3 ms·

Why so many programming languages?

• dwc 16y ago
  The reason why there are so many languages is that there's no committee deciding which languages are ok and which are not. And that's really, really awesome, because committees would stink at that. Instead, we have the ability to choose among all these languages, or even to make our own. This is a good thing. Different languages approach things in novel ways, and learning new languages is a good way to broaden your skill as a developer. Learning new languages also gives you more tools in your toolbox, so you hopefully won't be stuck pounding in a screw with a hammer. Some developers chose one language and stick with it, maybe for their whole career. This can be you, if you want that. But don't be surprised if your career becomes maintaining legacy apps, not that there's anything wrong with that.
  • Kristo5747 16y ago
    @dwc: sensible answer to a "problem" that made my head spin. Thanks for your insight. Much appreciated. As far as "broadening your skill as a developer", it's another story. Maybe I should become self employed so I can finally master one trade and stop being the jack of so many.
• timrobinson 16y ago
  This isn't particularly scientific, but the list of "2000s languages" on Wikipedia is relatively short: http://en.wikipedia.org/wiki/Timeline_of_programming_languages http://en.wikipedia.org/wiki/Timeline_of_programming_languag...
• madhouse 16y ago
  There were quite a lot of languages even ten years ago. There was shell, perl, python (albeit fairly young at that point), tcl, lisp, ada and so on and so forth. The trend merely continued.
• Kristo5747 16y ago
  Why shouldn't there a one size-fits-all type language? To the un-initiated (i.e me), this makes sense: one for mobile devices (java), one for the web (PHP?), one for UIs (javascript) and one for low-level work (c/asm). How can someone who makes a living writing code be expected to keep up with the trend that @madhouse talks about?
• cd34 16y ago
  Almost every language fixes a certain problem in a way that the authors felt was better than any existing method. In college I took part in an experimental Ada course - two semesters - taught by arguably one of the first dozen or so experts in Ada. Ada was going to save the world - everything in the US Government was going to move to Ada, learn it or be a dinosaur. Ada had some very unique traits that made it very flexible. Dealing with financial figures/budgets? overload your math operators to handle currency. Dealing with extremely accurate numbers for missile tracking? overload your math operators to handle extremely precise numbers. The problem was, after receiving this Ada training, replacing Cobol (which was the main intent), never took place. Now, the US Government allows programming groups to choose the 'right' language for the job. Ada is still used in some government and civillian tasks, but, it certainly didn't live up to the original hope. I believe portions of the Boeing 777 use Ada. It was still cheaper and easier to maintain Cobol code than to rewrite - and still is. I know two recruiters that occasionally mention their passing need for Cobol programmers - and it's been 25 years since Ada was going to replace Cobol. PHP was originally developed to get away from Perl CGI and allow pages to be easily scripted and customized. Every language has a particular strength or weakness and their own following. Not that you can't accomplish almost any task with any of them, it is selecting a language that might handle that task more efficiently. Even for something like webapps, if you had 30 people in a room, I seriously doubt you would get any consensus on the best language for web scripting. Add on top of the language choice, frameworks, and you've got another religious war. I don't think I have any projects that use a single language in deployment, even discounting bash from the equation. There are things that I find easier to accomplish in different languages.