The Rise and Fall of Programming Languages in 2011

[Wnderer]

Some fodder for the bestest coding debates.

http://drdobbs.com/mobile/232400093

It's about the TIOBE index for 2011. It uses search engine usage to figure out what languages are being heavily used.
http://www.tiobe.com/index.php/content/ ... index.html

[Proginoskes]

Interesting that C++ took a nosedive around 2005 ... That was about when some of the standards were changed (including the addition of namespaces).

And I'm surprised that Perl isn't more popular; it's powerful because every algorithm has been programmed in it. About the only thing missing is:

Code: Select all

print IsContinuumHypothesisTrue();

[EvanED]

Interesting that C++ took a nosedive around 2005 ... That was about when some of the standards were changed (including the addition of namespaces).

Um, no?

The first standard was published in 1998 (including namespaces), and the second in 2011. There was a revision around 2003 but those were only minor, and the publication of TR1, but that was only library additions.

Edit had "editions" instead of "additions"

[TNorthover]

Interesting that C++ took a nosedive around 2005 ... That was about when some of the standards were changed (including the addition of namespaces).

Eh? Namespaces were in C++98, and I don't think they were one of the features compilers especially lagged in. C++03 exists, but really didn't do much to the language apart from clarifying points -- I don't usually draw a distinction between '98 and '03, and I write compilers.

It does look like TR1 came out in 2005, but that was just a bunch of libraries (superceded in large part by C++11 now).

And I'm surprised that Perl isn't more popular;

Not quite the word I'd use. "Gratified" is closer.

Edit: Sorry for the similar post to EvanED. There are a couple of different bits and I couldn't be bothered to rewrite to disentangle.

[EvanED]

And I'm surprised that Perl isn't more popular;

Not quite the word I'd use. "Gratified" is closer.

I thought about responding to that part of the post but couldn't come up with anything succinct that wouldn't make me mad writing it. Yours is good.

[Wnderer]

I think the story here is what happened to Java in 2004. Java took a big dip, losing coders to PHP, Python and Visual Basic. Java recovers in 2005 taking back some from Python and Visual Basic but also capturing some from C++. PHP holds onto it gains. I don't know why this happened but the drop in Java seems to coincide with the closing of the Sun Microsystems' Newark California plant.

[Great Justice]

The real debate is how meaningful that chart is. Very little or tiny amount?
It's based on how much noise people are making about a language without any real context.
How can you compare C with perl with javascript? Completely different industries and domains.

SQL is below Lisp and Pascal?? at least 90% of our products use a DB and require SQL knowledge, whatever the main language is.
I'm strongly inclined to say there's something akin to Simpson's paradox in play for a few items there, being "normalized" between such different sources.

Langpop.com is broken down better, but still has many shortfalls, imo.
Eg. job listings: for PHP/Java monkeys, we almost constantly have something up to catch decent prospects; for C/C++ engineering, we'll go to universities to recruit or move up a talented coder. The high skill positions will rarely be posted publicly.

Interesting that C++ took a nosedive around 2005

what happened to Java in 2004

Try reading the FAQ on the same page... which basically proves my point - statistics based on google skillz are now ACCURATE TO THREE DECIMAL PLACES!

[Wnderer]

what happened to Java in 2004

Try reading the FAQ on the same page... which basically proves my point - statistics based on google skillz are now ACCURATE TO THREE DECIMAL PLACES!

Yes. you're right. It is explained in the faq.

Q: What happened to Java in April 2004? Did you change your methodology?

A: No, we did not change our methodology at that time. Google changed its methodology. They performed a general sweep action to get rid of all kinds of web sites that had been pushed up. As a consequence, there was a huge drop for languages such as Java and C++. In order to minimize such fluctuations in the future, we added two more search engines (MSN and Yahoo) a few months after this incident.

[Proginoskes]

Interesting that C++ took a nosedive around 2005 ... That was about when some of the standards were changed (including the addition of namespaces).

Um, no?

The first standard was published in 1998 (including namespaces), and the second in 2011. There was a revision around 2003 but those were only minor, and the publication of TR1, but that was only library additions.

The first C++ book I got (C++ by Till Jeske) didn't mention namespaces at all, and it was published in 2005 2002.

It took me a long time to figure out what I was doing wrong; I couldn't even get a "Hello world" program to work in C++. In fact, the very first program mentioned in that book is

Code: Select all

#include <iostream.h>
main () {
   cout << "C++ programming is fun." << endl;
}

Edit had "editions" instead of "additions"

I almost did that in another post tonight ...

[EvanED]

Interesting that C++ took a nosedive around 2005 ... That was about when some of the standards were changed (including the addition of namespaces).

Um, no?

The first standard was published in 1998 (including namespaces), and the second in 2011. There was a revision around 2003 but those were only minor, and the publication of TR1, but that was only library additions.

The first C++ book I got (C++ by Till Jeske) didn't mention namespaces at all, and it was published in 2005 2002.

Then it wasn't a very good book. (To be fair I'm not sure what proportion of C++ books at that point were good.)

It took me a long time to figure out what I was doing wrong; I couldn't even get a "Hello world" program to work in C++. In fact, the very first program mentioned in that book is

Code: Select all

#include <iostream.h>
main () {
   cout << "C++ programming is fun." << endl;
}

I actually had the same experience because of the AP CS curriculum when I tried to use, say, the standard string class instead of apstring. Took me a while to figure out what was going on.

[phlip]

Sounds like the book was just quite out of date when it was printed.

One thing I've seen with some C++ implementations is that eg #include <iostream> will bring in all the iostream declarations in the std namespace, while #include <iostream.h> will bring them in in the global namespace. The with-extension header basically just including the standard header with a bunch of using declarations. Same for all the other standard C++ library headers. AFAICT that isn't actually specified anywhere in the standard, and those libraries are just doing it for compatibility (with older versions of the implementation that predated the standard proper, and had headers with .h extensions and no namespaces).

The C++ standard does say that similar alternate headers exist for the standard C library (eg <stdlib.h> puts everything in the global namespace, and <cstdlib> puts everything in std) but I don't think the C++-library part is standard.

[EvanED]

Sounds like the book was just quite out of date when it was printed.

Which makes it not very good in my book on this point. I mean, I guess it could be okayish if it was totally awesome in most other respects.

One thing I've seen with some C++ implementations is that eg #include <iostream> will bring in all the iostream declarations in the std namespace, while #include <iostream.h> will bring them in in the global namespace. The with-extension header basically just including the standard header with a bunch of using declarations.

I don't know for sure that this is true on modern implementations (or if the implementers got tired of maintaining both versions), but I think there are some other technical differences as well regarding error conditions and maybe a couple other things. There at least were between the historical iostreams libraries and the standardized one.

AFAICT that isn't actually specified anywhere in the standard, and those libraries are just doing it for compatibility (with older versions of the implementation that predated the standard proper, and had headers with .h extensions and no namespaces).

Correct.

[phlip]

I don't know for sure that this is true on modern implementations (or if the implementers got tired of maintaining both versions), but I think there are some other technical differences as well regarding error conditions and maybe a couple other things. There at least were between the historical iostreams libraries and the standardized one.

Possibly for some implementations, I dunno. Here's the backward/iostream.h that I have on this machine (GCC 3.4.4 MinGW):

Code: Select all

#ifndef _BACKWARD_IOSTREAM_H
#define _BACKWARD_IOSTREAM_H 1

#include "backward_warning.h"
#include <iostream>

using std::iostream;
using std::ostream;
using std::istream;
using std::ios;
using std::streambuf;

using std::cout;
using std::cin;
using std::cerr;
using std::clog;
#ifdef _GLIBCXX_USE_WCHAR_T
using std::wcout;
using std::wcin;
using std::wcerr;
using std::wclog;
#endif

using std::ws;
using std::endl;
using std::ends;
using std::flush;

#endif

Meanwhile, MSVC doesn't have a iostream.h, and neither does the GCC 4.5.2 on my home machine.

[EvanED]

This page has some information regarding the differences between the legacy <iostream.h> and standardized <iostream>. And thus endith your C++ history lesson, and sorry for hijacking the thread.

(The differences are actually a lot more than I thought. For instance, the iostream classes were actually classes and not templates in the legacy version.)

[Proginoskes]

Interesting that C++ took a nosedive around 2005 ... That was about when some of the standards were changed (including the addition of namespaces).

Um, no?

The first standard was published in 1998 (including namespaces), and the second in 2011. There was a revision around 2003 but those were only minor, and the publication of TR1, but that was only library additions.

The first C++ book I got (C++ by Till Jeske) didn't mention namespaces at all, and it was published in 2005 2002.

Then it wasn't a very good book. (To be fair I'm not sure what proportion of C++ books at that point were good.)

I found out the first part later. After I bought it and started reading through it, I found LOTS of typos. I mean obvious things. Like

Code: Select all

cout << "'num' has a value of " << count << endl;

[btiffin]

COBOL!

[userxp]

I find it hard to believe that Javascript is going down with all that "HTML5 is the bestest standard ever, let's do everything in it" going on.

[Kitutal]

I picked C, because I figured it was the language version of microsoft paint, which I quite like. Also, I like the name, and I'd heard nice things about it and couldn't be bothered to research all the rest.
anyways, looks like most of them are going slowly down, that's what you get when people keep coming up with new ideas, there should be a line for other and see just how high that shoots up towards the end. we need one universal standard language, any suggestions..?

[Dason]

we need one universal standard language

Why?

[Iranon]

we need one universal standard language, any suggestions..?

Italian would be nice.

[wumpus]

The first C++ book I got (C++ by Till Jeske) didn't mention namespaces at all, and it was published in 2005 2002.

It took me a long time to figure out what I was doing wrong; I couldn't even get a "Hello world" program to work in C++. In fact, the very first program mentioned in that book is

Sometime around then (2005ish) I needed to use some C++ libraries. After checking out whatever "learn C++ in FIXNUM days" were at the library, off I went.

I don't believe any of them mentioned the use of the "new" keyword and the difference in placing objects on a heap vs. stack. Doesn't make a difference in hello_world type programs, but you never understand why nothing else works at all... Eventually found the problem through a long google search.

PS. MMIX should be the universal programming language. Any other programming language should first be compiled into MMIX.