Python's popularity statistics

[Aaron K. Johnson]

In message <3DF80544.6C9F70A7 at engcorp.com>, Peter Hansen wrote:
> "Aaron K. Johnson" wrote:
> > 
> > In message <mailman.1039658234.21778.python-list at python.org>, Chad Netzer
> wrote:
> > > On Wednesday 11 December 2002 17:24, Mike Dean wrote:
> > > > On Wed, 11 Dec 2002 14:49:11 -0600 Aaron K. Johnson
> > > <akjmicro at yahoo.com> wrote:
> > > > > python 9647
> > > > > clipper 8960
> > > >
> > > > Now this is what I find real interesting.  Clipper, right below
> > > > Python? I thought Clipper was quite dying, in this client-server
> > > > world...  does it still have that big a following?
> > >
> > > The most important step in any statistical analysis (except perhaps
> > > for gross accuracy of the numbers themselves), is interpretation.
> > > Drawing any subjective conclusions from these numbers, such as
> > > "popularity of language", etc. is essentially meaningless.  Others
> > > have already pointed out that "popularity of language" and
> > > "activity of newsgroup" could be inversely correlated, as far as we
> > > know.  Furthermore, we have no knowledge of how on-topic any
> > > specific newsgroup is (not from the raw numbers anyway).
> > 
> > Do you really think beta might be more popular than Java?
> > I still contend that in some cloudy way these number directly represent
> usage.
> 
> You mean in some cloudy-but-*real* way, right?  ;-)
> 
> I appreciate the effort in posting these numbers.  You'd probably
> be better off letting people draw their own conclusions though: 
> several of your posts contradict each other, as near as I can 
> tell.  Either Python and Clipper are both very popular and 
> well-engineered, or the meaning of the statistics is very open
> to wide interpretation.
> 

If you read carefully, you'll find that I said that popularity had nothing to
do with quality of engineering: my quips at perl for example.

Everyone has raised some interesting points, though. ;)

For what's it's worth, here's the data from freshmeat.net about what languages
are used for open source projects. There is a strong correlation between my
initial newsgroup data and these numbers, proving my point.

And again, I used python to sort all this stuff out, very easily ;)

-Aaron.


C 4509
Perl 2354
C++ 1932
Java 1693
PHP 1531
Python 856
UnixShell 459
Tcl 308
SQL 245
JavaScript 167
Assembly 143
Other 141
Ruby 84
OtherScriptingEngines 61
Scheme 60
ObjectiveC 57
Lisp 56
PL/SQL 46
Delphi 37
Fortran 30
Ada 30
Pascal 29
Awk 28
Emacs-Lisp 27
Zope 25
ML 23
Haskell 21
Eiffel 17
C# 17
ASP 16
Smalltalk 15
Forth 11
Basic 11
VisualBasic 10
OCaml 10
ColdFusion 9
Erlang 7
Rexx 6
ObjectPascal 6
Prolog 5
YACC 4
Modula 4
Dylan 4
Pike 3
APL 3
XBasic 1
Simula 1
Pliant 1
PROGRESS 1
Euler 1
Logo 0
Euphoria 0