Ecdat (version 0.3-9)

CRANpackages: Growth of CRAN

Description

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.

Usage

data(CRANpackages)

Format

A data.frame containing:

Version

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

Date

an object of class Date giving the date on which the count of the number of CRAN packages was determined.

Packages

an integer number of packages on the CRAN mirror checked on the indicated Date.

Source

A factor giving the source (person) who collected the data.

Details

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.

Examples

# NOT RUN {
plot(Packages~Date, CRANpackages, log='y')
# almost exponential growth
# }