[BangPypers] Python grows in 2007

[Anand Balachandran Pillai]

Hard book sales need not indicate growing popularity of a language.
In a way it could indicate that the language is more used in the
software industry, because the typical buyer of an open source
language book does it for a company project. For Python the
lesser book sales I think indicates that the language is so easy to
learn from the Internet and from groups like this, that there is no
real need of a hard copy of a book. Also Python has more free
books (ebooks) out there then Ruby.

 Regarding the absolute percentage growth if language A grows
from 10 packages to 15 it really means it *has* grown by 50%
by itself and it is impressive. In this case the hypothetical language
is real young, so growing from 10 to 15 is to be also considered
as growth. There is no real way to compare this  kind of a percentage
*across* languages, which differ widely in their user base and code base.

I think in the case of Python, PyPI is pretty incomplete. There are lots of
excellent Python packages out in the wild which have not registered with
PyPI. I am not a Perl monk, but I guess this is quite different in the
case of CPAN which starts with "Comprehensive" as its first word anyway :)

--Anand

On Jan 4, 2008 4:23 PM, Indrajith K <indrajith.k at gmail.com> wrote:
> Number of packages to a good extent can be taken as a indicator of the
> reach of the language (though it makes sense more to languages like
> python, perl, ruby etc). Perl has a long history. CPAN has almost
> every thing you require. The growth won't be really feeble, still
> number of packages for perl increased by 1140, python increased its
> stretch by 857. I agree that python standard lib can replace many CPAN
> modules, but it is hard to believe Python std lib + cheeseshop  = perl
> std lib + cpan. Python growth is great, but it has a long way to go.
> To me ruby is not in my radar (though oreilly says the ruby books
> sales out numbered python books in 2007, it is where Python was few
> years ago).
>
>
> On Jan 4, 2008 4:01 PM, Anand Balachandran Pillai <abpillai at gmail.com> wrote:
> > I think you made a mistake by dividing over the total number of
> > packages. The percentages shown by dividing the incremental
> > growth of say Perl by the total growth does not give any indication
> > of an index. It is like adding apples to oranges.
> >
> > The more valid index is the relative growth of the package size
> > w.r.t the previous package size i.e (p2-p1)*100/p1. That is very
> > simple growth percentage of each language and here it is.
> >
> > Ruby: 3.5%
> > PHP: 5.2%
> > Perl: 9.8%
> > Python: 35.8%
> >
> > Interpret the numbers yourself!
> >
> > --Anand
> >
> >
> > On Jan 4, 2008 3:37 PM,  <baiju at muthukadan.net> wrote:
> > > Last year (May 2007) I posted a blog about comparing different package
> > > repositories (http://baijum81.livejournal.com/20775.html).
> > >
> > > The total number of packages in different repositories was like this:
> > >
> > > 1. Perl (11643) http://cpan.org/
> > > 2. Python (2392) http://cheeseshop.python.org/pypi
> > > 3. Ruby (1587) http://raa.ruby-lang.org/
> > > 4. PHP (442) http://pear.php.net/packages.php
> > >
> > >
> > > When I read this thread I thought just revisit the repositories.
> > > Here is the new status:
> > >
> > > 1. Perl (12783) http://cpan.org/
> > > 2. Python (3249) http://cheeseshop.python.org/pypi
> > > 3. Ruby (1642) http://raa.ruby-lang.org/
> > > 4. PHP (465) http://pear.php.net/packages.php
> > >
> > > Here is the increment details:
> > > Disclaimer: I am not a statistical guy :)
> > >
> > > #!/usr/bin/env python
> > >
> > > # May 21st 2007
> > > perl1 = 11643
> > > python1 = 2392
> > > ruby1 = 1587
> > > php1 = 442
> > >
> > > # Jan 4th 2008
> > > perl2 = 12783
> > > python2 = 3249
> > > ruby2 = 1642
> > > php2 = 465
> > >
> > > perld = perl2 - perl1
> > > pythond = python2 - python1
> > > rubyd = ruby2 - ruby1
> > > phpd = php2 - php1
> > >
> > > total = perld + pythond + rubyd + phpd
> > >
> > > print "Perl", (perld/float(total))*100
> > > print "Python", (pythond/float(total))*100
> > > print "Ruby", (rubyd/float(total))*100
> > > print "PHP", (phpd/float(total))*100
> > >
> > > # Output:-
> > > #
> > > # Perl 54.9397590361
> > > # Python 41.3012048193
> > > # Ruby 2.65060240964
> > > # PHP 1.10843373494
> > >
> > > Again, I am not going to any conclusions ;)
> > >
> > > Regards,
> > > Baiju M
> > >
> > >
> > >
> > > _______________________________________________
> > > BangPypers mailing list
> > > BangPypers at python.org
> > > http://mail.python.org/mailman/listinfo/bangpypers
> > >
> >
> >
> >
> > --
> > -Anand
> >
> > _______________________________________________
> > BangPypers mailing list
> > BangPypers at python.org
> > http://mail.python.org/mailman/listinfo/bangpypers
> >
> _______________________________________________
> BangPypers mailing list
> BangPypers at python.org
> http://mail.python.org/mailman/listinfo/bangpypers
>



-- 
-Anand