Much of this book has focused on the essential need for people to understand more about science and technology and how these fields relate to their lives. One factor that contributes greatly to lack of scientific understanding, and which complements it well, is widespread unfamiliarity with math, or "innumeracy" (to use John Allen Paulos's term). The inability of so many people in our society to handle even the most basic quantitative information is not merely intellectually unsatisfying; it also contributes to gross inefficiencies in the way we work.

I had the misfortune to experience this firsthand when, in an incident reminiscent of the cartoon "Dilbert," I learned that I would be required to use a 26-digit account code in order to purchase items for the graduate group in which I work. I decided that publicly embarrassing those who had devised the codes was the best way to handle the situation, and sent a letter to the campus newspaper with this aim in mind. Below, I reproduce the letter in its entirety.

I have just been informed that my graduate group's purchases, from nails to computers, will now require a 26-digit code to specify our purchasing account. It is slightly surprising that the complexities of the university's accounts require no less than 26 digits. The old six-digit account codes provided for up to one million different account numbers, which would enable every student and employee of the university to have nearly forty distinct accounts. Only seven digits are needed to provide unique numbers for every office, home, and fax in Philadelphia and some of its suburbs. The Social Security Administration uses nine digits to specify each person in this country; it has done so since its inception during the presidency of Franklin Roosevelt, without yet needing to re-use the numbers of deceased Americans. Ten digits would provide ten billion distinct account numbers, so that every person on Earth could have two of their own.

I understand that part of the reason for the account codes' prodigious length is that they indicate the school, department, and professor with which the account is associated. Even so, a 26-digit code entails a great deal of redundancy. Two digits should suffice to characterize up to one hundred different schools, should the university grow so large; three additional ones would provide for each school to have up to 1000 different departments within it. Three more digits would allow for a unique number for up to 1000 individual professors within each department; if a professor should require as many as ten separate accounts, appending just one more digit would provide each of them with a distinct number. Thus, with just nine digits, accounts could be specified for a university many times larger than our own.

Having to write 26 digits on a regular basis is a nuisance for individual purchasers, who surely have better ways of utilizing their spare moments. For business adminstrators and stockroom managers, however, this represents a serious infringement on their ability to get their work done. Typing in 26 digits--and dealing with the typographical errors that will inevitably ensue--promises to be a time-consuming activity that will exacerbate existing barriers to information transfer around the university. If the intent of developing 26-digit codes is to reduce productivity and thereby create jobs, it is an admirable policy indeed.

I write this letter for several reasons. Partly, I hope that bringing the new 26-digit codes to the attention of the general public will help to avert future administrative fiascos, by tempering the colossal hubris of the persons responsible for this idea. (I have no illusions that 26-digit codes, however unwieldy, will be scrapped while their progenitors are still employed by the university.) To a great extent, though, I am impelled by sheer curiosity about the reasoning behind this new policy, and the mathematical backgrounds of those who created it. I look forward to learning more about their perspective in future issues of this publication.