#+INCLUDE: ../org-templates/level-1.org
#+title: GNU/Linux users per 1000 windows users
#+date: <2020-12-22 Di>
#+LANGUAGE: en

TLDR: In 2020 the fraction of Linux users doubled compared to Windows
      users. Compared to 2008 the Linux-user fraction rose by factor 5.

Every year people ask “is it finally the year of the Linux Desktop?”.

In 2020 the answer is clear: since 2008 almost every year was a year
of the Linux Desktop: Compared to Windows users there was reliable
superlinear growth:

#+begin_export html
<div class="sourcebox" markdown="1" >
<a href="gnu-linux-desktop-share.pdf"><img title="PDF" src="../assets/pdf-thumbnail.png" /></a><br />
<a href="gnu-linux-desktop-share.pdf">PDF</a> (drucken)
</div>
#+end_export
 
#+begin_src gnuplot :var data=gnu-linux-per-kw :file gnu-linux-desktop-share.png :exports results
reset
set title "GNU/Linux users per 1000 Windows users"
set ylabel "GNU/Linux desktop share / GNU/kw
set terminal png size 420,320
plot data u 1:2 notitle with lines lw 2 lt rgb "#1E90FF"
#+end_src

#+RESULTS:
[[file:gnu-linux-desktop-share.png]]

The source of this graph is data I collected over the past decade
every July from [[http://www.netmarketshare.com/os-market-share.aspx?qprid=9][OS market share]]. They do browser detection, so the
data is far from perfect, but at least they have data we can use:

# TODO: replaced this first line due to a bug in org breaking the table extraction. Should be fixed in 9.5.3
# | 2008 | 8.6 | GNU/Linux users per 1000 Windows users (GNU/kw) |


#+name: gnu-linux-per-kw
| 2008 | 8.6 | GNU/kw |
| 2009 |  11 | GNU/kw |
| 2010 |  10 | GNU/kw |
| 2011 |  11 | GNU/kw |
| 2012 |  12 | GNU/kw |
| 2013 |  14 | GNU/kw |
| 2014 |  19 | GNU/kw |
| 2015 |  18 | GNU/kw |
| 2016 |  22 | GNU/kw |
| 2017 |  25 | GNU/kw |
| 2018 |  21 | GNU/kw |
| 2019 |  24 | GNU/kw |
| 2020 |  41 | GNU/kw |
| 2021 |  20 | GNU/kw |

*GNU/Linux is growing.* (though there is a dip during the pandemic)

When will you migrate? Did you migrate already?

------

/This is based on an older German article I wrote in 2010 and updated regularly: [[http://www.draketo.de/licht/politik/leserbriefe/gnulinux-gewinnt-marktanteile][GNU/Linux gewinnt gegenüber Windows und MacOSX an Marktanteil]]/

* FAQ

** Could the increase be due to Windows users moving to mobile platforms?

Windows users did not go down enough, and MacOS did not rise along, to
warrant a 100% increase of GNU/kw. It’s at 87% while GNU/Linux in July
was only 3.57%. If this was due to an exodus of Windows-users, I would
expect MacOS to rise along.

The results are filtered for Desktop. When including mobile, Windows
did not have sufficiently high losses compared to mobile to explain
the Linux numbers differently than with an increase of Linux users.

# #+toc: headlines 2


#+setupfile: ../org-templates/level-1-politik.org
# Local Variables:
# org-confirm-babel-evaluate: nil
# org-export-default-language: en
# org-babel-noweb-wrap-start: "{{{"
# org-babel-noweb-wrap-end: "}}}"
# org-latex-images-centered: nil
# End:
# override template options:
#+options: toc:nil H:1 num:nil ^:nil