DyingProgrammingLanguages
ThoughtStorms Wiki
Contrast : ProgrammingLanguageSuccess
A sad but interesting story about why and how one company that was very into Elm Language, eventually gave it up : https://kevinyank.com/posts/on-endings-why-how-we-retired-elm-at-culture-amp/
Quora Answer : What are some of the dying programming languages in the next decade?
OK. Just to stir things a bit, I wonder if Ruby has peaked :-P
Ruby's burst of fame was almost entirely based on Rails. But current trends don't seem to favour it :
- the rise of rich HTML5 apps. that do more of their work in the browser in javascript.
- the rise of serverside javascript via node.js etc.
- the trend towards more dynamic queue-based message-passing services like WhatsApp which encourage developers to play with Erlang
- the general rise of functional programming as the new hotness (Clojure / ClojureScript, Haskell / Elm-Lang, Scala, Erlang etc.)
- the persistence of PHP as a reliable workhorse for a lot of more pedestrian web-work
Could Ruby go the same way as Perl?
Python suffers from similar trends, of course, but unlike Ruby seems to have some other strong power-bases : as a system scripting language, as a beginners' learn-to-program language (Python is the new BASIC), in scientific and academic computing. I'm not sure if Ruby has any equivalent areas where it's the preferred choice.
Quora Answer : Is Ruby dying?
Ruby has an odd history. It blew up because of Rails. Which was a state of the art web framework about 10 years ago. Many Ruby coders were Rails programmers. And most Ruby work is Rails.
It's still fantastically popular (see other answers here). And there are going to be a lot of legacy Rails sites around for a long time.
BUT ... as Giles Bowkett pointed out a couple of years ago, it's no longer the new hotness.
Python (Ruby's great rival) has found itself a more diverse range of niches : system scripting (replacing Perl), scientific / statistical / data processing (replacing Fortran / complementing R), teaching language (the true successor of BASIC, less hassle than Java) as well as web-sites.
The rise of node.js has launched a dozen hot new web-frameworks based on javascript running on the server and taking advantage of fast VM research and event-driven programming.
The rise of mobile devices has actually pushed programmers back to compiled "system" programming languages like Java (on Android) and Objective C (on iOS).
Today's cool kids are pushing on into Functional Programming languages like Haskell, Erlang and Clojure. And the cool enterprise kids are starting to make a big bet on Scala. That's also because the "web" is becoming more about juggling huge numbers of streams of notifications being pushed down to "fire-hoses" like Facebook and Twitter rather than serving up custom but static pages.
So, even though Ruby is popular, I think there are structural reasons why it might suffer a collapse almost as sudden as its rise.
OTOH someone may have pioneered a new niche for it that I'm unaware of.
Update Nov 2018 :
Ruby isn't, of course, "dying".
But it's shocking to see how, say, Python has pulled ahead of it in this Github report : The State of the Octoverse: top programming languages of 2018
Obviously that's all about data. But back in 2014 Ruby and Python were still about neck and neck.
But look at them now, only four years later.
Quora Answer : What are some of the dying programming languages in the next decade?
OK. Just to stir things a bit, I wonder if Ruby has peaked :-P
Ruby's burst of fame was almost entirely based on Rails. But current trends don't seem to favour it :
- the rise of rich HTML5 apps. that do more of their work in the browser in javascript.
- the rise of serverside javascript via node.js etc.
- the trend towards more dynamic queue-based message-passing services like WhatsApp which encourage developers to play with Erlang
- the general rise of functional programming as the new hotness (Clojure / ClojureScript, Haskell / Elm-Lang, Scala, Erlang etc.)
- the persistence of PHP as a reliable workhorse for a lot of more pedestrian web-work
Could Ruby go the same way as Perl?
Python suffers from similar trends, of course, but unlike Ruby seems to have some other strong power-bases : as a system scripting language, as a beginners' learn-to-program language (Python is the new BASIC), in scientific and academic computing. I'm not sure if Ruby has any equivalent areas where it's the preferred choice.
Quora Answer : What language can you think of that will kill Java programming language and why?
Java won't be killed by one language.
Java is like Cobol. It will diminish very slowly. It's very unlikely that anyone is going to come along and replace a huge monolithic Java application with a huge monolithic application written in something else.
What will happen is this.
GUI applications, some of which have been written in Java, get replaced by services with an in-browser web front-end. Initially the front-end is dumb and all the intelligence sits at the back-end. But over time, the front-end becomes richer and more powerful. It does more of the front-line validation. It does more of the pre-processing of inputs into, say, an appropriately shaped chunk of JSON. It does more of the manipulation and post-processing of data (eg. turning query results into graphs or letting the user pivot and crosstab higher dimensional data-sets).
All this in-browser stuff will be written in Javascript or a lighter compile-to-javascript language like CoffeeScript, TypeScript, Elm or ClojureScript etc.
There'll also be iOS and Android apps as alternative front-ends. On Android these might be Java. But simple front-ends will be written in Javascript with app-builder type software. And on iOS it will be exclusively Swift (or Objective C).
Meanwhile monolithic back-ends start to be broken up into multiple microservices. Services that handle high throughputs of data, or require scalability or high availability might be written in Erlang or GoLang or Scala etc.
Caches and secondary databases and message-queues etc. are fairly decoupled from the actual business logic of the application. So they are relatively easy to strip out of the main Java code-base. And once again can be written in robust and high-performance functional languages.
The fragmentation of the back-end to microservices probably means more "integration at the glass" ie. the browser pulls data from multiple back-end microservices and integrates / cross-references it in the UI.
Slowly, the monolithic Java application is being boiled down to a set of classes that represent the business logic. The data and the state-transitions it undergoes. But it will become possible to write new logic to describe new processes and transitions in compile-to-JVM languages like Scala and Clojure. We'll see Scala and Clojure being used to write unit and integration tests. We'll see new (more robust) process modelling being done in these languages (even as they use the old Java Beans for core data representation.)
Meanwhile, more of the enterprise management will get automated through continuous integration and delivery tools. Some of this will be written in Java. But it might also be written in Python or Ruby (or Clojure)
Statistical and data-analysis and machine-learning algorithms may also be "outsourced" from the code-base to secondary systems written in Python / R and Julia.
Within 10 to 15 years, I'd expect to see today's large enterprise Java applications distilled down to some core data classes synced with external databases. But they'll be deeply embedded within a much wider ecosystem of integration / testing / caching / queuing / specialist microservices / front-end apps / IoT devices and sensors / statistics and data analysis services etc. which will be written in many different specialist languages.
What are some examples of good programming languages that have failed to catch on and just faded away?
Arguably Smalltalk is the language that invented the GUI and the object-oriented style of programming; both of which took over the world for 20 years or so. And yet Smalltalk mysteriously failed to become as popular as it deserved, or as the things it spawned.
Instead C++, Java, Delphi, Visual Basic, Python and Ruby all went on to become far more popular, mainly using the tricks they copied from Smalltalk.
Why it didn't go mainstream is hotly debated and probably due to a variety of reasons. But I'd say the biggest was its (perceived) refusal to have anything to do with the file-system or other operating system features, and its stand-offish isolation in its own world of virtual machine / image.
OTOH Smalltalk is, still, a cult-classic, so perhaps doesn't quite fit your criteria of fading away
Backlinks (1 items)