'5 Programming Languages That Are Probably Doomed' (dice.com) 390
An anonymous reader shares a report: Not all programming languages endure forever. In fact, even the most popular ones inevitably crumble away, as new generations of developers embrace other languages and frameworks they find easier to work with. In order to determine which programming languages are likely doomed in the medium- to long-term, we looked at the popularity rankings by TIOBE and RedMonk, as well as Dice's own database of job postings. If your career is based on any of the following languages, we suggest diversifying your skill-set at some point: Ruby, Haskell, Objective-C, R, and Perl.
ORLY (Score:5, Interesting)
Re:ORLY (Score:5, Funny)
Can we mod main articles flamebait? Just askin. Totally nothing to do with this particular one. Promise. Really.
Re:ORLY (Score:5, Funny)
Can we mod main articles flamebait? Just askin. Totally nothing to do with this particular one. Promise. Really.
I'm totally in favor of having the UN set up preserves for Perlophones, perhaps abandoned warehouses filled with reconditioned Unix workstations and tear-stained mattresses, in order that one of the world's most unique languages shall not die out.
Re: (Score:3)
Re: (Score:3)
Good thing dinosaurs rule the earth until the next asteroid comes along. I prefer my languages without a political position.
Re: (Score:3, Insightful)
There is no such language as "C/C++".
Oh really? Both Intel's compiler and the GNU compiler will compile both C and the C with object extensions known as C++. If they aren't at least almost the same language I wonder how and for that matter why they use the same codebase to compile both. I sometimes write code in a .c file and sometimes in a .cpp file. The code in each case is almost identical when objects are not needed.
In natural languages it would be sort of like saying Brazilian Portuguese is a completely different language from European Po
R.I.P. (Score:2, Funny)
As in nearly everything else, Ruby developers will be the last to find out.
Re:ORLY (Score:5, Funny)
Re: (Score:2)
OTOH, it not exactly what it once was.
IMO: if you can't write good readable code in PERL, you should find a new business to work in.
check the job description (Score:3, Funny)
Writing unreadable perl is the business.
Re: (Score:3)
Perl has been "doomed" for over 30 years now, hasn't stopped it.
I love Perl, but today it is mostly small throw-away scripts and maintaining legacy apps.
It makes little sense to use Perl for a new project.
Perl won't disappear, but the glory days are in the past.
Re:ORLY (Score:4, Interesting)
I write new code in perl all the time. Cleanly written, well formatted and completely maintainable. Simply because YOU can't write perl in such a manner, that doesn't mean others can't.
Re: ORLY (Score:2, Insightful)
Do you have someone else who is saying that about your code or is that your own opinion?
Re: ORLY (Score:4, Insightful)
I happen to read a lot of Perl in my day job, involving reverse-engineering a particular Linux-based appliance for integration purposes. I seldom come across scripts that are too actually bad.
It's important to understand that Perl has a different concept of readability. It's more like reading a book than reading a program, because there are so many ways to write any given task. A good Perl programmer will incorporate that flexibility into their style, so intent can be inferred not just from the commands used, but also how the code is arranged. For example, a large block describing a complex function would be written verbosely for detailed clarity.
A trivial statement could be used, if it resolves an edge case.
Conversely, a good Perl reader will be familiar enough with the language to understand the idioms and shorthand used, so they can understand the story as written without being distracted by the ugly bits. Once viewed from that perspective, a Perl program can condense incredible amounts of description into just a few lines, and still be as readily-understood as any decent novel.
Re: ORLY (Score:5, Insightful)
Since you brought it up...
In building several dev teams, I have never tried to hire everyone with any particular skill. I aim to have at least two people with each skill, but won't put effort to having more than that at first. After the initial startup of the team, I try to run projects in pairs, with an expert starting the project, then handing it to a junior (in that particular skill) for completion. After a few rounds of that, the junior is close enough to an expert, and somebody else takes the junior role. That way, even with turnover, expertise is shared among the team, and there's always someone who can be the expert.
Back to the subject at hand, though...
My point is that Perl is a more conversational language that others, and its structure reflects that. It is unreasonable to simply look at Perl code, see the variety of structures, and declare it "unreadable" simply because the reader doesn't understand the language.
As an analogy, consider the structural differences between Lord of the Rings and The Cat in the Hat. A reader who is only used to The Cat in the Hat would find Lord of the Rings to be ridiculously complex to the point of being unreadable, when Lord of the Rings is simply making use of structures and capabilities that are not permitted in the language of young children's' books.
This is not to say that other languages are wrong to have a more limited grammar. They are simply different, and learning to read a more flexible language is a skill to be developed like any other. Similar effort must be spent to learn other languages with sufficiently-different structure, like Lisp or Haskell.
Re: (Score:3)
Re: (Score:2)
Perl doomed? <sobs softly>
<sigh> Been using Python lately anyway.
Re: (Score:2)
Perl has been "doomed" for over 30 years now, hasn't stopped it.
I'm on the edge of my seat waiting to see whether Perl 6 releases, Perl dies, or the next Duke Nukem ships.
Re: (Score:3)
Most likely I'd use NodeJS. It isn't that JS is a good language (it's a terrible language), but it's an easier language, ubquitious, readable, well supported by IDEs, well understood by developers, has an enormous ecosystem of packages, is internet savvy and works as well
Re:ORLY (Score:5, Insightful)
Here is what it says:
Well, what's your evidence-based rebuttal to that?
Re: (Score:2)
If you actually look at the TIOBE rankings, it's #11 (not #12 as claimed in the article), and back on the upswing. If you look at RedMonk, which they say they looked at but don't reference with respect to Ruby, it is a respectable #8, being one of the top languages on GitHub and Stack Overflow.
We are certainly past the glory days of Ruby, when it was the Hot New Thing and everyone was deploying Rails, but to suggest that it is "probably doomed" seems a somewhat hysterical prediction.
Re:ORLY (Score:4, Funny)
How do they know how many Ruby jobs there are? Maybe how many Ruby job openings announced, but not the actual number of jobs. Or maybe they are finding Ruby job-applicants and openings via other means.
Maybe there is a secret list of Ruby job postings only available to the coolest programmers? Man! I never get to hang out with the cool kids.
Re:ORLY (Score:5, Insightful)
Perhaps the devops/web programmers is a dying field.
But to be fair, Ruby had its peak about 10 years ago. With Ruby on Rails. However the problem is the "Rails" started to get very dated. And Python and Node.JS had taken its place.
Re: (Score:2)
Honestly, while RoR gets most of the mind share for Ruby, I imagine the larger use of Ruby is in server management tools, where it replaced Perl for a certain subset of the industry.
Re: (Score:2)
You can't have Puppet without it (unfortunately)
Re: (Score:2)
However the problem is the "Rails" started to get very dated. And Python and Node.JS had taken its place.
The Rails ecosystem has kept up to date. The big complaint was that Rails is a ghetto. That might be true, but Node.js is a bunch of mud huts. I spent the last week using Node.js Sequelize. It's an ORM, similar to Rails, but what a piece of garbage it is in comparison.
Re: (Score:3)
I agree. I maintain large Rails and Node projects. The nodeproject is definitely without any databases. Running
a database with Node is an excercise in pain
Rails's main advantage is ActiveRecord. That thing is Magic Squared, and it enables a killer app I wrote ... not good, but React/Angular/Vue has
Rails's main disadvantage is that is is slow. And the web template engine is
probably that moot by now.
I am looking at Rails 6 now, it certainly has kept up.
NodeJS? FFS, the language cannot even do things with conc
Re: (Score:2, Insightful)
Not really, DevOps is phasing out Ruby in favor of Python only. Ansible ignores Ruby altogether.
In web programming, people have gone back to Java/C# for job reliability or newer stacks like Elixir.
Ruby is dying, it won't be dead for a long while but the amount of Ruby specific jobs are falling far behind other languages.
Re:ORLY (Score:5, Insightful)
I don't see Ruby dying anytime soon, but I do get the feeling that Python is the go-to scripting language for all the things now. I learned Ruby and wish I spent that time learning Python.
Perl is perl. It will live on, but anybody writing new things with it probably needs to have a talkin' to.
Re:ORLY (Score:4, Insightful)
I learned Ruby and wish I spent that time learning Python.
Ruby and Python are basically the same thing. With a little google, you can literally start programming in Python today. Search for "print python" and you can easily write a hello world. search for 'python for loop' and suddenly you can do repetitious tasks. Search for "define function python" and you can organize your code.
After that do a search for hash tables and lists in Python and you'll be good enough to pass a coding interview in the language.
Re: (Score:3)
> What language can you not easily write 'hello world' in?
From the Preface of How to Think Like a Computer Scientist [openbookproject.net]:
"The first example from the text illustrates this point. It is the traditional hello, world program, which in the Java version of the book looks like this:
"in the Python version it becomes:
"Even though this is a trivial example,
Re: (Score:3)
Is it really? DevOps is mostly moving to Kubernetes or Docker or other YAML-declarative-ish approaches. DevOps using Vagrant, Chef, and Puppet (and thus Ruby) is a bit dated.
Re: (Score:2)
The statics in TFA tend to disagree dropped from the top 10 to #12 and saw 56% decline in job posts asking for experience in that language over an 18 month period shows a popular language in decline. Doesn't mean it's dead yet but they are betting the decline will continue.
Re:ORLY (Score:5, Funny)
Ruby almost killed Twitter before it even got started.
Damn you, Ruby, couldn't you have tried just a little harder?
Re: (Score:2)
Devops only use Ruby derived DSLs
https://www.intigua.com/blog/p... [intigua.com]
No one actually likes it .. or uses it for anything he is not required to use it for.
Hint: devops means you know linux/unix and shell, firewalls, ip, routing etc. Not Ruby.
Re: (Score:2)
My company is developing most of it's code in Ruby which is suitable for the task and we find it to be expressive and clean.
We don't care if it is the greatest language of the world, we don't follow the last trend and we don't lose time refactoring for the latest new trend.
On the contrary, the farther the crowd of trendy "developers" gets from it, the better.
Python is scalable? (Score:2)
Maybe compared to R, and I've heard that's a tremendous problem with Ruby, but in general, is Python scalable, if you aren't handing off the hard stuff to complied code below it?
Re: (Score:2)
Like most things, it depends. Its being used pretty heavily in machine learning now but what the scale of that is I don't know as it's not my field. Python is a odd language in the way people use it. Reminds me a lot of the old Perl days when people just rammed it into every use case they could come up with.
Re:Python is scalable? (Score:5, Informative)
The thing is, the idea of Python is (was?) that the compute heavy sections should be written in C, with "Python versions" used for debugging purposes. It's actually quite a good idea, and lead to the Python "batteries included" libraries. But it seems to have lost a bunch of support with the transition to Python3. It's still easier to do that in Python than in most languages, but it really needs a lot more attention.
(See Pyrex, Cython, etc.)
Re:Python is scalable? (Score:5, Insightful)
FORTRAN on big iron doe the best analytics, hands done.
But it take actual skill, discipline and understanding.
I frequently run into programmer who don't actually understand how code works or what it is doing.
Can we go back to needing a math degree to program? because that would be great, thanks.
Re:Python is scalable? Math degree? (Score:3, Interesting)
I have a math degree, and I was a computer programmer for many years, but I never thought I needed the math degree to program. Now if you're a Fortran Programmer (and Fortran was the 1st computer language that I learned) in this day and age, then the applications it's used for are heavy on the math.
If you want to say that logic, and Boolean algegra, and things like state machines are part of math, then maybe that's something you should know. But that stuff is pretty simple math actually.
If you are restric
Re: (Score:3)
When you're talking about analyzing a massive dataset via a cluster of computers, the communication between the nodes is the usually the limiting factor on your speed. And that is much slower than the difference between byte-compiled and really-compiled code.
IOW, C and Python wait for the network at the same speed.
Re: (Score:2)
Please tell that to those maniacs living over in microservicestan.
Re: (Score:2)
If you're using a high level language and not handing off the hard stuff to a lower level language, you're making a mistake, no matter what scale.
In Python you hand off to Numpy/Scipy/Tensorflow. In R everything hard is done by C modules. I don't know much about Ruby, but I assume it's the same.
Re: (Score:2)
Rumors of 's demise... (Score:5, Insightful)
I'll give them Obj-C since Apple was the only one really using it and they've now moved on.
I've never followed Ruby (outside of the cross over for writing Puppet and Chef modules) as it never offered me anything significant over languages I was already using, so I can't speak to it's life expectancy.
Haskell and R are too useful for the areas they were designed for, so I highly doubt they are going anywhere. Just maybe they won't be used for things where they aren't the optimal language.
People have been predicting Perl's demise for over 20 years, but it's still kicking. While Python has definitely eaten away at Perl's kingdom, Perl is still kicking.
Even though I think their specific conclusions are incorrect, the suggestion they offer is sound and how I've made a solid and stable career. Don't tie yourself to any one language. Be able to work in anything so you aren't arbitrarily limited in either job selection or picking the best tool for the job.
Not even Obj-C (Score:4, Informative)
I'll give them Obj-C since Apple was the only one really using it and they've now moved on.
On the face of things Apple is pushing Swift, sure.
But Apple has been careful to make sure that Obj-C has kept fairly current with language updates to help bridging to Swift. So the language is still very usable, and it's very easy to mix Swift and Obj-C.
On top of that, Apple still has a TON of internal stuff all built on Objective C, that will take decades (or longer) to unwind from. There's no reason to though, and that alone would keep Obj-C alive if nothing else, for those that care to use it... it's not like Xcode will stop supporting Obj-C anytime soon, so how can you really say the language is dead? It will be really easy for a long time to find work in Obj-C if you wish.
Big Difference (Score:2)
Just because some mainframes from the 70's are still around and require programming by punch cards does not mean that programming via punch cards is not dead as hell.
Such cards are dead, yes. Fortran or Cobol? Not so much.
You confuse language with tooling. Languages can last beyond the current means of development.
It would not hurt though (Score:2)
All true, but I wouldn't suggest starting a new project in Obj-C for iOS these days
Apple would certainly agree, but there are still a lot of people very much into new Obj-C development.
in the short run there will start to be new features that are only available in swift
You can already see that with Swift UI - but even there you can have Swift UI interact with traditional Obj-C based UIViews. Apple has made it super easy to interact with other Swift components, so for a long time you'll be able to mix and m
Re: (Score:2)
"Haskell and R are too useful for the areas they were designed for,"
So are FORTRAN and COBOL, but those areas aren't exactly growing.
Demise in this industry doesn't mean gone. It means low usage, adoption, and shit upon by ignorant "Computer engineers"
Re: (Score:3)
Re: (Score:2)
the new crop of programmers don't want to touch a mainframe with a barge pole
The pay is kind of low. At least, the COBOL programmers I know don't make nearly enough for me to be willing to switch to that (otherwise it would be fun).
Re: (Score:3, Interesting)
Haskell is incredibly useful for exploring type theory and the relationship between category theory, the typed lambda calculus, and computation. I don't see it getting much more than limited use in production environments: in the hands of smart people it can yield incredibly structured, maintainable, extensible, high-performance applications, but it has a high barrier to entry. It's like LISP or Clojure or APL-derived languages that way.
Re:Rumors of 's demise... (Score:5, Informative)
Haskell is a test lab for the next generation of programming languages. You've probably never used Haskell for anything commercial, but I bet you have used a language feature that came from Haskell, and in ten years, you will be using a language feature that is in Haskell today.
Haskell's unofficial motto is "avoid success at all costs". To claim that Haskell is doomed is to miss its reason for existing in the first place.
Comment removed (Score:5, Insightful)
Re: (Score:2)
Lots of analysts/quants/statisticians work with R exclusively.
Re: (Score:3)
The thing is, R has never been a mainstream language as such; it's always been for those few people that really need it.
It will survive because those few people really need it.
Wrong, Haskell does not run on the JVM (Score:2, Informative)
No it does not run on the JVM. You may be thinking of another well-known Functional Programming (FP) language, Clojure.
But Haskell is unquestionably the world's leading language for research in FP, and it also happens to be the fastest one, performing within 17% of C speed in applications suited to FP algorithms, and it's also the best supported with libraries and tooling. Its compiler GHC is widely recognized as world-class.
So no, Haskell is doing the opposite of dying, its use is
Re: (Score:2, Informative)
You're thinking of Scala. Haskell doesn't use the JVM in any way at all.
If your career is based on ... (Score:3, Interesting)
Perl: Even if RedMonk has Perl’s popularity declining, it’s still going to take a long time for the language to flatten out completely, given the sheer number of legacy websites that still feature its code. Nonetheless, a lack of active development, and widespread developer embrace of other languages for things like building websites, means that Perl is going to just fall into increasing disuse.
First, Perl is used for many, many more things than websites -- and the focus in TFA is short-sighted. Second, I've written a LOT of Perl in my many years, but wouldn't say my (or most people's) career is based on it. Yes, I have written applications in Perl, but more often used it for utility, glue and other things that help get things done, monitor and (re)process data. Nothing (or very few things) can beat Perl for a quick knock-off script to do something or another.
Perl's not going anywhere and it will be a useful language to know for quite a while. Languages like Perl (and Python) are great tools to have in your toolbox, ones that you know how to wield well when you need them. Knowing when you need them, and not something else, is important.
Perl? No. (Score:4, Insightful)
I don't know what will become of Perl 6, but Perl 5 will continue to exist as long as varieties of Unix do.
Lol, how little you know (Score:4, Informative)
R is very popular, in fact it wiped out S and other pretenders.
But it's for statistics. If you don't need that, you don't code in it.
Re: (Score:2)
Re: (Score:2, Insightful)
A lot harder finding stable Python code, actually
Re: (Score:2)
R is an implementation of S Plus that took over because it was free. Lots of those biostatistics people were using R for things other than stats, and there seemed to be a burgeoning "data science" market for it, but they seem to be realizing that R is a cranky old language primarily useful for it's large array of academic contributed stats libraries.
Re: (Score:2)
Yes, so you admit that it dominates.
Re: (Score:2)
Sure. Nobody uses S anymore. R is the pretender though.
Re: (Score:2)
Look, sour grapes over flavor choices by individuals never looks good.
R won. S lost. It's why blade servers run on Linux and not on Unix.
Not everyone needs everything some people want. Artificial measures like measuring language success by total market share, which cause high priced languages to have a distorted share compared to low cost languages, or measuring per seat when most open source language implementations are rolled out enterprise wide from 2-3 masters with 1-2 language support licenses, might
Re: (Score:2)
I'm not sure how any of that is relevant.
I don't care at all about S. R is free and, despite being a terrible language, is quite useful. It doesn't really have anything going for it besides the libraries and user base though.
Re: (Score:2)
Exactly this. R is a terrible language, but good luck finding high quality replacements for its libraries.
Re: (Score:2)
Well... some effort is being made. StatsModels does a decent job of implementing GLMs in Python for example. There's also RPy2, which spares you much of the pain of R.
It's unfortunate that the R language model is so crufty. It would be a lot of work to independently wrap the libraries.
Confirmation required (Score:2)
R is going away?!? (Score:5, Interesting)
Please, give me a break. Someone knows NOTHING about scientific computing, eh?
R is used for pretty much ALL scientific stats and number crunching.
The only way R is going ANYWHERE, is if you nuke all the Universities and research labs.
Barring that, R's future is pretty damn secure.
Re: (Score:3)
Well said. R is on the same exact trajectory that Linux was. It's equivalent to saying that "Linux is doomed." Also witness the proliferation of new books written about it. It will be academia's go-to for statistical analysis no matter what python becomes.
The thing that propels R is that it is the only open source (GPL) alternative to proprietary, closed source. and expensive software packages. It can link to practically anything and recently got a JIT compiling.
R's future is quite secure.
Re: (Score:3)
Re: (Score:2)
R is picking up speed in academia (Score:2)
R is going nowhere. I don't give a fuck how popular it is with contractors and generic SW developers. At the moment, most experimental researchers are learning the basics of R. We are talking about a gigantic number of people, here.
DIE Perl, DIE (Score:3)
DIE Perl, DIE
It means, "The Perl, the" in German.
But I still want perl to die.
It's uncomfortably terse (needlessly terse, IMHO) and it's easy to write code that is damn near indecipherable without even trying. Yes, you can do that in C/C++ and PHP and Ruby and Python, etc etc etc, but it's hard to do it by accident. With perl it seems to come naturally as part of the language.
But perl is cool because you can open a file in 763 different ways, so it's got that going for it...
Re: (Score:2)
Perl 5.x still a core part of system utilities in a lot of BSD, Unix(TM) and Linux, it'll be around more than a decade from now, and likely 20+ years.. Funny even some enterprise management and backup wares from HP have Perl in them.
Used at Microsoft, too (Score:3)
It also (at least as of a few years ago) is/was a core part of the Windows build toolchain. Microsoft was so reliant on Perl in the NT development space that they ported it to Windows on ARM (back in the pre-Windows RT and Windows on IOT days) rather than just relying on cross-compiled tools and tests run through the debugger.
No idea if it's still there, though. I don't actually know if anybody was currently writing any new Perl code, either... but a working Perl installation was a vital part of setting up
Re: (Score:2)
I ran into this problem just this weekend. I have a couple old scripts that do a little resource fetching, processing, and then build a simple report. They've been running on my server since I wrote them about 15 years ago.
One of the URLs they got changed so I had to update them. I have never been much for writing High Magic code as I don't want to outsmart my future self. But fuck me were those scripts hard to read. There were a bunch of character patterns I was sure were line noise or file corruption but
they're all doomed (Score:2)
Career based on *a* programming language? (Score:4, Insightful)
Re:Career based on *a* programming language? (Score:4)
Anybody whose career is based on a single programming language is doomed already. Programmers know how to write code. The language they use is beside the point. A good programmer can write code in whatever language is asked of them.
A good programmer can write <programmer's favorite language> in any language asked of them.
Re: (Score:2)
that said I know people who went for decades with COBOL, and also some Java EE devs nearing retirement age. Some languages may not be cool or great but they have ability to endure.
That is the dumbest sentence ever said to me (Score:3)
The language they use is beside the point. A good programmer can write code in whatever language is asked of them.
That is horseshit and dangerous. Name one example. I've never seen a programmer that could code professional-grade code in several languages. I've seen a bunch of arrogant amateurs who know remedial programming in many languages. I've never seen a full stack programmer that could code as well as a professional backend programmer, front-end programmer, and DBA. I have seen some overconfident ones who thought they could and turned in amateurish garbage that was rewritten once it was handed off to actual
Who are they? (Score:2)
And this is somehow the definitive source to everything related to whether a language is doing well or not. I hate to break it to you but companies looking for COBOL developers will probably not bothering with the above mentioned people / sites / whatever. So this is a skewed perception at best.
Re: (Score:2)
TIOBE counts the number of search engine queries for a language, LOLZ.
Yes, every time a person is hired for development of a language or ten lines of code are written, the person engaged in such activity does a web search on the language.
That makes as much sense as counting web queries for food to see what is eaten the most.
Python is overrated (Score:3)
I had this naive idea that Python might substantially displace R until I learned more about the Python internals, which are pretty nasty. This is the new generation's big data language? If true, sure sucks to be young again.
Python isn't even really used to do big data. It's mainly used to orchestrate big data flows on top of other libraries or facilities. It has more or less become the lingua franca of high-level hand waving. Any real grunt is far below.
R commits a substantial scale crime by being so dependent on memory-resident objects. Python commits major scale crime with its single-threaded primary interpreter loop.
If I move away from R, it will definitely be Julia for any real work (as Julia matures, if it matures well), and not Python.
Five Useless Job Sites that are Doomed (Score:2)
If you want to open yourself up to 3rd party job seekers (they saw an opening, so they want to become middle man, never mind 2/3 the time they're listing another middle man), put yourself on Dice.
If you want to be bothered by 3rd party head hunters working out of a hut in a 3rd world toilet, put your job on Dice.
If you want to get bad and incorrect advice on what languages are in decline, e.g. R, listen to article on Dice.
If you want to know popularity of language for employment and amount of use, do NOT li
Re: (Score:2)
Banks still use mainframes from the 80s? Name one. While virtually all large banks use mainframes, the majority of those systems are less than 5 years old.
Re: (Score:2)
Re:Diversifying (Score:5, Insightful)
The writer of this article should consider diversifying his skillset at some point, as not all bloggers endure forever and his popularity ranking on Slashdot has recently tanked.
I'd suggest that this writer quit their day job and take up stand up...
Old languages never really die until the platform dies. Languages may fall out of favor, but they don't usually die until the platform they are running on disappears and then the people who used them die. So, FORTRAN, C, C++, and COBOL and more are here to pretty much stay.
Specifically, PERL isn't going anywhere being fundamentally on Linux, neither is Ruby, the rest to varying degrees have been out of favor for awhile now, but none of the languages in the article are dead. They are, however, falling out of favor and because of that, it might be a good idea to be adding other tools to your programmer's tool box if your livelihood depends on one of them.
Re: (Score:2)
They are, however, falling out of favor and because of that, it might be a good idea to be adding other tools to your programmer's tool box if your livelihood depends on one of them.
Got it. Not dying just "falling out of favor" and so the conclusions of TFA are absolutely on the money.
Re: (Score:2)
Re:Not seem to be "doomed" (Score:5, Funny)
Perl? What about awk?
Clutching at their single perl... Somebody sed a word and It was an awk word moment...
Re: (Score:2)
Re: (Score:2)
I'm a huge fan of awk.
I really like it. Taught some guys how to use it. Since then they export their excel sheets and awk them. One made a Phd mostly based on simple scripts she wrote herself calculating over Excel sheets (agriculture and life stock breading research, 20 years ago).
Re: (Score:2)
Yes, and such articles often conflate Ruby with Ruby on Rails
We use Ruby (and a couple other langauges) where I work but no Rails, that stuff is not going away in the next decade since it moves money for banking and taxes. I'd bet the code is around in 15+ years.
Re: Perl ain't going away (Score:2)
The vast majority of production perl code is managed by people who aren't perl devs, though. Very little new stuff getting written, and the old stuff doesn't need a perl dev to work on it.