Standardization with regard to the operational safety of computer software

Standardization in the field of software dependability (SdF) is the answer to two successive questions facing the various fields in which software plays an important role in terms of the safety of people and property.

Why do we need to standardize software SoTL?

To say that software has been omnipresent in the systems that surround us for many years now is almost an understatement. The stocktaking exercise necessitated by Y2K fears highlighted this trend, which has continued unabated ever since.

The functions performed by such software are becoming increasingly critical in terms of the safety of people, the environment and property, and it is vital for all those involved (designers/developers, purchasers, control bodies, operators/users) to have the greatest possible control over the operation and malfunctions of these software-intensive systems.

To achieve, or attempt to achieve, such a goal, it became necessary to try and standardize the practices to be implemented. This awareness has developed on a field-by-field basis, in parallel with the increasing preponderance of software in the field in question.

What are the general standardization principles for software dependability?

Whereas the quality approach must find an optimum between under-quality and over-quality, the SoF approach can be a permanent bidding war for a single objective: zero defects.

This objective, which we know to be unattainable, and above all virtually unprovable in view of the increasing size of software, is increasingly being replaced by a more controlled objective: zero risk.

In this context, the main aim of SoTL standardization is to identify the nature and consequences of risks, and, depending on the level of risk identified, the properties to be respected, the methods imposed or strongly recommended, and the evidence expected of their application, with the main aim of ensuring that customers, certification bodies and project managers are able to control all these risks.