It is not easy to be perceived as the bad guy. In a room full of management, I once looked my employer's CEO straight in the eye, and told him that the software he was ready to deliver did not work. It did not work, and was one of the ugliest pieces of user interface I had ever seen. In fact, I told him, I had seen school projects that were better quality that the junk the company was about to deliver to a major client. I will never forget that definitive moment of corporate terror, the reflection of panic from the sweat beads streaming down his crinkled face; the death of silence in the room.

What went so wrong that this product was almost allowed to ship? Why was a mid-level software engineer telling the corporate head his watch had just created a pile of software crap destined for the company's best customer? The CEO should have been sweating; he was faced with the very real prospect of contractual breach, and litigation. The testing team had been copious in their documentation of the software verification results, and sent them to the quality manager on a weekly basis. The reports showed consistent failure of major components, and data corruption. Why didn't top management see the damming information that was so pervasive?

This company had a system in place that allowed defective products to ship. But the problem was not technical, it was political. The issue was a quality manager who had more concern for the creation of a façade, than with reporting of reality. Perry and Rice call this situation the "lose-lose" and describe how the quality manager in this case, did not want to be viewed as the "inhibitor to progress" (1997, p. 151). The manger responsible for quality let his employees down; he disappointed the entire company. He did not fulfill his role, showing unethical behavior by blocking the true state of product quality to be reported. Had he been a securities broker, a bank employee, or an accountant--he would have been the subject of a criminal indictment.

And so it was necessary for a subordinate to blow the whistle in an attempt to prevent a travesty--at great personal risk. Thankfully, the company did realize the individual contribution and ultimately fired the offending manager. The defective software was not shipped to the customer, but was completely reworked by a different project team. The client was not pleased at the delay, but was appreciative of the circumstances and the efforts expended in delivering a quality product.

This type of situation cuts to the bone of the corporate ethical culture. The Web site College Journal, analyzing the presence of a formal ethics system suggests questioning a prospective employer "Are workers at all levels encouraged to take responsibility for the consequences of their behavior or to question authority when they are asked to do something that they consider to be wrong?" (No date, par. 6). Another question presented on this site is about formal mechanisms that protect the whistleblower, "Are whistleblowers encouraged and are formal channels available for them to make their concerns known confidentially?" (No date, par. 6). My situation with the CEO was not under any umbrella of formalized protections--he could have easily told me that I was out of line, and fired me without hesitation or recourse.

Thankfully, the incident became legendary in company circles, and served as an archetype for product quality in the organization. When prospective employees evaluate the ethics of this company in the future, they might ask the questions presented, again, on the College Journal Web site, concerning the informal corporate ethics system in place:

"Does analysis of organizational stories and myths reveal individuals who stand up for what's right despite pressure, or is conformity the valued characteristic?", and "Do people get fired or promoted in these stories?" (No date, par 7).

From my experience, the answers to these questions was that the company I worked for did reward me to stand up and speak in the name of quality; that no one will be fired for doing so. The system worked for me, not because of any formalized set of ethical polices, but due to the conscience of top management and a whistleblower who was willing to make a stand. It is unfortunate that this company had to come to the brink of crisis to realize basic product quality attributes. The situation was in need of an engineer who understood how to communicate defects in software, not sweep them under a rug.

The manager who was terrified to be the bearer of bad news had much to learn about software verification and validation. Perry and Rice discuss the difficulty of having to report problems with software, identifying it as "Challenge #1" (1997, p. 161). But such reporting goes with the territory, and those who lack the fortitude should be employed in another field. The proper approach is to depersonalize the problems, and to manage by fact. Perry and Rice present a set of guidelines that will facilitate management test reporting in any situation:

tell the truth, stick to the facts, deliver a solution for each problem reported, and include test reporting as a regular part of the testing process (1997, pp. 172-174).

Staring down the CEO should not be part of the job description for a mid-level quality engineer. However, the organization's informal and formal ethical policies should provide a network of support for the situation, should it be the only recourse. A policy of protection for the whistleblower, the ability to question authority in the name of product quality, and a culture that is supportive of the customer's interests are all elements of an employer's ethics. When you look the boss in the eye to report the worst, make sure your position is concretely supported with facts, and a solid foundation of ethics. "Truth is incontrovertible; malice may attack it, ignorance may deride it, but in the end, there it is." --Winston Churchill.
References

The College Journal, How to Assess An Employer's Ethics. (No Date), Retrieved September 20, 2003, 2003 from http://www.collegejournal.com/resourcecenter/nace/20030115-nace.html .

Perry, William E., Rice, Randall W., (1997). Surviving the Top Ten Challenges of Software Testing. New York: Dorset House Publishing.