TIOBE language of the year

[ccc31807]

TIOBE has awarded its 'language of the year' designation to Python.

Yes, I know that a language's 'popularity' is a meaningless concept,
that the metrics are suspect, and that at best it only depends on the
change in rank of the language. Still, it counts for something. It
also must count for something that schools like MIT and Ga. Tech. have
switched to Python for their introductory programming course.

I used Python some about six years ago, and liked it a lot. I have
since played with Ruby and have used it some. Of the three, I find
Perl best suited to my needs and to my personality. That said, during
the past year a couple of friends, non-programmers who find themselves
writing scripts quite a bit in the performance of their jobs, have
switched to Python.

My question is this: Is the (seeming) increase in popularity of Python
due more to any intrinsic quality of the language (i.e., is it
'superior' in some way to Perl), or is it more related to appearance
factors?

This isn't intended to start a flame war or any other kind of war.
Languages exist in a state of competition with each other, with some
languages achieving long term success (like C and Java) and others
falling into disuse (like Ada and Lisp). ISTM that some thinking on
this issue is appropriate on clpm.

CC.

 

[Nate Bargmann]

Like you I find that Perl fits my personality and way of thinking. I
also get along well with C and some Bash shell. I've spent time self-
studying Python in the past but have never managed to get my head around
the almost required full dive into OOP that using a lot of the third
party packages requires. I have no formal training in programming and I
am strictly a self-taught hobbyist--I do it for fun and my own
edification not for profit.

I find Perl's regular expressions an integrated part of the language. My
impression of RE in Python is that it's something they needed to add to
play in the same space as Perl and somehow seems to be an extra. Also, I
have found old Perl 5 scripts that run perfectly well on later Perl 5
versions. This is as it should be, IMO. I recently tried to get a
Python module working on a Linux distribution that has standardized on
Python 3 and the script failed. That was almost to be expected as the
script was shown as being tested on Python 2.4 and 2.5. As it turned out
the distribution had Python 2.7 installed as its Python 2 version and the
module threw an exception for some reason that I couldn't spend a lot of
time trying to figure it out so I gave up and ran the script on another
box where the module was packaged with the packaged Python interpreter.

It just seems to me that in a major version series that backward
compatibility should be a given. A script written for Python 2.0 should
not be broken by Python 2.7 even with its features and bug fixes.
Jumping to 3.0 is another story, of course. So that has become my
biggest gripe about Python--the need to continually update a script for
point releases. I kept wanting to like Python and it kept getting snaky
on me. ;-)

The fact is that different programming languages exist to solve different
problems. Also, they appeal to different mindsets and problem solving
approaches. I enjoy the freedom to use what suits me best. Long live
Perl.

- Nate >>

 

[ccc31807]

whatever.

If you google 'programming language popularity' you get about 2.4M
hits. If you look at the first ten hits, you get about the same kind
of thing, with roughly the same ranking. TIOBE is the most recognized,
but that's probably due to more to longevity than anything else.

What does 'programming language popularity' mean? Is the fact that one
language is more widely used than another mean that it's 'better' in
some sense? Many more people use Basic and PHP than use Erlang, yet
Erickson has used Erlang for years to do things that Basic and PHP
couldn't even dream about.

Last year, Go was the TIOBE LOTY. Python has been the TIOBE LOTY
multiple times. Perl has never been the TUOBE LOTY, but Perl has made
the top ten TIOBE list much longer and consistently than Python or Go.

Another thing. Using dice.com, 'perl' returns 4576 results, 'python'
returns 2209 results, and 'ruby' returns 1231 results. Obviously, this
is a gross measure (in both senses of 'gross' when you look at the
actual job postings), but it /IS/ a measure. Does this mean that Perl
is twice as 'better' than Python or four times 'better' than Ruby?

One last point. IMO, vi (vim) is the most powerful and most effective
editor in use today, but in my Windows shop, people have either never
heard of vi or if they have consider it as they would some loathsome
disease. Perl is old, fat, and ugly, but it cooks your meals, cleans
your house, does your laundry, and loves you more for it. As they say,
beauty is only skin deep, but ugly goes all the way through. Maybe the
'popularity' of Python is a strike against it.

(This isn't a slam against Python, which probably deserves its
popularity -- I agree with you that we shouldn't read too much in to
the TIOBE LOTY prize, but IMO it's profitable to discuss.)

CC.