Data Visualization for R packages at Github #rstats

I noticed this article sometime back by the most excellent hacker, John Myles White ( author Machine learning for Hackers)

http://www.johnmyleswhite.com/notebook/2012/08/12/the-social-dynamics-of-the-r-core-team/

Professor John Fox, whom we have interviewed here as the creator of R Commander, talked on this at User 2008 http://www.statistik.uni-dortmund.de/useR-2008/slides/Fox.pdf

I also noticed that R Project is stuck on SVN ( yes or no??, comment please) while some part of the rest of the World has moved on to Git. See http://en.wikipedia.org/wiki/Git_%28software%29

Is Git really that good compared to SVN http://stackoverflow.com/questions/871/why-is-git-better-than-subversion

Maybe, I think with 5000 packages and more , R -project needs to have more presence on Github and atleast consider Git for the distributed and international project R is becoming.

Again this is how you can move from SVN with history to Github

http://stackoverflow.com/questions/79165/how-to-migrate-svn-with-history-to-a-new-git-repository

Can we move the R project from SVN – CRAN to make one mirror at Github ( note I would still keep CRAN, as a big redundancy measure)- there are 90 CRAN mirrors already http://cran.r-project.org/mirrors.html though I like the 0 cloud mirror by R Studio ( it elegantly redirects you to mirror)

But the data visualizations at Github are lovely for tracking projects! Lets see R Studio

https://github.com/rstudio/rstudio/graphs

One point of concern that has also been voiced before is the lack of succession planning with the R Core team.

Having better and more transparent commit data can show how the Project has become overtly reliant on a few gorgeously hard working men, and what can be done to build in redundancy.

Senior professors should move to articulating the roadmap vision than stick to old habits of commit addiction is what I feel as an opinion and I accept that I may be wrong or impertinent or both here.