Two-day in-person team planning and learning sessions

in-person
meeting
planning
learning
Agenda for our two-day in-person sessions.
Author

Luke W. Johnston

Published

October 23, 2025

Details

  • Team building, social activities, and dinners on October 23 to 24.
  • Main aims:
    • Team building: Some social, non-work-related activities.
    • Share some knowledge, skills, and tools with each other for some peer learning.
    • Align on strategy and roadmap.
    • Hear some details about the DP-Next collaboration.

Schedule

The schedule is meant as a guide only. We might spend more time on some topics and less on others.

October 23

Time Item
09:30 An overview and discussion of the strategy, roadmap, and collaborations
11:00 Lightning rounds to share tools and tips
12:00 Lunch
13:00 A presentation on a structured approach to writing code for software and analyses
14:00 ☕ Coffee break
14:10 Continuing from previous session
15:00 Break (Luke has a meeting at this time). Fussball?
16:00 Social activities (escape room challenge) and dinner

October 24

Time Item
10:00 Introduction to Denmark Statistics work in DP-Next
12:00 Lunch
13:00 Debrief from discussion meeting
13:15 Continue lightning rounds
15:00 Social activities (Ovartaci Museum)

Informal dinner afterwards.

Sessions

Overview and discussion of the strategy, roadmap, and collaborations

Review the overall strategy and roadmap again, especially in the context of our collaborations with ON-LiMiT and DP-Next as well as having Joel with us. We’ll make sure we’re all aligned in understanding what we want and hope to achieve in the next few months. We’ll also talk about the different “roles” and teams we have currently as well as in the future. Especially we’ll want to discuss how to prioritise effectively.

While this session is more informal, this is the general agenda:

  • Review/revisit strategy and roadmap.
  • Review tasks and projects in ON-LiMiT, including current repositories, work, and boards.
  • Review DP-Next, including repositories and GitHub, as well as potential/expected projects and tasks.
  • Share what we all would like to work on (mostly to hear from Joel).
  • Align what we want to work on and prioritise over the next ~6 months.
  • Review team design principles, as well as Team Topologies. (Didn’t get to this.)
  • Discuss what teams we’ll have, our roles in them, and how we’d like to work together with them. (Didn’t get to this.)

Lightning rounds to share tools and tips

  • Marton: Demo debugging in Python package using VS Code.

  • Kris: Show off REDCap and the data dictionary from it for ON-LiMiT.

  • Joel: Walk through what Vim is, some basic keybindings, and how to use it in the Terminal or VS Code.

  • Luke: Using gh dash, a GitHub CLI extension, to list PRs to review, approve, merge and so on, all from the CLI.

  • Luke: High level introduction to Rust, mainly to share some features, benefits, and simple examples of some code. Walk through personal learning package waltrs as an example.

A presentation on a structured approach to writing code for software and analyses

Luke will do this presentation that will last about 2 hrs and will include lots of time for informal questions and discussions. We’ll advertise to and invite others to attend, so the presentation will be targeted to a more general audience.

We’ll cover a basic introduction to functional programming for doing data analysis and building software. We’ll go into the conceptual and (simple) mathematical foundations of it in a programming language-agnostic way. Then I’ll show some examples in Python and R. Later in the session, I might show how it might look in a stricter, more advanced language like Rust.

Slides will be posted to the Community website (link will be added here later).

Introduction to Denmark Statistics work in DP-Next

Since we’re collaborating with DP-Next on tooling and some of the data side of things, the aim of this session is to provide information on Denmark Statistics (DST), how it relates to work in DP-Next and to Steno Aarhus in general, and how the Seedcase team would like to contribute in this area.

This session will be a 2 hour meeting and will have the following general agenda:

  • Luke: Introductions to this session, overall aim and desired outcomes.
  • Anders: Describing DST and demonstrating/showing how they organize their data (files, folders, format).
  • Luke: Explain what Parquet is, why it’s great, and what problems it solves in DST.
  • Kasper and Marie: Sharing any updates on what Steno Aarhus wants to do in this area with registers.
  • Kasper/Marie/Anders?: Explaining how StatTransfer works, give a demonstration of it, and expected way of using it.
  • Luke: Give overview of my code/package, what it does, and how, then compare it to StatTransfer.

Debrief from discussion meeting

This session is to debrief from our previous session, to get aligned with each other and discuss and decide on next steps.

Minutes

Overall, we enjoyed and thought the learning sessions (lightning round and functional programming talks) as well as the team building activities were useful and a good use of our in-person time together. The learning sessions were new this time, and we decided to continue including them in our in-person sessions.

Overview and discussion of the strategy, roadmap, and collaborations

  • We didn’t have time to cover everything during this session because of technical issues with the screen sharing setup.
  • We agreed that we want to continue focusing and prioritising work that revolves around building up the Seedcase software and collaborating on projects that want to use it.
  • We also would like to spend some time developing educational resources, including our guidebook, though we aren’t sure how much to prioritise this right now.

Introduction to Denmark Statistics work in DP-Next

  • We learned a lot more about what Steno Aarhus aims to do with the register data from Denmark Statistics (DST) and how DST organises and structures their files and folders. This also was relevant to better understanding how the “meta” register project, the Register Project Database, that Marie manages is structured too.
  • We determined that converting the registers2parquet package into a formal CRAN package would be useful to them and a good use of our time (as well as something we are interested in pursuing).