It's not bad at all, though the organization is a bit funny. I get the impression of an organization that knows a great deal struggling to get it all down on p

It's not bad at all, though the organization is a bit funny. I get the impression of an organization that knows a great deal struggling to get it all down on paper. Specific references to programming languages, tools, RTOS choices, etc are obviously dated, and specifics of NASA's processes in 2004 won't necessarily apply to whatever you're doing. Their best practices checklists for different programming languages were not exhaustive at the time, and are written like suggestions - I would supplement with more recent approaches there too. That said, methods of safety analysis have not changed much, and neither has what-makes-an-RTOS-an-RTOS, for example.