The Wrong Key

In your discussion of encrypting and decrypting messages (see the sidebar "Security: Who's Got the Key?" in the February cover story), author Michael Nadeau states that "the private key encodes the message, and the public key decodes it." This is correct for digital signatures but not for secure e-mail. In secure e-mail, you use the recipient's public key to encode the message; the recipient uses his private key to decrypt it. The problem is not "how to make those keys available to only the people you want using them." The public key should be accessible to everyone! To read the message, the recipient uses something that only the recipient has: the private key. At no time is there a need for secure channels to transmit information to anyone in the transaction.

Note that an e-mail message sent to many users cannot be bulk-distributed as in a nonencrypted scheme. The messa ge must be individually encrypted using each recipient's public key. This does increase traffic, but there is no known way around it that doesn't require a secure channel for key dissemination, short of creating public/private key pairs for eve ry conceivable combination of recipients needed. This still requires that private keys be distributed to several people, opening up the possibility of compromised communications.

Thomas Paul Karrmann
tkarrman@giddings.com

You are correct. I confused the terminology. The point I was trying to make was that most companies don't want the public keys to be truly public. To make an analogy between public keys and phone numbers, most individuals don't mind that their phone numbers are published, but few companies make their internal phone lists public. They don't want unsolicited traffic on the network. Your point about bulk mailing and encryption is an important one. While it may be pos sible to create "group-level" sets of keys, it is often too cumbersome to be practical. -- Michael Nadeau