Cheers, Michael! I'm sharing this with the powers that be at UW-Madison, whom I'm joining to plan a new institute for data science (lowercase intentional).
Also, your model for growing students into helpers into instructors is exactly how our thriving community of volunteers has trained hundreds of students per year in Software and Data Carpentry workshops since 2013. Lauren Lauren Michael - Research Computing Facilitator, Center for High Throughput Computing <http://chtc.cs.wisc.edu/>, University of Wisconsin - Madison Other Affiliations: Advanced Computing Initiative <http://aci.wisc.edu/>; Wisconsin Institute for Discovery <http://wid.wisc.edu/profile/lauren-michael/>; ACI-REF <http://aci-ref.org/> [email protected], tinyurl.com/LMichaelCalendar, Discovery 2262, (608)316-4430 On Mon, Oct 8, 2018 at 1:00 PM Michael Koontz via discuss < [email protected]> wrote: > Hi all, > > The Graduate Group in Ecology at the University of California, Davis just > added a required, quarter-long, Carpentry-style scientific computing skills > course to its curriculum. This was a result of a grass roots, student-led > effort to ensure that instruction in these skills would be reliably offered > to future students. We embraced the collaborative lesson design philosophy > and collated ~20 hours worth of content (plus some take-home assignments) > from a few sources, but mostly the DataCarpentry material. The foundation > of the content is here if you are interested: > https://gge-ucd.github.io/R-DAVIS/ > > Before the details, I want to acknowledge the folks that made it happen. > The course just became a requirement, but I think that ultimately took 10+ > years and the efforts of many, many people. > > 1) Aviva Rossi <https://www.usfca.edu/faculty/aviva-rossi> (the grad > student representative to the ecology group’s executive committee and > learner in the first full iteration of the course; Aviva led the charge in > codifying the requirement) > > 2) Ryan Peek <https://ryanpeek.github.io/aboutme/> (former grad student, > now PhD, and badged Carpentry instructor that taught a pared down version > of the course 2 years ago as a favor to the graduate group and then > co-instructed the first full iteration of the course last year) > > 3) Taylor Reiter <https://taylorreiter.github.io/aboutme/> (a grad > student and badged Carpentry instructor that taught a session last year and > was a helper for many other sessions) > > 4) Michael Culshaw-Maurer <https://mcmaurer.github.io/About_Me/> (grad > student and newly badged Carpentry instructor that was a helper for the > first full iteration of the course and will be co-lead instructor next year) > > 5) Martha Wohlfeil <https://marthawohlfeil.github.io/aboutme/> (a grad > student that was a helper for the first full iteration of the course and > will be co-lead instructor next year) > > 6) Ted Grosholz <http://www.des.ucdavis.edu/faculty/grosholz/index.htm> > (former > grad group Chair, who supported the course and advocated for codifying the > requirement) > > 7) Andrew Latimer > <https://psfaculty.plantsciences.ucdavis.edu/plantsciences_faculty/latimer/index.htm> > (ecology > professor and newly badged Carpentry instructor that will backstop the > course) > > 8) All of the scientific computing and teaching enthusiasts over the years > in Davis, Carpentry-affiliated and not, that created a thriving community > of practice and generated the momentum for our grad program to say “we > think it is important”. My institutional memory only goes back so far, so > if you were a part of this group in any capacity please let me know so that > I can document it as part of this effort! > > > *Here’s how the course is organized:* > 1) Two graduate students that are badged Carpentry instructors run the > course. They are currently paid as TA’s (salary + tuition), but the program > is looking for a way to get “instructor of record” credit to those students. > > 2) A few graduate students that are enthusiastic about teaching scientific > computing volunteer to be helpers during as many of the 10 sessions as > they’d like. > > 3) Those helpers take Carpentry instructor training and become the next > year’s instructors. > > 4) Enthusiastic learners become next year’s helpers. > > 5) The course is backstopped by a Carpentry-badged professor who could > step in during lean years with respect to grad student interest in being > instructors or helpers. > > 6) Making the course a requirement opened up new avenues for institutional > support and ensured that support would have to be found. > > There’s many reasons why I think we’re in a good position to maintain high > levels of learner and instructor motivation, even though the course is a > requirement. Here’s a few: > > *Keeping learners motivated:* > 1) The course is grad student led. We’ve had feedback that this promotes > engagement and reduces some of the anxiety inherent in graduate coursework. > > 2) There is a pretty thriving community of practice in scientific > computing skills in Davis and in the ecology program. Many alumni (some on > this list!) laid the groundwork for this and I think the normalization of > these skills keeps motivation high for new learners. > > 3) Students can opt out of the course with permission from their committee > > *Keeping instructors motivated:* > 1) Recent data from surveys of early career researchers in our field > suggest teaching is a key part of the day-to-day work of ecology > practitioners (not just in academia) (Hampton and Labou, 2017 > <https://esajournals.onlinelibrary.wiley.com/doi/10.1002/ecs2.2031>) so > there is motivation for the grad students that want to keep doing ecology > to gain formal training and experience in teaching. (funny, I just saw that > one of the authors of this awesome paper, Dr. Labou, posted earlier in this > thread!) > > 2) There is a thriving Carpentry community in Davis, so there is a > motivating and normalized ongoing conversation about teaching scientific > computing skills. > > 3) The distributed nature of our Graduate Group (it comprises lots of > departments) means that TAship funding is really valuable, even just for a > quarter. This model basically created a new TAship (or two) because no > single department has to shoulder the full financial burden every year. > > 4) The course being grad student led helps here, too, I think. The thing I > love best about Davis ecology is the collaborative nature of the group. As > an instructor, it’s definitely a motivator to be able to help fellow > students in a direct way. > > All of this said, I’ve really appreciated everyone’s thoughts on the > challenge of making something like this sustainable and successful. I think > these are good things for us to actively reflect on as potential points > where this “ecosystem of practice" might collapse. > > Best, > Michael > > > -- > Michael Koontz — website <https://michaeljkoontz.weebly.com/> > ORCID: orcid.org/0000-0002-8276-210X > Pronouns: he/him > > Graduate Group in Ecology > Plant and Environmental Sciences > <https://www.google.com/maps/place/Plant+and+Environmental+Sciences+Bldg,+University+of+California,+Davis,+Davis,+CA+95616/@38.5427751,-121.7570251,16z/data=!4m2!3m1!1s0x808529085c693a13:0xe20b0046ee4eb4f7>, > Room > 2211 > University of California, Davis > Davis, CA 95616 > > > > > > On Oct 5, 2018, at 6:24 AM, Reid, Andrew C.E. via discuss (Fed) < > [email protected]> wrote: > > On Thu, Oct 04, 2018 at 11:16:19AM -0500, David Bapst wrote: > > April, All, > > And yet, people teach first-year algebra every year. > > > Yes, but... should they? Why do we require students to take algebra? > Why would we (possibly) require students to take a class on data > analysis, or (possibly) require that they attend a carpentry workshop? > Recently I attended a teaching workshop where we were told that we > needed to communicate to students what they were supposed to be > learning in each class, and that there was a crisis of transparency > where students want to know why they are taking the classes they are > required to take. I think such instructor transparency is extremely > important, but I think there's also crisis where institutions aren't > transparent about why degrees requires such-and-such a course. And I > think part of that is because many academics don't contemplate why > courses are required or not. > > > [ good stuff elided... ] > > > I don't know, maybe I'm making up a dichotomy between the exploration > of a subject versus technical skills that doesn't exist? > > But if such a dichotomy is real, than it makes sense for different > approaches to be ideal for maximizing benefit to the learners -- that > one pedagogical environment would not be ideal for both. And I think > that whether that dichotomy exists, or not, has implications for how > we think about where Carpentries exists in context with longer-term > courses in higher education. > > > > I think this dichotomy absolutely is real, but it manifests > strangely in pedagogical practice. In an ideal world, you could > teach "critical thinking and problem solving in [field]" on the one hand, > and "skills and best practces on [specific activity relevant to field]" > > But you can't *actually* do either of those. The given field's > broad approach to grasping its subject matter and approaching > problems are informed by the field's history, which is contingent > to some degree on the problems that the field faced initially, and > how it solved them. Likewise, it's hard to teach the skills needed > to solve specific problems without admitting that, much of the time, > the reason a particular skill is used in practice is because it's the > one that worked on other problems. > > In cartoon form, for example, Physics favors foundational laws and > totalizing models, from which the solutions to important problems can be > derived, because early physicists were trying to figure out ballistics > problems and the solar system and clocks, which share an underlying > stratum of mechanics and gravity. In similarly cartoon form, early > biology confronted a bewildering variety of lifeforms from trees to > insect colonies to charismatic megafauna to humans, Computer science > is more recent, but no less contingent -- I was recently reading a book > about this ("The Great Formal Machinery Works: Theories of Deduction > and Computation at the Origins of the Digital Age", fwiw), and was > reminded that advances in logic and computer science went together -- > Godel, Turing, and Russel were contemporaries. (Also, there is such > a thing as "advances in logic") > > All of which is to say, you can't teach critical thinking and > problem solving in physics without at least implicitly invoking > ballistics problems. > > I think the converse is also true. The skills needed to solve a > specific problem are embedded in the field's broader culture of practice. > Physicists solve rigid-body motion problems with conservation of energy > and momentum, because deriving specific solutions from foundational > principles is the Physics Way, whereas (more cartoony, bio is not my > field) biologists may determine a metabolic pathway in an animal by > finding some biochemical markers they've seen in nearby (in the Linnean > sense) animal, then proposing (and testing) that the new pathway is > analogous to the known one. > > So, unable to teach "the field" without examples or "the method" without > the history of the field, what you *actually* do is ... teach first-year > algebra every year. And workshop problem sets in the tutorial sessions. > Algebra is a particular sub-field, but you hope it's exemplary of the > broader mathematical approach, and that students don't get bogged down > in the specifics. The problem sets help hone Algebra-specific skills, > and might not have a lasting impact on students that don't stay in > mathematics, but you do it anyways because it's impossible to just teach > "math". > > For the Carpentries, we are of course in a time-constrained > corner at the workshoping-the-specifics end of this spectrum, but > we see the tension in the lessons. We promise learners primarily > that they'll be able to build software tools that will make them > more efficient at wrangling their data and getting results, and > secondarily that the tools will capture some of the capabilities > and allow this efficiency to persist across changes in personnel. > > These skills are built on a foundation of computer science, which > has a reductionist philosphy built in -- your data-wrangling task > is made up of smaller tasks, which can be expressed as a sequence, > possibly with some conditionals, and has a functional representation > in terms of code. You live in a Read-Execute-Print loop, and the > E of this REPL is where these tasks happen. The solution to your > problem is the correct identification, aggregation, and sequencing > of sub-tasks, and it manifests in the form of code. > > That last bit is our connection to the broader world. We > maybe sometimes don't see it, because it's the water we swim in, > and the vast majority of our learners are on-board with it, so it > generally doesn't need to be mentioned or acknowledged, but it's > still there. Learners who don't grok this will struggle, possibly > stuck on why our approach is so reductionist. Learners who come > to us without an immediate need might wish there was more emphasis on > this, maybe thinking that if we emphasized the principles more, they > could build their own solutions. > > -- A. > -- > Dr. Andrew C. E. Reid > Physical Scientist, Computer Operations Administrator > Center for Theoretical and Computational Materials Science > National Institute of Standards and Technology, Mail Stop 8555 > Gaithersburg MD 20899 USA > [email protected] > > ------------------------------------------ > The Carpentries: discuss > Permalink: > https://carpentries.topicbox.com/groups/discuss/Tad9e416c2ec4742e-Mf5263512c0f89869d3f018a7 > Delivery options: > https://carpentries.topicbox.com/groups/discuss/subscription > > > *The Carpentries <https://carpentries.topicbox.com/latest>* / discuss / > see discussions <https://carpentries.topicbox.com/groups/discuss> + > participants <https://carpentries.topicbox.com/groups/discuss/members> + > delivery > options <https://carpentries.topicbox.com/groups/discuss/subscription> > Permalink > <https://carpentries.topicbox.com/groups/discuss/Tad9e416c2ec4742e-Mecaf36b38cc2416c31039cbd> > ------------------------------------------ The Carpentries: discuss Permalink: https://carpentries.topicbox.com/groups/discuss/Tad9e416c2ec4742e-Mb012b09d1898a5ecff6d7b6b Delivery options: https://carpentries.topicbox.com/groups/discuss/subscription
