NULL is the culmination of five years of work by a team of language designers and computer science mathematicians. The final breakthrough occurred when operating system expert Hugh Nicks suggested that if removing GOTOs was good then why not scrap IF statements as well, since they usually required typing too many characters anyway. This brilliant concept was extended through a series of complex mathematical theorems that form the basis of the NULL language.

Put in layman's terms by Sally Kahn-Vallee, electrical engineer and PROM reader, "Like first we tossed out the bath water, then the baby, and like, finally, the whole tub." The elegance and conciseness of NULL can thus be proven to be a direct consequence of the fact that the language as defined contains no statements at all. While at first glance this may seem a drawback, in fact, it is a major improvement over any other language. A few of the numerous reasons are:

• Highly structured constructs.
• Advanced data hiding techniques.
• A NULL compiler can be written first in NULL without ever needing to be written in a lower level language.
• Since there are no statements to compile, in fact, no compiler need ever be written in the first place, saving time and money.
• Since there will be no compilers, no new releases will ever be issued hence maintenance is reduced.
• NULL programs are highly portable and totally machine independent.
• NULL programs compile and execute rapidly. An important point to note is that with the addition of a small amount of language dependent code, e.g. PROC/END etc., all NULL programs can be compiled by any other language compiler.
• Since there will never be new releases of NULL, all programs are upwardly and downwardly compatible.
• NULL can be parsed top-down, bottom-up, left-right, right-left, inside-out, and over-easy.
• NULL programs are both self-documenting for clarity and self-concealing for security.
• NULL programmers are easy to find and once found can be fired since they are not needed.
• If desired, specialized NULL hardware could be designed implementing the code in firmware. Of course, such hardware may require years of development. One suggestion from Bell's VLSI experts Nora and Andy Gates was to take an existing available chip and remove all the instructions except NOP. While this should work in theory, they acknowledged that it is probably not the most efficient implementation.

These are just a few of the many ways NULL is superior to all current computer languages. You can, no doubt, think of more. For further reading consult any of the numerous books and articles by Donald Knuth, David Parnas, and of course, the basis of all modern computer language theory, "The Emperor's New Clothes."

By John R. Andrews, University of Illinois at Chicago.