Teaching basic lab skills
for research computing

The Endless Sprint

The recent Library Carpentry sprint exceeded my expectations, and then some!

Around 107 people signed up on our organising etherpad, either to work remotely or to contribute at one of the 13 sites in seven countries around the globe. That’s a big step up from the 2016 numbers of six sites and around twenty people working on the material.

The sprint was organised as part of the 2017 Mozilla Global Sprint. Library Carpenters comprised around one-sixth of that sprint workforce, and our more than 850 GitHub events - pull requests, forks, issues raised, commits and merges - outpaced the rest of the field by miles.

We used the sprint to amend, update, and extend the existing Library Carpentry lessons, get draft lessons on SQL and Python into better shape, and develop a new lesson on web scraping. We welcomed contributions from librarians and archivists, as well as from other information professionals, not to mention Software Carpentry Executive Director Jonah Duckles himself who worked on the git lesson alongside Gillian Elliott’s team in Otago.

We used our chatroom and this GitHub repository to organise work during the sprint. These were the issues we worked on, with lesson maintainers making themselves available to answer questions either via the chatroom or on Zoom (video call) sessions. The Zoom sessions sometimes ran for hours and proved to be an efficient channel for the different sites and people working remotely to network with others or to resolve tricky questions.

Greg Wilson popped up in one Zoom session and we also had a visit from Raniere Silva - both welcome presences at any Carpentries’ hackathon.

We used Zoom for the daily handovers as the baton passed to the next team coming online. The sun never set on the project as momentum moved from earlier risers New Zealand, into and across Australia to the Netherlands and the UK, then on to Canada and the US, and back again. US librarians truly committed to the sprint, with seven sites overall, some of whom were still going strong when I clocked in on Saturday morning (my time) to say thanks and goodbye (getting snapped in my pyjamas by Elizabeth Wickes in the process).

Librarians who had attended the instructor training in Portland run by Tim Dennis and me (half of the newly minted #TeamSpatula) were particularly well-represented. Portlanders - we love you guys! There were just too many of you to name here but more than half the Portland cohort sprinted, and some are still chiming in with contributions now. I will give a shout out to Scott Peterson for being not only a sprinter, but also a site organiser at UC Berkeley.

Sprinters were free to work on whatever lessons or issues they liked. In New Zealand and Utah, librarians worked through lessons to familiarise themselves with what for many was new material. What eventuated was not just lesson knowledge, but a sense of community, and a feeling of ‘buy-in’ that this was useful knowledge worth spreading further.

Non-coders were free to raise issues, correct typos, suggest fixes and devise scenarios to try to make the material more relevant. That this worked well could be seen in our email in-boxes: GitHub notification numbers were off the charts for the lesson maintainers.

Not that we are complaining - the level of engagement was just mind-bogglingly gratifying.

Led by Nora McGregor, the British Library had a team working on git, and Software Carpentry Steering Committee member Mateusz Kuzak led a big team at the national library of the Netherlands where the Python and SQL lessons got a solid working over. Owen Stephens and Carmi Cronje led the charge on OpenRefine, and Jez Cope worked on incubator lessons and built a page for reporting our many workshops. James Baker worked on the data intro lesson, helmed a lot of Zoom action, and provided a steadying hand and answered a lot of questions in the chatroom. James also made sure that issues around standardising README and CONTRIBUTING files across the many repositories were not forgotten.

Just as in 2016, the indefatigable Cam Macdonell was on board all hours, and Elizabeth Wickes distinguished herself by being blocked on GitHub - they mistook her hard work for something more spammy and sinister and closed her down for a few hours. Thanks to Thomas Guignard for the web scraping lesson we used as a basis for ours and to Lauren Ko for hacking away at it and running a site in Texas as well.

The Brisbane site hosted Richard Vankongingsveld who developed the new Python intro lesson. He made good use of Zoom to consult with sprinters in New Zealand and the Netherlands as the sprint got underway.

As for me, I baked two cakes, wrangled pull requests, worked on the git and web scraping lessons, reported on progress to Mozilla, dug people out of Git-holes, talked to people on Zoom, haunted the chatroom, answered queries, and matched people to tasks that needed doing. I also tweeted a lot.

All in all, it was a thoroughly rewarding two days, and it was truly sad to see it end. We are now into the consolidation phase, with a lot more work ahead.

HUGE thanks to everyone who took part. Great effort all round.

Never seen Library Carpentry? Here are the links:

There is a new Data Intro lesson specifically geared towards the needs of archivists:

An incubator lesson exists within a separate repository for tidying spreadsheet data.

Dialogue & Discussion

You can review our commenting policy here.