Slashdot Mirror
Programming Languages You'll Need Next Year (and Beyond)
Nerval's Lobster writes: Over at Dice, there's a breakdown of the programming languages that could prove most popular over the next year or two, including Apple's Swift, JavaScript, CSS3, and PHP. But perhaps the most interesting entry on the list is Erlang, an older language invented in 1986 by engineers at Ericsson. It was originally intended to be used specifically for telecommunications needs, but has since evolved into a general-purpose language, and found a home in cloud-based, high-performance computing when concurrency is needed. "There aren't a lot of Erlang jobs out there," writes developer Jeff Cogswell. "However, if you do master it (and I mean master it, not just learn a bit about it), then you'll probably land a really good job. That's the trade-off: You'll have to devote a lot of energy into it. But if you do, the payoffs could be high." And while the rest of the featured languages are no-brainers with regard to popularity, it's an open question how long it might take Swift to become popular, given how hard Apple will push it as the language for developing on iOS.
Over at Dice
But we are at Dice, sir:
Pros: Today's article has more content than the usual Dice front page linkage. Great article if you're not a programmer but feel stymied by the wide assortment of languages out there. Although instead of hemming and hawing before making your first project you're better off listening to Winston Churchill and sticking your feet in the mud: "The maxim 'Nothing avails but perfection' may be spelt shorter -- 'Paralysis."
...
Cons: It barely scratches the surface of an incredibly deep topic with unlimited facets. And when one is considering investing potential technical debt into a technology, this probably wouldn't even suffice as an introduction let alone table of contents. Words spent on anecdotes ("In 2004, a coworker of mine referred to it as a 'toy language.'" like, lol no way bro!) could have been better spent on things like Lambdas in Java 8. Most interesting on the list is Erlang? Seems to be more of a random addition that could just as easily been Scala, Ruby, Groovy, Clojure, Dart -- whatever the cool hip thing it is we're playing with today but doesn't seem to quite pan out on a massive scale
My work here is dung.
CSS3 is not a programming language. No more then HTML is.
Visit the Arcade Restoration Workshop @ http://www.arcaderestoration.com
Not even a mention of VB
Kill me now.
I am surprised that Scala isn't mentioned.
It is strongly typed, object-functional and compatible with java.
Swift syntax is basically a cut and paste from Scala, which benefits from being more mature (and having access to all the Java libraries)
Scala is also much faster than erlang, while also supporting the actor based model.
http://www.scala-lang.org/
C#, Java, PHP and JS will continue to dominate, C# and Java being mutually exclusive in a web app. Swift? Maybe. It's a shot in the dark. Apple is apple but the iOS development market is dried up for most developers, unless you can get a job at some company that wants an iOS app. It really depends on how well the Apple/IBM partnership works out in dislodging Microsoft from business apps. It's a long uphill battle that I suspect will take a lot longer than 2 years.
Erlang? Fuck off, Dice. How much did the one company that wants it pay you to write this garbage article?
I am surprised that Scala isn't mentioned.
It is strongly typed, object-functional and compatible with java.
Swift syntax is basically a cut and paste from Scala, which benefits from being more mature (and having access to all the Java libraries)
Scala is also much faster than erlang, while also supporting the actor based model.
http://www.scala-lang.org/
Learn C++, Java or C# and get yourself a job at a big corporate.
But hey, if you want to be a hipster coder and dick about all day doing "groovy" websites at some here today gone tommorow startup and earning fuck all by all means go down the web development route along with every other 14 year old school kid.
Erlang? Nice language but too niche. Never really got momentum outside telecoms and its probably too late for it now.
Who cares about a fucking language. I know more that 20 different programming languages, I don't care about one more or less. What can it do, that I can't do already ? Can I program faster, or better? Or is it just an other syntax, for some obscure system ?
How do we even know it's going to be popular in the first place? Does it solve any problem I can't do with C# or Python and/or on more platforms? It'll be a language for little hipsters who hope to be the next Steve Jobs by releasing yet another crappy useless iOS app. I don't know anyone who still bothers with iOS apps.
What's the point, dice? Further proof that recruiters and by extension recruiter-infested jobsites and everything around them are clueless nitwits? As if we didn't already know that.
Note that the jobs on dice outside of the US typically require US government clearance, so the thing typically is entirely useless to the 95% of the world that isn't the US -- ./ is dice owned and (a shade too) US-centric, true, but has considerable non-US readership. To me dice are worse than useless in fact, since it presumably is their nitwit management that keeps on flogging that dead horse "beta".
Just from what I've seen (not a user yet) Swift seems like an intelligently put together language. But I'd like it a lot better if somebody ported it so it can be useful across multiple platforms. I'm a C# developer who uses Xamarin's platforms, and compiling my code for all the major mobile and desktop platforms is a really good feeling. (Paying Xamarin's licensing fees isn't, but you know how that goes.)
Please, no more Erlang world domination news.
I went through that 3 years ago already. We had a project that a fanatic asked us to rewrite in Erlang.
it took 9 months with 2.5 people.
Tons of issues, mostly with very lacking library support, tooling. Obnoxious stuck up community too.
In one case, I had a guy tell me online "hire me as an Erlang consultant and then I will help you".
In the end we set screw it (once the Erlang fanatic left).
We rewrote this 9 months of Erlang development in 3 weeks (!) using one senior Java developer.
it worked like a charm and still runs flawlessly in production today.
Erlang = HYPE
Everything is immutable is beautiful for fairy tales, but not for real-life software (trying building a DOM in a language which is 100% immutable).
All modern languages have learned from Erlang's mistake. They do immutable by default, but allow mutable if there is a need for it (e.g. Ceylon, Rust, etc)
Next year, the languages you'll need will still be C, C++ and Java. Maybe some C#, Python or Bash. The year after that, you'll still be using C, C++, and Java. Maybe some C#, Python or Bash.
By 2020, the main difference is that you'll be working with machine-learning DSLs and libraries to program/train memristor based devices. But you'll still be using C, C++, and Java. Maybe some C#, Python or Bash.
When you write code and declare a variable, dynamic languages let you change the type of data held by the variable when the program is running; those languages that don’t are known as “static” or “strongly typed.” Languages such as C++ and Java are strongly-typed languages, while JavaScript, PHP, and Perl are dynamic languages.
"Staticness" and "strongness" are orthogonal properties. Python, for instance, has strongly typed values (you can't convince an int that it's a str), but dynamic variables (a=123;a='foo' is valid). And while C++ is statically typed, I'd be hard pressed to describe something with void* and unions as strongly typed.
TL;DR: Words have meaning. It's OK to disagree about whether a particular language is strongly or weakly typed, but it's not OK to claim that two different concepts are the same thing. When you make a fundamental mistake in the third paragraph, I'm likely to ignore anything else you have to say in the rest of the article.
Dewey, what part of this looks like authorities should be involved?
Or how about learn about the fundamentals of computer science. Actually learn what pointers are, pass by reference, multi-threading, type safety, and all of the things that implies. Then express those in whatever language you want. If you truly understand how computers and languages work, and what an enterprise system is composed of, you will likely have future proofed your career. If your language doesn't support many of those ( I am looking at you, JavaScript), then perhaps consider how much those jobs are likely to pay in the long run....
It's been a while since I've had a language paradigm course but strong/weak is not synonymous with static/dynamic. Thoughts?
Swift is a strongly-typed language, that operates inside a runtime with garbage collection.
There is no GC in iOS. Also the GC in OS X is deprecated. Swift uses Automatic Reference Counting which is something... completely different.
an example of dynamic and loosly types, TCL.
There is only one type: string. However internally it also knows about numbers and objects, but it converts those to strings willy nilly, and interprets string to those numbers and objects.
TCL should not have existed, but companies that make hardware design tools fucking love it. However everyone in the industry that works with these tools have moved on to Python. But this is an industry that still doesn't have a working simulator and synthesiser for VHDL2008.
date_default_timezone_set(‘America/Los_Angeles’);
Since when is ‘America/Los_Angeles’ a time zone?
When our name is on the back of your car, we're behind you all the way!
All I need is PIC assembly, DOS batch files and AWK.
Mostly random stuff.
... hey dice, this sort of cross posting between online entities is lame and irritating. Oh well, the time where I can use slashdot is diminishing as the unusable and unwated beta looms overhead.
P.S. Don't use Dice.com they'll SPAM the shit out of you forever then deny it.
Holy fuck that article was bad. Slashdot, what have you become?
Half the people reading this are going to need to learn PowerShell in the next year.
Next year, unless you are at the bottom of the skill and payment pyramid, you will need C and derivatives (C++, Objective C) and maybe Python or Perl.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
CSS/JavaScript/HTML5 is plainly obvious. Everything from Microsoft to mobile hybrid development relies on this these days.
C# is the standard language of the Microsoft stack --- in fact, the bulk of MS-stack training is in C#, with only a smattering in VB.NET.
Java is the COBOL of the early 21st Century. It isn't sexy anymore but it will always be around.
PHP is used in a lot of web applications. I wish it weren't. In fact, I'd really rather see Ruby on Rails take over this space.
If you're going to program native code, you could learn Swift, sure. You could also learn Rust (Mozilla's systems-level language with significant buy-in from Samsung) for device programming. If your goal is to write native apps, your best bet for Android is actually Java. By the way, one can also design native apps in Java (the code is Swing-like) and compile them to native apps for iOS or Android using Codename One, and I imagine a few shops will pick up that practice.
I like Erlang as an honorable mention. I'd also add two others: Python (especially for data analysis) and PowerShell (which will set the grown-up Microsoft sysadmins from the point-and-click kids).
Finding God in a Dog
C. Plain old C.
Entire Operating Systems are written in it. Userland tools for those operating systems are usually written in it. Any self-respecting developer knows at least C. The rest is just like fashion tips: next year they're outdated.
Although, as much as I hate to admit it, the same could be said for Java...
And on the Eighth Day, Man created God.
Learn Glue. As the once great Charles Barkley said "anything else would be uncivilized"
This reads like it was written by a student in a Writing 101 class who chose to research programming languages as a weekly assignment.
Hell yeah, I played that at lan parties back when people used to lug desktops around every weekend just to play games.
Languages don't matter of, in the age of google, the concept of "learning a language" is obsolete, dont learn language, learn good programming logic, and understand the different types of languages in general, everything else you need to be able to adapt to and learn on the fly. don't bother sitting there and trying to learn code and memorize crap, you're wasting your time, just understand programming and understand how to learn code quickly and you will be 1 step ahead of everyone else who waste's money learning 1 or 2 languages.
When you are successful at doing this, languages mean nothing to you, they are all the same (with a few exceptions), some have advantages over others in certain fields. Just understand that syntax is not meant to be memorized anymore, its meant to be looked up and used.
Back in the day when there were only a handful of languages and the only references were books, it was worth memorizing, these days where the number of useful languages that are used everyday are by the dozens, you need to forget the concept of learning a language.
...when you can use Qt and compile the same code frickin' _everywhere_?
PHP is forecast to be very popular going forward. That means your employment prospects are good!
#DeleteChrome
With Flash you can make: Destop/Web/Android phone+tablet/ios phone+tablet all in the same code!
Flash is a lot like C/C++ and Java, except it allows you to a couple things easier.
I think of all the languages though, Java has to be the winner. It has about the power of C/C++, except you don't need to dot the is and cross the ts. Java is superior to C/C++ in strings, garbage collection and arrays. For most projects you'd take code that has less bugs in it and develops faster than code that executes a slight % faster.
God spoke to me
The only time I've seen Erlang was back when I used eJabberd (which at the time was better than the regular jabberd).
Depending on one's industry, I'd say that GLSL would also be a useful (and interesting) thing to learn.
I spent over two years working every day with Erlang on a project, and I still don't consider myself to be anywhere near an "expert" at the language. It's just too weird and special case for a lot of the functionality I was trying to code, so while certain tasks were easier than they would have been in Java or another procedural-object language, others were damned near impossible and took obscene amounts of time to get working at all -- never mind working efficiently.
Personally I'd avoid it like the plague unless you have some special case need for it's features. Even with regards to concurrency, it's not really any better than any other language's concurrency features. They aren't really baked into the language as the summary suggests, but provided by frameworks in the API libraries, much as they are by other languages.
The main difference with Erlang concurrency is that the concurrent models are the "normal" way to program Erlang, so you're likely to find a lot of good examples of how to do it. I've found the documentation for other language's concurrency features to be somewhat limited in comparison, and less "real world" in their examples.
The main thing that I found neat about the Erlang framework was the ability to specify auto-restarts of failed threads. It takes all of about 4 lines of configuration to get a thread to be persistent/self-starting. That's the densest code I've ever seen for achieving such a task.
The big downside to Erlang is that it's almost as bad as LISP -- everything is a list. Even "structures" are just lists of objects with tags that identify the list indices for accessing the members. Be prepared for a nightmare of tail recursion if you get into this field of programming.
That said, it can be a fun and entertaining language to work with. For the things it is good at, it can be a joy to use. Much as with any language.
I do not fail; I succeed at finding out what does not work.
I wish it were Pascal. Or at least some decent extensions, and some *real* macros, updating C. And maybe a little learning from history: I could do things in IBM Assembler macros in the 1970s that you still can't do in C++ templates. We spent too much time being cool and iconoclastic and "new" that we threw out the good with the bad - except C, which has hung around forever and you STILL don't know how big your values are.
As long as we are hyping Erlang, the Erlang community is getting some disruption from a language developed by a prominent Rubyist called Elixir. Clojure-inspired metaprogramming, a Ruby-ish syntax, and it all compiles down to the same VM code that Erlang compiles into.
C# does not run in a virtual machine. It is just-in-time compiled to native code. The C# compiler compiles to IL (intermediate language) that at run time is compiled to native code. Depending on what you are doing, multiple runs may have the native code cached.
One great thing about C is that if you don't like, you can use it to program a compiler for a new language that you prefer. (I haven't done that yet but I have written a parser/interpreter in it, as well as written meta-code in C -- i.e., C code that writes my C code for me. C gets a rep as an old fogey language, but it's quite fun really.)
That sounds nice.
If your code doesn't use any strings.
A functional language is one whereby the functions themselves can be stored in variables and passed around as parameters to other languages.
What in the actual fuck. That may be the worst definition of a functional language I've ever heard. Even if I try to interpret it as something that could make any sort of sense, I just get that storing functions in variables makes a language functional, which the author goes on to debunk by pointing out that C++ isn't a functional language. Why bother even trying to describe them if you have no idea what the hell they are?
Yeah. We need to grow he Cyber War Domain.
NOT !
Entire operating systems are written in C -- as they should be.
But C is a low level language. Not the best tool for writing applications.
Higher level languages and managed runtime systems have gained so much traction for a reason. They are very productive to use. They protect you from simple mistakes. The relieve the burden of memory management. GC simplifies library APIs by making the question of who should dispose of what become irrelevant. We could still be programming in assembly language instead of C. Why aren't we? Why aren't OSes written in assembly? Because C is more productive and higher level. Similarly, there are higher level languages than C, and they have their place. C is not the end all of abstraction.
I'll see your senator, and I'll raise you two judges.
Maybe you grow up, find a macro processor of YOUR choosing and do what you want with C++ ? There is no law forbidding you to use your OWN generic programming technique in your C++ code.
Needing to learn a projects pet Marcos is why I don't bother with C projects.
it's a semantic argument, IMHO...but you totally missed the joke
if the defining criteria is your code language can do something another cannot, then have C++ render a hypertext document with linked photos & text into a browser
still it's *semantics*...it's all machines following code pounded by a monkey
Thank you Dave Raggett
Thank you for helping to reinforce our Android-first meme.
We are still a long way off, but with your continued effort we can shape the thoughts of developers to our advantage.
$0.10 has been deposited into your Google Wallet.
Thank you,
Google.
we can get pedantic about the difference between "coding" and "programming" languages...but others have gone down that road & it's never-ending
we need a consistent paradigm
HTML & CSS are coding languages...they are a defined set of symbols that contain commands run on a computer (specifically by a browser)...yes they are not "original gangster" coding languages that you can brag about..but they are a unique set of code that commands a machine...it's "code"
not all people who use HTML/CSS are "coders"
it's like back before facebook.com came online, one of the only options was Myspace...kids would get into the HTML & CSS via an interface and just randomly change color numbers and hack it up
**that's not coding**
whoever made the CSS for a site like...well facebook.com....is surely a coder
keep your bragging rights, bro...you're more of a 'real' coder...but it's reductive and pedantic to say "CSS is not a programming language"
Thank you Dave Raggett
http://en.wikipedia.org/wiki/E...
it's a relevant language to the conversation
Thank you Dave Raggett
Erlang's becoming at least slightly trendy because it's used in several sets of Cloud Stuff, and Cloud Stuff is heavily enough management-buzzwordy that HR departments have figured out they need to hire some Erlang programmers.
It's especially useful for some of the orchestration tools out there, and it's useful if your management likes Cloud Stuff Buzzwords that don't start with "v" or "V".
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
Especially when "Pacific Time" may mean different things in different countries on different sides of the Pacific Ocean.
So would free Swift rely on free Cocoa?
Clearly if someone considers that CSS is essential for a programmer then they have limited exposure to situations where it is not (ie. anything other than applications to generate web pages). The only sad thing is trying to force that attitude on programmers who have nothing to do with web based applications.
Kind of a lame article. Other than Swift, everything listed has been around for some time, is fairly well established or obvious. Also, Couchbase is moving away from Erlang due to lack of performance. No mention of some of the more interesting languages to come about recently, like Go, D, or TypeScript. No mention of the CSS framework languages like SCSS and LASS.
One great thing about C is that if you don't like, you can use it to program a compiler for a new language that you prefer.
As if that was somehow not true for languages other than C...
CLI paste? paste.pr0.tips!
Chances are, if you need fucking Dice to tell you what languages to learn, you probably aren't a very good programmer.
Check out the FreePascal project. It's still kicking. Not many reasons not to use it.
Using erlang i always end up solving my stuffs recursively,i love the language :D. Way to go Erlang :p.
In 10 years, it will be like Symbian. Dead and buried.
10 PRINT... oh wait
MacPro 4,1 2.66GHz/Radeon HD 4870/Mac OS X 10.6.x
That article was beyond useless. Says a 20+ yr. exp software engineer who's actually thinking pretty hard about the language trends.
Still languages are tools at best. I think next year you better start checking out the coping saw, as the vice grip fad seems to be pretty well played out.
Basing one's choice of language to learn on its current popularity - though this may be economically prudent - retards progress in programming languages almost as much as new languages' emphasis on backwards compatibility and ease of learning for novices.
"Backward compatibility" gives us the backward languages that predominate today. Making things easy for beginners gives us languages mostly only suitable for newbies.
Contrast this novice-oriented, backward-looking orientation with the little-understood idea that a language can be a tool of thought, that it can provide us with useful conceptual building blocks for thinking about computation at a high level.
It's not true for many languages other than C, presuming you want something compiling natively.
Edit: I meant such that you have fine-grained control over the natively compiled code -- e.g., you can write to the SIMD registers, massage the assembly code, etc. Something like compiling Java doesn't have that, unless of course you use it to wrap C.
Yeah. We need to grow he Cyber War Domain.
NOT !
...what?
C++ is a better C, a fully complete and compliant superset with, these days, better optimization and much better safety features. Just say no to C unless you are in an embedded system with no other compiler.
Well, any language that can write out a binary executable can do the things you mention. It's just a pain to have to implement your own assembler to get the results you're talking about! :)