Typically what ends up in wide use is historical in nature. C became dominant because it closely maps to the hardware that was available when personal computing went mainstream. A whole generation of programmers learned it, and that influenced language design for decades to come. When everybody is used to doing things a certain way, it can take a long time for new ideas to gain popularity. People often get set in their ways and have a hard time adapting to new ideas.
A lot of the popular languages are fungible because they come from the same family. For example, Ruby, Perl and Python are all extremely similar in nature and offer no fundamental advantages over each other. Then we have system languages like C, C++, and Rust which offer better control over resource utilization.
As a long term trend what's going to drive language popularity is how well the language is suited to tackle popular problems. Functional languages are perfect example of this, as they have been around for many years, but they simply didn't fit well with the available hardware.
Back in the 70s you had single core CPUs coupled with very limited memory. Things like GC were simply not an option and you needed languages that were good at squeezing out every bit of performance possible. Imperative languages addresses this problem extremely well.
However, today we have different kinds of problems. In many cases raw performance is not the most important goal. We want to have things like reliability, scalability, and maintainability for large code bases. We want to be able to take advantage of multicore processors and network clusters.
The imperative paradigm is a very bad fit for these problems because it makes it very difficult to localize change. On the other hand, functional paradigm is a great fit for these problem thanks to its focus on immutability and composition. Today we're seeing a resurgence of interest in functional programming, and Clojure is a perfect example of a Lisp that's quite successful.
A lot of the popular languages are fungible because they come from the same family. For example, Ruby, Perl and Python are all extremely similar in nature and offer no fundamental advantages over each other. Then we have system languages like C, C++, and Rust which offer better control over resource utilization.
As a long term trend what's going to drive language popularity is how well the language is suited to tackle popular problems. Functional languages are perfect example of this, as they have been around for many years, but they simply didn't fit well with the available hardware.
Back in the 70s you had single core CPUs coupled with very limited memory. Things like GC were simply not an option and you needed languages that were good at squeezing out every bit of performance possible. Imperative languages addresses this problem extremely well.
However, today we have different kinds of problems. In many cases raw performance is not the most important goal. We want to have things like reliability, scalability, and maintainability for large code bases. We want to be able to take advantage of multicore processors and network clusters.
The imperative paradigm is a very bad fit for these problems because it makes it very difficult to localize change. On the other hand, functional paradigm is a great fit for these problem thanks to its focus on immutability and composition. Today we're seeing a resurgence of interest in functional programming, and Clojure is a perfect example of a Lisp that's quite successful.
reply