Where We Are (June 2012 edition)

This post originally appeared on the Software Carpentry website.

Last week, a double dozen friends and colleagues gathered physically and virtually to review what we've done and where we should go next. I'll postpone discussion of the biggest "what next" items to separate posts, but here's the status report itself.

Successes

1. We've reached a lot of people
   • We have delivered 12 workshops to 376 learners in 4 countries (curriculum)
   • Numbers will be 18 workshops, 520 learners, and 5 countries by the end of July
   • Have actually lost count of the number of online tutorials we've run...
2. What we're doing is helping
   • 96% of participants happy with training
   • Significant improvement in computational understanding
   • We get mail
3. Scientists want more
   • 17 workshops in the pipeline
   • 2-4 new contacts/month
4. We can scale
   • 20 co-instructors
   • 7 have gone on away missions
   • 5 have run online tutorials
5. We can become self-sustaining
   • Have raised $25K in other funding
   • Shifting to "host pays costs" model for workshops
   • Will try to count Software Carpentry training toward NSF requirements for training in Responsible Conduct of Research
   • Working with the Software Sustainability Institute to create a driver's license for the DiRAC supercomputing facility

Room for Improvement

1. Very little new video content (because very little demand)
   • But some interesting experiments using video for formative assessment
2. No learner-led challenges
   • Most grad students feel their research is challenging enough
   • Having more mentors, from more disciplines, may help us help more people with their research problems
3. Badging delayed pending a software release (mid-July)
   • Initially badge instructors rather than learners (greater immediate value to recipient)
4. Never-ending installation/setup/configuration difficulties
   • Particularly for Windows users (20-40% of our learners)
   • All technical solutions have proven to be unsatisfactory in different ways
5. Some material is going stale and/or not as useful as we thought (e.g., Subversion, databases)
   • Will update/replace
6. Pacing continues to be a problem
   • 20% find it too fast while 20% find it too slow
   • More instructors will allow us to stream people appropriately
7. Online tutorials have had mixed success
   • "Only" 10-20% of learners still participating after 3 weeks
     • Which is actually higher retention than many online courses
   • Partly because of technical difficulties and time pressures
   • But online instruction remains less effective than in-person instruction with today's technology
8. Workshop participants are predominantly male (even when drawing from disciplines that are gender-balanced)
   • Now in contact with WiSE/diversity groups in the US to try to arrange workshops for their members
9. Still don't know what to teach scientists about the web