Our Indiana U Workshop Went Well

This post originally appeared on the Software Carpentry website.

Our two-day bootcamp at Indiana U. on March 7 and 8 went well—while less than half of the people who registered actually showed up (more on that in a moment), everyone who did attend the two days seemed to get a lot out of it. Special thanks are due to Prof. Andy Lumsdaine (host), Jennifer Richards (organizer), and Ben Martin, Randy Heiland, Joe Cottam, Mike Hansen, DongInn Kim, and Purdue's own Jeff Shelton for helping out.

At the end, we asked everyone to give us one good and one bad thing about the bootcamp:

Good Bad
  • Shell scripting
  • Data provenance
  • Tips on efficiency and coding
  • The five-step cycle (update, write tests, make them pass, refactor, commit)
  • Where to find more info
  • Got started with basic Unix
  • Good structure
  • Comfortable environment
  • Good examples
  • Help team was awesome
  • Website is useful
  • Saw lots of new tools
  • Examples were talked through
  • Energetic instructor
  • Good breadth
  • Test-driven development
  • Version control
  • Enjoyed the stories (even though they weren't all true)
  • Didn't get to web stuff
  • Didn't get to look at code examples long enough
  • Went too fast through Python
  • Less support for Windows users
  • Not enough exercises
  • Not enough time to practice
  • Didn't see how to create databases (only how to query them)
  • Too much breadth
  • Didn't see how to create a repository
  • No reference guide/cheat sheet
  • Didn't explain keyboard shortcuts
  • Printed instruction would have helped (ESL)
  • Free lunch is nice, but what about free breakfast?
  • Some parts too advanced
  • Too much typing: should be more pre-cooked data files for download
  • Time was too limited
  • Didn't show IPython
  • Too many "drive-by" topics
  • No capstone exercise to bring everything together
  • Dove into databases too quickly

I agree with most of what's in the second column (especially the comment about breakfast). What I want to fix first, though, is the high no-show rate that plagues any free event. One suggestion is to charge people $20 or so when they register, then refund it when they show up, so that only no-shows wind up paying anything. What would your reaction to that be?

We'll start online tutorials with our Indiana students next week—like their predecessors from the Space Telescope Science Institute, they'll meet once a week to work through a few simple problems that are directly relevant to their research. Next up: the Monterey Bay Aquarium Research Institute and Lawrence Berkeley National Laboratory.

Dialogue & Discussion

Comments must follow our Code of Conduct.

Edit this page on Github