add workflows

workflows.md

@@ -0,0 +1,43 @@
+# Workflows and computing!
+
+Biology is increasingly data intensive!
+
+Command line skills are the new pipetting!
+
+The most important skill to develop is the confidence that the solution
+is out there and you can find it. **Everybody learns something new
+every day they do data analysis!**
+
+---
+
+Why are we using the cloud??
+
+* easy install of software!
+* big compute power and disk.
+* general skills that "travel" - not specific to one location or computer type.
+
+Why do we teach command line??
+
+## Workflows, and responsible conduct of computational research
+
+The basic goals:
+
+* readers and reviewers of your work should be able to discover what
+  you did to analyze your data, starting from the raw data and going all
+  the way to your final figures.
+
+* if you can automate the entire process, this will serve the above purpose
+  and also lead to efficiency gains for yourself in updating analyses.
+
+* automating small data analyses will seem unnecessary, but the skills you
+  gain in doing this will enable you to handle much larger data analyses
+  much more easily.
+
+  Think - could you do a particular task on 1 data set? 5 data sets? 50 data
+  sets? 5000 data sets?
+
+## Reading material
+
+[Computing Workflows for Biologists: A Roadmap](http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1002303)
+
+Repeatability considerations - see [2016 Oslo workshop](https://2016-oslo-repeatability.readthedocs.io/en/latest/)