The language wars are back!
Well, one war -- between C++ and Rust.
Perhaps "war" is too strong a word. A better description is "a discussion".
Programmers do like to participate in language wars. Or they used to, back in the days prior to open source and the internet. Back then, a programmer worked with the language that was chosen by his employer, or a language for which he (and programmers were overwhelmingly men) had spent money to acquire the compiler or interpreter. One's programming language was fixed, either by company mandate or by finances. That caused false pride in their programming language.
Open source and the internet made it possible for programmers to easily switch. They could try one and change if they found it better than their current language. With the ability to change, programmer's didn't need false pride, and didn't have to argue the merits of a language that they were most likely unhappy with. The language wars faded.
Until now. There were two events this past week, both concerning programming languages.
The first involved the Linux kernel. Linus Torvalds, the chief maintainer for the Linux kernel, announced that he would allow code written in Rust (an up-and-coming programming language) to be part of the kernel. (The kernel, up to now, has been written exclusively in C.) This announcement angered the C++ advocates, who would have preferred that language. Various arguments bounced across the internet, extolling the virtues of each. (Mostly "Rust is a safe language, designed to prevent many mistakes that can happen in C and in C++." and "C++ is a time-tested language with a mature toolset and a large base of experienced developers.")
It wasn't a war, or even a battle, but a discussion with lots of emotion.
The second event was an announcement from Microsoft's CTO for Azure. He stated that organizations and individuals should stop choosing C++ for new projects, and instead pick other languages. (I don't think he listed the languages, but I suspect he would prefer languages supported by the Azure platform.)
That announcement received much less interest. But still, it counts as a volley in the language disputes.
(I suspect that the Microsoft CTO is right, but for a different reason: the size of applications. C++ was designed for large systems, and today's cloud-based services are much smaller. They don't need C++; they need a language larger than C, smaller than C++, and safer than both.)
I find the timing of these two announcements interesting. It may be that we are seeing the beginning of the end for C++.
It may be that historians, in some distant future, draw a line and say "Here, here in 2022, is where C++ began its decline. This is the time that the IT world started to move away from C++."
I don't expect C++ to disappear. Some programming languages have "disappeared" in that they are used for nothing outside of a few museum exhibits and systems run by die-hard fans. Programming languages such as Flowmatic, Neliac, BASIC, and Modula-2 are all but unused in the modern world.
Yet other old languages continue to be used: Cobol, Fortran, and even RPG are running systems today. I expect C++ to join their ranks. They will remain, they will continue to do useful work, and they will be present in popularity surveys. But they won't be at the top.
