Letters

Most of us take for granted that the bridges we drive on, the airplanes we fly in, and the medicines we use in our lives are the products of careful design and peer review. Unfortunately, as many of us in the software-development field know, the strict checks and balances required by state and federal law for public works (and other engineering products) are not present in the software-development process.

Core to the concept of professional engineering is accountability, and the practice of a senior engineer stamping a set of design drawings and becoming "accountable" for the design. This places a very strong link between a particular engineer and a particular design. Later steps during construction and manufacturing also require a responsible party to stamp design changes and "take ownership."

This is something sorely lacking in today's software-engineering community. I have worked on many projects where the ultimate responsibility for the success of a design is safely hidden in a committee, and management choices (often based on slick advertising media) are used as substitutes for engineering analysis. A licensing requirement would give software developers leverage against unreasonable schedules and pressure to ship unsafe products, since by law, the company would be unable to ship such software without an engineer's "stamp."

Those of us who read Dr. Dobb's are already slightly ahead of the pack, since we at least take some interest in continuing education and maintaining the state of the art. Unfortunately, a large body of the software-development community avoids learning new theories and techniques. A licensing requirement would ensure that the responsible party for the software design maintain a certain number of continuing education units.

As software comes to play a larger role in systems that are core to the safety and well being of the public, I think we will see stronger pressure as a community to become licensed professionals. Rather than fight this reasonable goal, we should embrace the opportunity and work with state and federal regulators to craft meaningful legislation that will help our industry move into the new millennium.

Brent A. Fulgham

bfulgham@debian.org

Dear DDJ,

I am a Registered Professional Engineer (PE) in the state of Texas. I read with some irritation Jonathan Erickson's "Editorial" (DDJ, June 2003), but additional thought led me to realize the man is an idiot. He had better hope that the Texas state legislature is also composed of idiots, and that I am the only one who reads his editorial. A few facts should illustrate my reasoning.

First, in addition to obtaining a college degree in engineering, passing several certification tests, and maintaining liability insurance, I must pay $2350 dollars a year to maintain my engineering license. The lion's share of this amount goes to the state treasury as a tax, as do the $3000-a-day fines.

Second, Texas has a budget shortfall of about $10 million. There are only about 49,000 PEs in Texas. I am sure the legislature would be happy to let every programmer, system analyst, or whatever use the term "software engineer" for about $200 a year.

Third, I have licenses in Oklahoma and Arkansas. Because of those licenses, I have to pay about $3000 a year in training to maintain my license in those states.

Fourth, just to grind home the point, the "Texas Engineering Practice Act" conforms to what used to be called the "model engineering act," which means that a lot of states have similar laws about the use of the terms "engineer," "professional engineer," "licensed engineer," "registered engineer," "registered professional engineer," "licensed professional engineer," and "engineered." See http://www.tbpe.state.tx.us/downloads/laws.htm. Most of those states also have budget shortfalls, so if Erickson wants to be really popular with the readership, he can send his editorial to all of the state boards. I, for one, think he ought to shut up.

Gregory R. Perry, P.E.

PerryG@Airmail.net

Jonathan responds: Gregory, thanks for your note—I think. We're in agreement that high standards are a good thing, and I have no problem with the concept of the Texas Engineering Practice Act. That said, my first complaint is that the Act targets individuals who have not unilaterally appropriated the word "engineer" in their job title, but who have been assigned it by the company they work for. The company who devises the job title ought to be held accountable, not the person who takes the job. Secondly, putting chili-dog fights aside, recent shenanigans by the Texas legislature answer your implied question of whether that august body is composed of idiots. Finally, I conducted an admittedly unscientific poll in the office and at home, and—by a wide margin—my coworkers and family concur with you about my being an idiot, too. Thanks. And on that note, I will, as you suggest, shut up.

Dear DDJ,

The first thing I read in the June 2003 issue of DDJ was Jonathan Erickson's "Editorial" on the use of the term "engineer." As someone with a B.S. in Electrical Engineering who has worked as a software "engineer" (but who would never publicly claim to be an engineer), I very much hope the Texas legislature does not let developers hijack the title "engineer." Why don't developers call themselves "software lawyers" or "software doctors"? Those titles would be as ridiculous. However, I do favor the development of a true Software Engineer title the same way a person can become a Professional Engineer in other fields.

Peter Hoffman

peter@opensourcerers.com

DDJ