>The problem is that safety critical systems is not (yet) a big enough "niche." >You would at least have to generalize it to "business critical" or >"mission critical" or simply "critical." I think we want to focus on >systems where errors are very expensive, even if no lives are directly >threatened. Good point, but if Ada were to own the position of "the safety critical language," that is if you asked the typical engineer on the street about Ada and he said, "oh yes, that's the safety critical language," wouldn't that be appealing to program managers in other market segments? Gee, if it'll keep a 777 in the air, it ought to work better than C++ on my project. Seems like this position provides some high ground that offers smooth transitions to other areas. Business critical and mission critical live next door and can be easily assumed if you own safety critical. But I wouldn't take business critical or mission critical as Ada's position. They don't mean anything because it can be said nearly any language is for business or mission critical applications. Safety critical is much more vivid and I think Ada is the only [potentially] mainstream language that can substantiate its superiority in this area. You're right though -- you wouldn't want Ada pigeon holed for only that market segment. It needs to be clear that the qualities that make Ada the safety critical language also makes it superior for other market segments.