High-pressure steam engines and computer software

It is argued that we do not want to impede progress by writing unachievable standards or inadvertently increase risk by implementing the wrong standards. We have not scientifically established the benefits and effectiveness of most of our software engineering techniques. Depending on a particular software engineering methodology to assure safety by assuming it will produce error-free or ultra-high reliability software is dangerous. And as the technology progresses, standards that require the use of specific approaches often lag behind. Manufacturers may feel no ethical or legal duty to go beyond what is required in the standard. Moreover, manufacturers or those who will personally benefit financially from particular techniques being included or not included in the standards sometimes play a dominant role in the drafting process. The result may be watered down req1~irements or the recommendation of techniques with more commercial than technical value. The alternative is to construct flexible standards specifying general criteria for acceptability of a methodology instead of a specific methodology and ensuring that those building safety-critical software have the competency and personal responsibility to use the best approaches available at the time and for the particular project characteristics. As Edison argued with respect to electricity, increased government regulation of our technology may not be to anyone's benefit; but it is inevitable unless we, as the technology's developers and users, take the steps necessary to ensure safety in the devices that are constructed and technical competencies in those that construct them.