Every single one of them, when viewed with a long enough time scale ;)
Someone, somewhere is still using almost anything you can think of. So it really depends on what you mean by "popular" and "relevant". eg. Java and PHP aren't "cool" but neither one's going away any time soon.
I know a number of places that have used (and some still use) Adobe ColdFusion for server-side development: en.wikipedia.org/wiki/Adobe_ColdFusion
I had thought most people using ColdFusion had switched to PHP over a decade ago, but there are still some holdouts from the pre-PHP days
At the moment, I think that old languages with small communities are slowly fading away. There might be people, who still use them, however the numbers shrink by the day. Imho, Cobol and Ada are some examples.
Imo, in the short-term, languages which do not provide security mechanisms and modern features, like proper multiprocessor support, and are inconvenient to the developer, will have a hard time and fade away. I think Perl, Object-Pascal and Objective-C are candidates for that.
In the long-term, most languages will be deprecated and replaced by languages which fit better into the tech-world of the time. There might be a few exceptions, which stick around very long, because they are standardized on a grand scale (for example ECMAScript derived languages, JavaScript in particular). This will even affect today's most important languages, like C++ (people rumor that Rust will deprecate it) and GLSL (with the introduction of SPIR-V, even C and Rust become possible languages for writing shaders).
This is a really good list to pay attention to, mostly reindexed each month by the amount of activity for each language.
I see on most benchmarks that Python is one of the slowest languages.
Todd
Software Security TechLead
We have to be careful with this because it's illogical for a programming language to really become "irrelevant" if you think about it hard enough. The reason why is because most of these languages are all Turing Complete which, in this context, means that each one can do anything any other can. Therefore, it's sort of like asking "which raft color is less popular at the raft store?"
The raft color question is a valid question because in reality, perhaps blue rafts sell way more quantities than gold rafts... However, every single raft will perform the exact same way at keeping you afloat on the water, regardless of its color. So while a green raft could become, as you say, "irrelevant," is that even really true? It depends on what you're looking at. I look at a raft's ability to keep me afloat in which case the green raft, even if it sells 0 ever, is still relevant in my eyes.
This is how I view programming languages. Languages which are good for some given task aren't so mostly due to the constructs of the actual language, but rather, the cultural following behind each language. I.E. "What was the purpose the language author conceived the language for?" and "Which community has taken to the language the most resulting in the most libraries?"
Another thing too is, many would consider assembly language to be "irrelevant" today because I don't know a single person who really programs in it for anything major... But it's not at all, for without it, you would not be able to effectively analyze binaries and find vulnerabilities and such, since it is an explanation of machine code essentially. Would you say now that we have marble flooring and beautiful chandeliers, that the ugly, boring, concrete foundation is irrelevant??