Data casually collected on the number of packages on the Comprehensive R Archive Network (CRAN) at different dates.
NOTE: This could change in the future. See Details below.
data(CRANpackages)A data.frame containing:
an ordered factor of the R version number primarily in use at the time. This was taken from archives of the major releases at https://svn.r-project.org/R/branches/R-1-3-patches/tests/internet.Rout.save, ... https://svn.r-project.org/R/branches/R-3-1-branch/tests/internet.Rout.save
an object of class Date giving
the date on which the count of the
number of CRAN packages was determined.
an integer number of packages on the
CRAN mirror checked on the indicated
Date.
A factor giving the source (person) who collected the data.
This seems to provide the most widely available source for data on the growth of CRAN, manually recorded by John Fox and Spencer Graves. For a discussion of these and related data, see Fox (2009).
For more detail, see the
CRAN packages data on GitHub maintained by Hadley Wickham.
This contains the description file of every
package uploaded to CRAN prior to the date of
Hadley's most recent update. The current
maintainer of the Ecdat and Ecfun
packages would consider contributions along the
following lines:
1. It might be nice to have a more complete
dataset or datasets showing CRAN growth.
This might include code fitting multiple models
and predicting future growth with error bounds
computed using Bayesian Model Averaging. These
model fits might make an interesting addition to
the examples in this help file. With a little
more effort, it might make an interesting note
for R Journal. Functions written to fit
those models might be added to the Ecfun
package.
2. It might be nice to have a function in
Ecfun to download the
CRAN packages
data from GitHub and convert it to a format suitable for
updating this dataset.
The current maintainer for Ecdat and Ecfun
(Spencer Graves) might be willing to accept code and
documentation for this but is not ready to do it himself at
the present time.
plot(Packages~Date, CRANpackages, log='y')
# almost exponential growth
Run the code above in your browser using DataLab