harmless clause for an individual, I guess. Very much in line with your earlier thoughts on the potential (non-) application of traditional libel/slander law to the Net -- would the Courts allow cyberspace to be that much more rough-and-tumble, with higher standards of abuse required for successful suits? They'll have to, with the current state of the art with respect to authentication and non-repudiation. A plaintiff will have to work pretty hard to prove a particular human generated a piece of email, given the ease of spoofing, remailers, anonymity and pseudonymity. I also thank Uni for posting. Jason Zions