May 2000

Pleased To Meet You, Hope You Guess My Name

BY ROBERT VAHID HASHEMIAN

At home, Caller ID allows us to screen calls and identify crank callers, as well as allowing us to return calls even if callers do not leave a message. And of course, there are the calls to be ignored, banished quickly and easily to the answering machine.

Credit card companies have been battling fraud by asking the card recipients to dial a number from home to activate the card. This not only allows them to identify the residence where the call originates before the credit card is activated, but also to do a little marketing to a captive audience.

Call centers too have reaped great benefits from Caller ID. Customers calling a service agency are quickly identified through Caller ID and routed by a skills-based routing system to the most appropriate group of agents for handling. The receiving agent gets a screen pop with all of the caller's vital stats before the call is even answered. Caller ID has also been of use in making outbound telemarketing more effective. A predictive dialer systematically dials appropriate telephone numbers until a person answers the call. At this point the call is transferred to a live agent and a simultaneous screen pop notifies the agent of the callee's identification for a more personal approach.

Now, Caller ID -- like any technology -- has seen its share of glitches, but by and large, considering its short life, it's been a pretty smooth ride. Why? In a word: Standards. Yes, the telephone companies are slow to move, yes they have red tape that stretches from here to the Milky Way, but let's be fair -- they do a good job of making standards and more importantly adhering to them. That's one reason we've become accustomed to reliable telephone service and that is why Caller ID service works well.

ID-ing WEB CALLERS
With Internet telephony becoming increasingly popular, companies have been busy devising methods to effectively set up a Caller ID system for VoIP calls. But obstacles abound. Let's look at some of issues that make the Internet such a hard place to set up a Caller ID system:

Dubious Identification: No matter how many pieces of information you try to elicit from users, there is still no way to ensure that they provide correct information. People have been conditioned to distrust the Internet, so they go through hoops to give wrong information about themselves.

Cookies: Cookies by themselves are of no use except to determine whether the user has already been on the site. Cookies can be used to identify people based on the information they previously provided about themselves but then that would put us right back to problem 1.

E-mail Address: More and more sites verify their users using their e-mail addresses. They not only check for a correct syntax of the e-mail address, but also use the given e-mail address to send a confirmation message with an account code, password, etc. thereby necessitating users to give their valid e-mail addresses. Okay, so the e-mail address is valid, but who's to say it's not one of those freebies provided by Yahoo, MSN Hotmail, Excite, etc. Many people use these services just to get a valid e-mail address, and they could easily use bogus information to open an account.

IP Address: If ever there were a useless method for identifying someone online, this would qualify. With dial-up IP addresses changing every time a user goes online, and proxy servers identifying whole companies under one IP address, it's easy to see why.

Hackers: With the countless number of hacking tools available online, even bumbling hacker wannabes can easily learn to fake their identities.

CONCLUSION
No matter how you slice it, coming up with an effective Caller ID system for VoIP will have to include some red tape. This should probably come in the form of a central certificate authority, which would involve having users authenticated first (perhaps against some known information such as their phone numbers) before using such a system. This would be not unlike the DNS (Domain Name Servers) used today to bind names to IP addresses.

Yes, it will be expensive, it will not be perfect, there will be privacy issues, there will be a risk of hacking, and consensus will be difficult to achieve, but it may be a worthwhile exercise in the end.

Do you have a better idea? Let's hear it.

Robert Vahid Hashemian provides us with a healthy dose of reality each month in his Reality Check column. Robert currently holds the position of Technical Director for TMCnet.com -- your online resource for CTI, Internet telephony, and call center solutions. He can be reached at rhashemian@tmcnet.com.