Forum

I enjoyed Stephen B. Jenkins’s "Technical Opinion" ("Musings of an `Old-School’ Programmer," May 2006) because to some extent I feel the same way. Like Jenkins, I, too, do not view design, development, and debugging as separate tasks. (We are also probably about the same age; I wrote my first program, in Fortran on punched cards, in 1971.)

But on two of Jenkins’s points, I would like to offer another perspective. First, the visual development tools and integrated development environments of recent years include several features I find much better than their older counterparts. One in particular is the automatic reference checking and code completion provided by Microsoft’s IntelliSense and other such programming innovations. IntelliSense facilitates programming so well that I never want to program again without it.

The second is how much programming is like building custom motorcycles. I know nothing about building custom motorcycles, so I take Jenkins’s word for it. But I do see much similarity with writing a well-structured essay, an attitude that may come from my having been an English major in college. When writing an essay, it is best to start with an overall design, then craft each section to fit within it, making sure everything fits properly, works well, and feels right. Like programming (and motorcycle building) writing requires creativity, the ability to visualize and systematically produce the final product, and a knack for knowing what works (in this case, writing in a way that attracts the reader and conveys what the author intended), along with a sense of ownership and pride. One major difference is that if the job is not done well, it is just bad prose on paper, rather than a motorcycle or program that doesn’t run.

Richard Veith
Port Murray, NJ

Back to Top

Nurture the Joy of Engineering

Jack Ganssle’s "Forum" comment "Before Coding, Learn Disciplined Engineering" (Apr. 2006) missed the mark. The idea that software developers should learn engineering discipline before learning to code is ludicrous and not supported by the two creative disciplines Ganssle cited. Instead, we should recognize that coding and software engineering are different disciplines, that not every coder is a software engineer, and that companies must employ their technical resources with these factors in mind.

Discipline is certainly a problem in all areas of engineering, and software engineering has always been problematic. But, as Ganssle pointed out, it has little to do with coding, and thus his examples involving university English and music departments are invalid.

As anyone with a background in English would attest, writing stories and novels is a skill that takes years of practice to develop and begins well before most writers enroll in an English class. Studying the masters is a task pursued in parallel; while there is much to be learned from the masters, there is no substitute for the act of writing. Music education proceeds along a similar course, with creative exercises in the first year of instruction, even as students study the masters of many genres.

If there is a difference in coder education, it is that no one expects great composers and novelists to produce masterworks after only four years of instruction.

The greatest barrier to the adoption of good process is the engineering community itself. Much of the knowledge of the practice of engineering is being lost due to the trend away from maintaining an engineering brain trust within many companies.

We have been moving from lifetime employment to the commoditization of engineering talent. Where once young engineers were mentored and nurtured within a company, companies now hire young engineers out of school at bargain rates and expect them to arrive fully fledged. Our colleges and universities have largely bought into this approach, and curricula have changed to emphasize specialization at an early age.

Engineering comprises many things: ethics, discipline, knowledge, experience, talent, and a touch of hubris. Behind it all, however, lies the joy of creating. There are many reasons for the lack of discipline Ganssle lamented, but the idea that we should teach discipline before coding is way off the mark. The joy of engineering begins with the first disassembled television, the first popsicle-stick bridge, or the first toy bulldozer in the sandbox.

Discipline is appreciated late, and companies must address this reality by maintaining a core of experienced, disciplined technologists to train and guide the new ones just entering the field.

Gregory Aicklen
McKinney, TX

Back to Top

Don’t Equate Bankruptcy and Identity Theft

Rebecca T. Mercuri’s "Security Watch" column ("Scoping Identity Theft," May 2006) was seriously misleading on the subject of U.S. bankruptcy. It claimed that $47.6 billion is attributable to identity theft and that the annual cost of bankruptcies is $34.5 billion. It further claimed: "Since scammers tend to take the road of least resistance…bankruptcy fraud may continue to seem a lot more attractive as a source of free buying power than identity theft." The next sentence added: "According to the FBI, up to 10% of bankruptcy filings may involve fraud," that is, $34.5 billion 3 10% = $3.45 billion. Therefore the "institutional cost" of identity theft—$47.6 billion—is more than an order of magnitude greater than bankruptcy at $3.45 billion. Despite this gross difference in magnitude Mercuri implied that the costs of bankruptcy are "not terribly far from" the costs of identity theft.

Worse, a pull quote in the column showcased the bankruptcy bogeyman, displaying Mercuri’s unsupported assertion that bankruptcy fraud is a choice vehicle for scammers. Finally, this quote included the phrase "…may continue to seem…" What precisely did that mean? It means that Mercuri and Communications should have realized the portion of her column pointing to bankruptcy was tenuous at best and should not have been included if Communications wished to remain a credible scientific journal.

John Knight
Salt Lake City, UT

Author’s Response:

John Knight’s criticism of my thoroughly fact-checked data on bankruptcy vs. identity theft is based on an out-of-context sentence he chose to highlight from a paragraph in which I compared identity theft with personal bankruptcies. But the larger numbers (and point) are spelled out in the subsequent paragraph where I described institutional bankruptcy figures as eclipsing identity theft losses by more than an order of magnitude during the years surveyed. Since both types of bankruptcy may be shrouded behind the corporate veil, these lines may actually be further blurred.

As for the concerns regarding Communications being a "credible scientific journal," although it is certainly both credible and scientific, it is, by its own description, "the internationally acknowledged premier magazine of the computing field," though of such high quality it might occasionally be mistaken for a journal.

I intend in my column to raise awareness of trends and issues in computer security, hoping that others might further investigate these matters and either confirm or refute my conjectures through further theoretical research. I encourage publication of such results in our sibling publication, the Journal of the Association of Computing Machinery.

Rebecca Mercuri
Philadelphia

Back to Top

What Malaysia Looks for When Funding CS Study

I was encouraged by what Mazliza Othman and Rodziah Latih had to say in their article "Women in Computer Science: No Shortage Here!" (Mar. 2006) about the high proportion of women in computer science and information technology in Malaysia, but I was also surprised that no mention was made of a particular aspect of Malaysia’s education system that might directly affect these observations. As in many countries, Malaysia offers government loans to some of its students. To ensure that these public funds go toward developing and supporting the economy of the country, the government reserves and sometimes exercises the right to decide which subjects students will study. So I would like to know what proportion of the students were assigned to their courses in this manner and what proportion chose the courses themselves.

D.J. Howorth
Hitchin, U.K.

Author’s Response:

The government of Malaysia does indeed offer study loans, but this practice does not influence students’ choice of which programs to pursue, as they apply for a loan only after enrolling in a program of their choice. Howorth is apparently referring to a scholarship program offered to students who wish to study abroad, in which the government determines how many scholarships are offered for a particular program. Since these scholarships are for students going abroad, it has no affect on the observations we made in our article. We must also point out that the number of scholarships offered for computer science programs has decreased steadily over the past decade.

Mazliza Othman
Kuala Lampur, Malaysia
Rodziah Latih
Selangor, Malaysia

Back to Top

We’re from Taiwan, Not from China

Robert L. Glass’s "Practical Programmer" column ("Is the Crouching Tiger a Threat?," Mar. 2006) raised an issue of concern to my colleagues and me. Many of us noted that in Glass’s Fact 4, concerning Asian universities beginning to dominate the Top Institutions portion of the annual Top Scholars and Institutions survey, his claim that "National Chiao Tung University of China is number two" should have been "National Chiao Tung University of Taiwan is number two," because Taiwan has its own identity. As the Dean of the College of Computer Science, I know that National Chiao Tung University is in Taiwan and does not belong to China. We would even like to invite Dr. Glass to give a talk here in Taiwan so he can see the big difference between Taiwan and China.

Chin-Teng Lin
Hsinchu, Taiwan

Back to Top