algebraist's Journal: Practice, not only theory

/. is running a story about a simulated impact of foam debris on the leading edge of a Space Shuttle wing which gave astonishing results. There is a more detailed story available as well.

While there are many questions that can and could be asked, most of which the CAIB will address in their report, there are questions of overall approach and technique and a philosophy of nature which, it seems to me, need addressing.

In particular, the wing impact experiment recently done is something that might have been done before. The looming question is Why wasn't it attempted? Anyone familiar with NASA's comprehensive approach to engineering and safety knows it must have been considered.

The answer, I believe, reveals a disconnect or fracture along lines distinguishing between "hip pocket engineering", or systems engineering as it is sometimes called, and experimental technique, science, and true engineering. During the Challenger disaster and subsequent investigation, the disconnect between these worlds became apparent, as well as those with decision-making authority having a fundamental misunderstanding of the role and application of statistics. In many ways, this is astonishing since all engineers are trained in the same basic syllabus and have access to the same techniques and outlook.

In my opinion, based in part upon 17 years of work with the now defunct IBM Federal Systems Company and my own decision to leave its successor for ethical reasons, engineers starting in many commercial concerns working in defense and aerospace engineering learn their advancement and success depend crucially upon their putting aside their engineering outlook and its associated ethics, as well as those of allied professions. The needs and terms of the business and of business dealings are discussed and recommended by mentors as more important. Designs are elected, not based upon superior service to a client, their needs, or their mission, but rather to the profitability of the company bidding a job and to their economic success in the long term. Designs which demand continuing and expensive upgrades are preferred, as are designs which are inflexible and difficult to replace.

This grounds my suspicion that in the long run, while capitalism goes well with engineering and scientific applications, corporate capitalism is incompatible, both in terms of ethics and safety, and in terms, ultimately, of its inability to innovate.