While I had used version control on earlier projects, I didn't start using version control for collaborative research projects until reading Code and Data for the Social Sciences: A Practitioner’s Guide by Matthew Gentzkow and Jesse M. Shapiro. If you haven't read it, it's a good read (I agree with the general guidelines and most of the specifics).
The first decision is which files to version. I version the dta, gph, tex, and log files. I chose not to version easily generated files such as different formats of outputted figures and tables and instead generate them automatically using my Makefile. I normalize the dta, gph, and log files before committing them so that changes are noted only if real content has changed.
Some miscellaneous tools: rm-non-svn.sh, svn_batch_rename.sh.
The first decision is which files to version. I version the dta, gph, tex, and log files. I chose not to version easily generated files such as different formats of outputted figures and tables and instead generate them automatically using my Makefile. I normalize the dta, gph, and log files before committing them so that changes are noted only if real content has changed.
Some miscellaneous tools: rm-non-svn.sh, svn_batch_rename.sh.
No comments:
Post a Comment