GNU

  • The GNU Project and the Free Software Foundation.
  • My ~/.emacs file showing some work habits and preferences in GNU Emacs. It includes my customizations to work with AUC-TeX, as an interface for the LaTeX typesetting system, as well as my use of the ESS package to interface with the GNU R Statistical System. The R Core Team are a brilliant group of experts in the statistical and computing sciences that are developing state-of-the-art software for dealing with most of the analytical challenges that biologists and other researchers encounter in their work. I have benefited enormously from their contributions, in particular those from Drs. Brian Ripley, Doug Bates, José Pinheiro, Andy Liaw, and several others, too numerous to list but who are very well known in their fields and in the R discussion groups.
  • If you are using Emacs, you probably know about emacswiki.org, where you can get tons of ideas for customizing the best text editor around ;-) The excellent planner-mode in Emacs is an invaluable tool for keeping track of projects and tasks. You can pick up my configuration files (planner, muse, and remember), called from my .emacs file above, which are slightly modified versions of those used by the maintainer. I have also included some of my elisp snippets at my Emacs Wiki user web site.
  • ess-rutils, an Emacs library for ESS that I wrote to make my work with R more efficient and comfortable. It's a hack on Dr. John Fox's init.el file for XEmacs users. It has now been incorporated into ESS.
  • fixAbbrs is a gawk script to remove useless entries in a file with full names and various abbreviations that may be repeated or with different punctuation, which is considered irrelevant. This is a typical need for creating so-called "term-lists" for Endnote and other bibliography managers. Once the database is clean, and if you use BibTeX, stringBib will generate @string commands to use as input in *.bib files and then use the full name, or various abbreviations for journal names, etc. If strings in the *.bib files need to be replaced by macro names (shortcuts), I generate them with tbl2bibtoolrsc, and then use bibtool, with its -r (resource) option:
$ tbl2bibtoolrsc.awk -abbr biosis.csv > bibtoolrsc
$ bibtool -r bibtoolrsc [bibfile]

diveMove: analysis of time-depth recorder data

I have developed this package to perform some basic analyses of these sort of data. The package is available as a contributed package from CRAN, which means it can be installed from GNU R with:

 R> install.packages("diveMove")

A quick overview of functionality can be obtained by executing the examples in ?diveMove or, more comprehensively, by reading its vignette (vignette("diveMove")). The vignette builds up on the article published in R News 7(3): 8-14 as an introduction to diveMove.

Comments for improvements, or about how it is being used, are always welcome.

Development versions of the package can be obtained from R-Forge, where a mailing list and forum are also available.

Major Features

  • Based on S4 formal classes and methods.
  • Formal classes to store depth with or without other concurrent information from TDR models.
  • A class to store information from various stages of analysis.
  • Methods to extract information from the classes above, and to plot the entire object, or a selection of dive(s). These include a plot method to zero-offset correct data, visualize given sections of a dive(s), and provide extensive summaries for the whole record, or different sections of it at various scales.

Screenshots

  • plot method for TDR class.
  • plot method for TDRcalibrate class, showing identification of activity phases.
  • Another plot method for TDRcalibrate class, subsetting dives, and showing results from dive phase identification.

Documentation

The vignette provides an overview of the package's functionality, so it is the suggested starting material. The help pages have more in-depth descriptions of each tool in the package.