I think you're looking at this from the wrong angle

I agree completely. Languages don't succeed because they have more features. They succeed because they're better at doing certain things (even if some of those things happen to be "sounding safe to management" :-)

There is no single language that has more features than all of the others.

Hmmmm... Personally I'd feel fairly happy saying, for example, that Lisp and Perl had more features than Java and C - in the sense that they support more styles of development in the core language.

How can C not support more styles of development than Perl or C++? both are, or were, written in C. The first "compilers" for C++ actually translated to C so C has to be able to do OO (albiet not in a fashion you would want to read).

How much of the Perl we write is really "out of the box Perl"? Pretty much anything we write should at least include "use strict; use warnings;", but straight off we are using modules, core for sure, but are they part of the language?

How much C is really "out of the box C"? Ever used printf or any other I/O in C? Is I/O part of the language?

At the end of the day what is part of the language is unimportant. What is more important, as has been suggested by others is, how well does the language fit the task at hand and the range of tasks handled.

While you can solve most problems in most languages, some are just naturally (or designedly) better at some tasks than others. For me Perl is an excellent tool that makes simple things easy. I'm not so worried by hard things and often for hard things there is a better candidate, but there are many more simple things that have to get done than there are hard things.

So, until all the simple jobs are done, I think Perl has a pretty firm place in the scheme of things.

---

DWIM is Perl's answer to Gödel

How can C not support more styles of development than Perl or C++? both are, or were, written in C. The first "compilers" for C++ actually translated to C so C has to be able to do OO (albiet not in a fashion you would want to read).

Of course you can do OO development with effort in raw C - as you could in perl 4 come to that. But there is no support in the core language for methods, classes, inheritance, etc. You have to build it yourself. So I think it's perfectly reasonable to say that Lisp or Perl or Ruby support certain development styles that C does not.

Personally I'd feel fairly happy saying, for example, that Lisp and Perl had more features than Java and C

Oh, absolutely. You'd be crazy to deny, for example, that Python has more features than FORTRAN. There are certainly languages that have more features than others. But would you be happy pointing at one language and saying that it had more features that all other languages? I don't think it can be done in any meaningful way. That's the point I was trying to make.

--
<http://dave.org.uk>

"The first rule of Perl club is you do not talk about Perl club."
-- Chip Salzenberg

But would you be happy pointing at one language and saying that it had more features that all other languages?

Common Lisp maybe :-)

I think (and preface most of the rest of this with these two words) the problem is using the word "features"; underneath it all you're just twiddling your infinitely long paper tape, so in that sense no language is more feature-full than any other.

The difference you're trying to highlight is the difference in the abstractions that the language provides that push whatever grunt work/complexity/minutia further out of the programmer's ken and in doing so make their job simpler, and those can be qualitatively compared between languages.