Safety Critical Linux - Working Group Proposal by Nicholas Mc Guire

Industrial Demand

The industrial demand for safety critical systems is apparent from the development of standards and the evolution of technical safety standards. Beyond this one can notice a clear push in the direction of COTS/OSS, not only for reasons of direct cost reduction (license fees) but also due to a number of other issues that are generally not so present in the public discussion on COTS/OSS:

• OSS is long term - there is no single company behind GNU/Linux that could go out of business
• Human resource base - it is substantially easier to find a good Linux kernel hacker than it is to find a VxWorks or QNX developer.
• Generality of technology - technological costs are reduced by utilizing know how in other (often unrelated areas) i.e. Engineers using UNIX on a day-by-day basis are more effective in building safety critical systems if these are based on the same technologies than if they must "context-switch" when working on safety critical projects.
• OSS/COTS based development can reuse existing tool know-how.
• The Free-software/Open-source community can improve system stability due to its peer-review nature
• The substantial works on GNU/Linux in related fields, notably High-availability systems and security will improve safety centric design and ease development of such systems