5 Programming Languages Doomed Over the Next 10 Years

If you want to build a long career as a software developer, you need to pay attention to those programming languages and other skills that will endure for years, if not decades. After all, it’s hard to earn a living off a language that nobody uses anymore. But which languages are in danger of fading away?

To answer that question, we drew on a number of data sources, including RedMonk and the TIOBE Index (which regularly track the popularity of various programming languages). We also turned to Burning Glass, which analyzes millions of blog postings from around the country, to see which languages face slackening demand over the next 10 years. 

Based on that analysis, we concluded that the following five languages (despite their fans) face significant headwinds in the years ahead. Keep that in mind if you decide to devote time to learning them. 

Perl

According to RedMonk, which ranks programming languages according to data from GitHub and Stack Overflow, Perl has plunged from 11th place in 2012 to 18th place today. That is… not good. Even worse, Burning Glass predicts that demand for Perl-related jobs will tumble 22.1 percent over the next 10 years. 

As readers have pointed out in the past, there’s still a robust developer community around Perl, with updates rolling out on an annual basis. But when we look at job demand, along with code contributed to GitHub, it seems that Perl is seriously on the decline.  

Objective-C

Over the past 12 months, Objective-C has plunged eight ranks on the TIOBE Index—the steepest annual drop of any language in the top 20 (it now sits in 18th place). That’s exactly what Apple, which launched Objective-C 36 years ago as the language for building apps in its ecosystem, really wants; for the past six years, it’s done everything in its power to convince developers to switch to Swiftits new-and-improved programming language.   

Despite that pressure, Objective-C has managed to cling to relatively top positions on the various programming-language rankings, largely because of three-plus decades of legacy code. At some point, though, Objective-C will fade away: Burning Glass predicts that jobs featuring the language will tumble 34.5 percent over the next 10 years. 

C

As a language older than many of the developers who work with it, C occupies a vaunted place within the programming world. It’s found its way into a mind-boggling number of applications, from embedded systems to supercomputers. However, Burning Glass suggests that jobs requesting C will dip 14.2 percent over the next 10 years, suggesting that developers are turning to other languages in order to efficiently tackle the building and maintenance of those applications.

Haskell

Although Haskell has its die-hard fans, especially longtime developers who appreciate some of its features (including type classes), the language hasn’t done well on RedMonk’s long-term language rankings, suggesting that there’s little developer chatter around it.

If Haskell faces declining use after three solid decades, at least its fans can comfort themselves by recalling the language’s impact on many of tech’s biggest sites and companies, including Facebook and GitHub.

Ruby

Signs of Ruby’s decline have been present for years. It’s slowly fallen on TIOBE’s programming-language rankings, and last year’s analysis of Dice job-posting data showed a strong dip in the number of companies that wanted technologists skilled in the language. But if you want a true warning sign that Ruby is on the way out, look no further than Burning Glass, which predicts that Ruby jobs will tumble 8.1 percent over the next decade. 

42 Responses to “5 Programming Languages Doomed Over the Next 10 Years”

  1. Karol Tesar

    I dont agree with C at all. Despite the language itself is 50 years old, there is no other viable alternative on the hardware level or in embedded domain, and it will not appear in the near future. Why? Its 50 years of legacy of all hardware vendors. Even new hardware designs are aligned with C standard and the C compiler is the only thing they provide in most cases for new hardware. The linux kernel is written in C, one of the biggest open source projects which runs the world (and is also base for Android). No other language is allowed in that code base. Rust is optional for non-critical parts only in the drivers. In the embedded domain C is over 80% market share. Check the statistics over last 20 years tracked by the BarrGroup or the Tiobe Index (1st place in the list and still going up). You will find only subset of microcontrollers which run Rust and C++ works only for high level MCUs which are expensive for most use cases. There are basically 3 types of embedded platforms: RTOS based, Linux based and Android based. The first one, with constrained memory up to 256KB and cheap components in the range of few cents, makes roughly 90% of all CPU deployments, every device around you has bunch of low power microcontrolers and optionally one powerful CPU. In the world of IoT the trend is quite opposite. You dont have resources to use more powerful hardware with Linux kernel and hence variety of other languages like python, C++ or javascript, but you have to be as low power as possible. Because of money (bulk manufacturing) or power requirements (battery powered). C will stay probably until mankind will change the whole hadware underneath with something different technology. Quantum computers? Maybe, but not in the next 10 years.

  2. Mike Synnott

    More ill-conceived, clickbait garbage. If you honestly believe, for one second, that C is ‘doomed’, you are incapable of thinking past high-level application development to low-level, embedded, device driver or OS development. Had this article claimed that C will continue to decrease in importance for mainstream application development, rather than predicting its utter demise within 10 years, I’d have agreed with you.

  3. Gian Constantine

    This is the dumbest article on programming I have read in a long time. Get a writer who knows more than Java, JS, and Python (which means the writer does very little beyond enterprise and web apps). The Internet runs more on C (and to some smaller extent Perl) than really anything else. You really have no idea.

        • Jeff H Silverman

          I think he miswrote. C is useful in the cybersecurity world doing compute intensive things such as brute forcing passwords. I think what he meant is that there are a lot of places where C can be, well, unforgiving. So, for example, if neglect to free() something after you malloc()’d it, or worse, free() it twice. If you make your buffer 1 byte too small, that could be lethal when somebody else’s pointer trashes that last byte with something other than a 0.

          Full disclosure: I like writing in C. My arduinos use nothing else.

  4. PRITESH M UGRANKAR

    Perl version 5.32 was released a few weeks ago. It’s still used a lot for automation. And I’ve come across many articles and updates that state Perl is still heavily used in finance and biotech.
    Perl is a fantastic language, excellent at regex, report creation, web development (Dancer and Mojolicious). Perl can create data structures on the fly, is faster than other similar scripting languages, and most importantly, the Perl community is a group of brilliant and helpful folks ready to teach you the idioms of the language at the drop of a hat. There’s new Perl code being written, and per a good friend of mine, Perl is used for many new projects…sorry but disagree with the points….your article seems a rehashed version on the internet echo chamber…

    • David Hagens

      I completely agree….perl is NOT going anywhere, just like C. It is the best regex language out there ever as far as I’m concerned it was designed in a manner for regex / strings just like you suggested, it’s even in the name of the language itself lol!

  5. Mahendra Singh

    C and C++ will never die, all the Basic coding right from OS for windows,unix and vast majority of software used in internet has been designed using much of above program. You writer have simply termed and classified based on earnings,not of value

  6. Ilia Kaliuzhny

    Dooming C is just laughable. The only viable alternative is RUST, but that will not pick up everything over next 10 years. The world is being run with stuff written in C. I think I wont live to see C disappear.

  7. I do not think C is going anywhere. Yes it has been around a long time but the basics that it is built on are solid. I also do not see Perl going anywhere. It is fast and the preferred method to get something in Unix done with low CPU overhead.

  8. Lee McGraw

    C will never die, and it’s usage will go up, in fact. Anyone that thinks C is going away doesn’t realize WHY C is and was used in the first place. C is extremely fast, and required minimum hardware to run it. That’s it’s key and it’s niche. C# is far better for enterprise applications running on servers as big as the average person, but C is wonder when hardware limitations exist, like IoT, drones, etc. C++ is a middle point between C and C#. C++ is often used when C would be used, but also comes with OOP principles and concepts. Until IoT, drones, and hardware exist with no limitations on memory and processor speed, C will always be around.

  9. Ray Andrews

    A neighbors child held up their hand slowly and said “butterfly”, it was in fact a butterfly. Their observation, research, and conclusions were more more thorough and accurate than the author here.

  10. To be fair the author does interpret “doomed” in the case of C as “jobs requiring it fall by 14.2%”. However I’m not sure how many of those people claiming C (or C++) will be here for ever have learned Rust to a reasonable level of competency.

  11. Robert Neuendorf

    C is the base for a number of operating systems. Your higher languages many times rely on C calls for system-level operations. C is not going anywhere until the operating systems that use it are rewritten. Professionally, I build winforms using Visual Basic. Within the next 10 years, VB will be a memory. I will be close to retirement, but I’m working on transition to C# and perhaps xaml as a windows-based form solution.

  12. In reality popularity of languages is also about their applications, embedded is a different world from mainframe and PC’s are different again.

    In the embedded/SBC/IOT world its hard to imagine C dropping out of use. Its logical alternative C++ for low power systems (IOT) is lowly becoming less and less practical as the ISO committee add more and more features. With cross compilers becoming less easy to come by for the embedded market due to cost and the ever changing specifications.

    But C is solid, unchanging and does exactly what its designed for, without interference from academia or over zealous standards committees!

    Yes C is a dangerous language for the careless, but its that ability to drill down to the metal unheeded and unprotected that gives it its power and usefulness.

    Perhaps C could be replaced with a C++ with a reduced and more practical specification – with backwards compatibility with C of course!

    The main risk to C usage is if even the smallest computing device has oodles of memory and power available, look at Android phones using Java – Certainly possible but not right now.

  13. Well Nick, I don’t know where you get your data from, nor do I know just how much research you’ve done to make your claims. But if I go on what you have placed up in your article as something which resembles authoritative, then I may as well shoot myself in the head.

    Now, with all that said, done and dusted, let me say this. Having been a programmer for more than 40 years up my belt, and knowing what the industry is like, I guess I just have it a tad over you. So when it comes down to the popularity basis of what a better or more popular language, and what is not, in reality is just opinion, and like any opinion its just purely subjective, and of which outside of the opinion holder, its just case of so what, who cares, means nothing.

    You see, the reality of the programmer, is that he has to use the tools that are available to him, and no choice in that matter at all. And if that programmer doesn’t know how to use that tool, then that programmer is a useful as a hip-pocket on a t-shirt.

    The point being made here, is that most devs don’t get the choice, or have the luxury to choice the language he wants to use, because they have to work for the company that uses its own favorite language. About the only time a dev gets to use his favorite language is if he has his own business.

  14. Thomas G

    I look at it this way. They have been threating that programing would go the way of the dodo for years now. Most who write this kind of list focus on what is the trending language of the day. We are dealing with languages. The only way a programing language falls out of use is that the number of people who know it drops to a point that no one remembers how to program in it. Plus, if one know anything about history. Sometimes it is a lot harder to replace something than just keep it working. For one has to realy understand what the original source does and how to get the data out of the system

  15. John French

    It is a tad amusing that the one listed programming language the other listed languages are designed and built from, is slated for demise.
    C, one of, if not considered the first, structured language is not headed anywhere other than being compiled by more programmers.
    next up will be assembly languages headed for that Write only RAM cloud.

  16. During the 1960’s and 1970’s I worked at Bell Telephone Labs (BTL)
    on the first computer-based telephone switching systems. For each switch type we tended to develop a new processor with more advanced features than the previous generation. Since these real-time systems had to get the maximum performance out of relatively weak processors, programs were written in assembly language, and so each processor type had its own instruction set and programming tools. By the late 1960’s we BTL engineers had cranked out millions of lines of assembly code, and the cost of maintaining and evolving that code was becoming a huge factor in our budget projections. Furthermore, since telephone systems were required by Act of Congress to run for 40 years (don’t ask why), we were facing the unhappy prospect of hiring and training engineers in the 21st century to write code for systems designed in the mid-20th. Tackling this problem, the BTL R&D organization concluded that we needed a programming language that would work across all of our processor types, in other words, a portable assembly language. Working towards that goal, Dennis Ritchie and his team came up with the C language, which was so simple that C-to-assembler compilers were easy to build. Several years later I left BTL and co-founded Lattice as a company developing C compilers for commercial processors from companies like Intel and Motorola. We also licensed our compilers to larger software companies such as Microsoft.

    The bottom line of all this is that C has a purpose today (i.e. portable assembly language) even more important than when it was invented, simply because the computer business produces many more processor types than the dozen or so that we designed at BTL. So, if you need to deploy high-speed real-time software using every machine cycle efficiently, chances are you’ll write that code in C.

  17. Keith Hussey

    I think that the prediction that C is doomed is indicative of what has been happening in our industry. Those of us who earned the title of “computer programmer” back when that was a thing (and before there was google or stack overflow – I know, I’m old) know otherwise. But the fresh ones who call themselves that and other, more lofty things can’t survive without a high-speed internet and these newer “training wheel” languages. It is for them very fortunate that resources are plentiful now so that all of that dead weight can be absorbed. I wish them well and I, for one, feel very fortunate that I made my bones in this industry when that still meant something.

  18. I find this article mostly amusing. Most of it is very predicable, in fact more so than many of those training languages ever becoming popular for developing finished applications. However to predict that a language is dead you would first have to be able to prove that other languages have died. Fortran is still around, as is Cobol, A, and B are still used by some people who still love them. And, you have to consider that the original language, machine language is not likely to ever die of old age.

    If you are going to make assertions in front of computer programmers, make sure you can prove them reasonably well 😉

  19. Simon C. Suñas

    I don’t think that C is doomed. Maybe now a days more school are using Python, but still C is a very good start for building up programmers foundation. In my school days we use Pascal and C. Programming in C is a programmer start-up road to learn C#, C++, JavaScript (all those framework) and Java. For me, if you know C all those well know application language today are easy for you to learn.

  20. Fred Flintstone

    if you like and use any of these programing languages you would be riled up by this article. So that would be a lot of people. Me personally, I just clicked the link to make sure mine was not on it.

  21. non of them will die.

    the only language that is going to die is Java.

    it can only pass by value simple types and is passing by sharing all composite types (passing a reference to). you can write of functional programming an such a language, because pure functions are impossible.

    it forbid procedural programming as a “lower species” and you will know when Java spits on it’s own face when some years after it reintroduces procedural programing. Kotlin has support, Go and Rust are procedural.

  22. Red Rooster

    What a nonsense list.
    C and even Perl will still be in use 10 years ago, just as they are nowadays even though their death was prematurely announced 10 years ago.
    Haskell never was, Objective-C was always a tool for masochists and found its demise when Swift came out (not that this is that great of an improvement in the first place) and Ruby always was a “wanna-be” that kind of died years ago, if it wouldn’t be for some web related companies that have bet their farm on it an now can’t (easily) switch to something use full…