Improving security in computational electronic mail systems

Computational electronic mail is essentially traditional electronic mail messages with embedded programs or interpretable scripts. At the recipient’s end, the mail reading software will distinguish it from normal mail, and executes it upon confirmation by the recipient. If appropriate, output from the program can be channelled back to the sender automatically for further processing. Despite its potential for many applications, it has however, not taken off in a big way. A major reason stems from security concerns amongst users. This paper examines security issues associated with computational electronic mail systems and suggest some techniques and approaches for improving the security of such systems.