{smcl}
{* 20Nov2008}{...}
{cmd:help wfrules}: Rules for best practice in workflow
{hline}

{title:Basic rules for effective workflow}
{p2colset 4 8 8 2}{...}

{p2col :#:}{bf:Provenance}: The provenance of every result should be
documented.{p_end}

{p2col :#:}{bf:Posting Principle}: Once a file is posted, it should
never be changed. Never change a variable that has been posted.
Never share a file till it is posted.{p_end}

{p2col :#:}{bf:Run order}: Do-files should be named so that when run
in alphabetical order they exactly recreate your datasets
and/or replicate your analyses.{p_end}

{p2col :#:}{bf:Basic backup}: Make at least two, bit-verified backup copies;
store them in different locations.{p_end}

{p2col :#:}{bf:Drukker's Dictum}: Never type anything that you can obtain
from a saved result.{p_end}

{p2col :#:}{bf:Long's Law}: It is always faster to document it today
than tomorrow.{p_end}

{p2col :#:}{bf:Label variables}: Create a note and variable label
for all new variables. Categorical variables should have value labels.{p_end}

{p2col :#:}{bf:Never final}: Never name it final!{p_end}

{p2col :#:}{bf:Test graphs}: Alway try them before you present them.{p_end}

{p2col :#:}{bf:datasignature}: Use {help:datasignature} before saving a dataset;
check the signature when you load a dataset.{p_end}

{p2col :#:}{bf:Getting help:} The clearer and more detailed the information
you provide, the greater the chance you will get the help you need.{p_end}

{p2col :#:}{bf:Too many files:} It is easier to create a file than to
find a file; easier to find a file than to know what is
in the file; and it is tempting to create a lot of files.{p_end}

{title:Rules for replication}

{p2col:1:}Backup and archive your dataset and do-files.{p_end}
{p2col:2:}Keep a research log.{p_end}
{p2col:3:}Document the provenance of all reported results.{p_end}
{p2col:4:}Always set the seed when using random numbers (set seed #).{p_end}

{title:Also see}

    {helpb wf:Workflow}
INCLUDE help wf_footer