> Will we just be stuck in a local maxima of Java for next 50 years?
1. Yes, if the extent of the imagination is languages like Idris and ideas like effect systems, that follow a gradient descent from Java, and always in the same direction: being able to express more constraints. What you get "for free" from such languages may not be significant enough to justify the cost of adoption, and the valuable stuff you can get is not much easier than the options available today, which are too hard for anyone to take. If you were to consider truly novel languages that think out of the box (e.g. Dedalus/Eve) then maybe one will stick and make an actual impact rather than just a change in fashion.
2. How do you even know that we can do much better? NASA engineers may not like it, but they don't complain that we're "stuck" at sub-light speeds. Maybe Brooks was right and we are close to the theoretical limit (that we know must exist).
> We talk about languages as a bag of feelings and fuzzy weasel words that amount to “It works for my project”.
Can you find another useful way, available to us today, of talking about languages?
> “Use the right tool for the job” Zero information statement.
That's right, but it's not a dumb cliché so much as it is a tool we've developed to shut down religious/Aristotelian arguments that are themselves devoid of any applicable, actionable data. One, then, is often confronted with the reply, "but would you use assembly/Cobol?" to which the answer is, "of course, and it's not even uncommon, and if you don't know that, then you should learn more about the software industry before giving any more Aristotelian arguments."
1. Yes, if the extent of the imagination is languages like Idris and ideas like effect systems, that follow a gradient descent from Java, and always in the same direction: being able to express more constraints. What you get "for free" from such languages may not be significant enough to justify the cost of adoption, and the valuable stuff you can get is not much easier than the options available today, which are too hard for anyone to take. If you were to consider truly novel languages that think out of the box (e.g. Dedalus/Eve) then maybe one will stick and make an actual impact rather than just a change in fashion.
2. How do you even know that we can do much better? NASA engineers may not like it, but they don't complain that we're "stuck" at sub-light speeds. Maybe Brooks was right and we are close to the theoretical limit (that we know must exist).
> We talk about languages as a bag of feelings and fuzzy weasel words that amount to “It works for my project”.
Can you find another useful way, available to us today, of talking about languages?
> “Use the right tool for the job” Zero information statement.
That's right, but it's not a dumb cliché so much as it is a tool we've developed to shut down religious/Aristotelian arguments that are themselves devoid of any applicable, actionable data. One, then, is often confronted with the reply, "but would you use assembly/Cobol?" to which the answer is, "of course, and it's not even uncommon, and if you don't know that, then you should learn more about the software industry before giving any more Aristotelian arguments."