One reason is that this is mostly a brain firmware phenomenon. My personal observation is that some people are able to nail a domain with good conceptual integrity, but there are also many (clever) people that quickly reach to a mechanical implementation, often even unable to think on a conceptual level about the problem domain. Languages with proper sum types are really helpful here, but Haskell was 20 years ago already 16 years old.
Nonetheless, the language is not the real impediment, it is the way of approaching a problem. Also, the software we are talking about was a standalone application installed on laptops, they were not part of the airplane's board computer.
Comments
One reason is that this is mostly a brain firmware phenomenon. My personal observation is that some people are able to nail a domain with good conceptual integrity, but there are also many (clever) people that quickly reach to a mechanical implementation, often even unable to think on a conceptual level about the problem domain. Languages with proper sum types are really helpful here, but Haskell was 20 years ago already 16 years old.
Nonetheless, the language is not the real impediment, it is the way of approaching a problem. Also, the software we are talking about was a standalone application installed on laptops, they were not part of the airplane's board computer.