Re: New IEEE Language Popularity Ratings

[Maciej Sobczak <see.my.homepage@gmail.com>]

> In the defense area in the US, "personnel reasons" is an excuse. The real reason
> is to maximize profits.

I disagree. I can see you beating the same argument over and over again, but I don't find it to be substantial.

In defense or safety-critical industry in general, there are enough ways to generate so called "work hours" that the selection of technology is irrelevant. In some industries the statistics is that only about 5% to 10% of time is spent on actual coding, the rest is spent on maintaining all other artifacts, the set of which can be arbitrarily large, with arbitratily complex interdependencies and kept in multiple incompatible systems, etc., that can take ages to modify to reflect even a trivial functional change. Of course, all activities must be fully manual with no chance of being automated. In a such a setup it does not matter what is the implementation language, it might be anything - the treadmill of "work hours" is spinning elsewhere.

> A
> poorly designed, error-prone language is an additional tool for maximizing profits.

Then everybody would choose assembler if that was the real reason, but somehow this is not the case.
Interestingly, the choice of technology is frequently driven by the client themselves (that is, contractors have no influence on this choice), which further invalidates your argument.

A poorly designed software development process is orders of magnitude more effective in generating profits.

-- 
Maciej Sobczak * http://www.inspirel.com