owned this note   changed 3 years ago  
Published Linked with GitHub uk-carpentries/uk-carpentries/meeting-notes/meeting-2022-10-24.md

Meeting - 24 October 2022

Meeting Info

This is the UK Carpentry Community space for the UK Carpentry instructors, helpers and workshop coordinators (or anyone involved in training tech to researchers in general and outside of the UK) to get to know each other better, update the commmunity about developments, discuss issues and ideas, and encourage collaboration.

During these meetings, we will be conforming to The Carpentries Code of Conduct.

Meeting schedule: meetings happen on 4th Monday each month, 16:00-17:00 UK time, BST (UTC+1) or GMT (UTC+0) depending on the time of the year

Meeting details are shared via local-uk mailing list and The Carpentries community calendar.

Zoom URL: https://zoom.us/j/95360073649

Meeting Minutes

  • Chair: Mario Antonioletti
  • Timekeeper:
  • Notetaker: Phil Reed

Notetaker:

Agenda

  1. Assign notetaker & timekeeper
  2. Sign in & ice-breaker
  3. Review of actions
  4. Announcements
  5. Guest speaker (Eirini Zormpa) + Q&A
  6. Instructor Training checkout questions
  7. Wrap-up/AOB

Sign-in

Name/pronoun if you like/ institution / optionally put "(checkout)" if you are here for the Instructor Training checkout:

  1. Gemma Turon / she,her / Ersilia Open Source Initiative (checkout)
  2. Adam Milsom / University of Birmingham (checkout)
  3. Felix Shaw / Earlham Institute (checkout)
  4. Nigel Green / he, him / University of Manchester (checkout)
  5. Lucy Hinnie / she,her / British Library / (checkout)
  6. Tom Marlow / AstraZeneca / (checkout)
  7. Mario Antonioletti, EPCC/SSI/University of Edinburgh, he/him
  8. Eirini Zormpa / she,her / Alan Turing Institute
  9. Andrew Walker / he,him / University of Oxford / on a coach with 35 undergraduates returning from a field trip (I'll be silent)
  10. Colin Sauze / he,him / Aberystwyth University+Supercomputing Wales
  11. Nilani Ganeshwaran / she,her / University of Manchester
  12. Phil Reed / he,him / University of Manchester
  13. Sarah Jaffa / she / University College London
  14. Winfred Gatua / she, her / University of Bristol
  15. Danielle Sieh / she,her,hers / The Carpentries
  16. Jeremy Cohen / he,him / Imperial College London
  17. William Graham / he,him / University College London
  18. Evgenij Belikov / he,him / EPCC, University of Edinburgh
  19. Lucia Michielin / she, her / University of Edinburgh Centre for Data, Culture & Society
  20. Matthew Bluteau / he,him / UK Atomic Energy Authority /
    Image Not Showing Possible Reasons
    • The image file may be corrupted
    • The server hosting the image is unavailable
    • The image path is incorrect
    • The image format is not supported
    Learn More →
    @mattasdata
  21. Aleks Nenadic, she/her, SSI

Notes

Announcements

SSI's Collaborations Workshop 2023

Meeting in Manchester. Hybrid event. Tuesday 2 May - Thursday 4 May 2023.

Instructor sign-up in AMY

Danielle is available for questions.

Guest speaker (Eirini Zormpa) + Q&A

ezormpa@turing.ac.uk

Training for community building at the AIM RSF. {Intro, Training needs, Training menu, questions.}

Intro

AIM - AI for Multiple Long-term Conditions. NIHR awarded £23m to AIM to establish research. Seven research consortia. Collaborations with other universities, NHS Trust, hospital databases. Have mixed group of people with Data Science, AI, medical backgrounds.

AIM Research Support Facility (AIM RSF) is a dedicated facility at Turing institute. For open, inclusive and collaborative practice. Five themes (reproducible, accessible, community, patient engagement, sustainability/legacy).

Training needs

Training underlies goal of fostering community of researchers. Can't expect people to be experts on all areas. Challenge: everyone needs

  1. Open research and reproducibility
  2. Open collaboration

Some already have expertise on data/AI or health reserach.

Gather needs from (1) project managers, (2) PIs/ECRs, (3) everyone in the consritia with a survey.

Training needs: Sharing code, interdisc collab, science communication, algorithm biases. Survey has 28 respondants, mostly ECRs.

  • 68% have not had any training on open research and reproducibility. Most requested was reproducible workflows, also FAIR. Very few wanted open access training.
  • 62% have had data science/AI training. Few wanted version control (may already have it, may not understand they need it).
  • 68% have not had biomedical data training. Want AI ethics in healthcare most.
  • 68% have not had collab/comms training. Most wanted GitHub and Markdown.

Training "menu"

2-3 hour workshops once a week, no full-day workshops. Keen to get project managers on GitHub too.

Git menu:

  1. Intro to version control with git
  2. Intro to GitHub
  3. GitHub for Collaboration
  4. GitHub for Project Management

R menu (R was most requested language), based on Data Carpentry:

  1. Intro to R
  2. Getting started with data in R
  3. Data wrangling
  4. Data visualisation
  5. Literate programming

Questions?

Questions from Eirini:

  • Any tips on resources on teaching programming to health researchers?
  • How do you find out what people want to learn?
    • The million dollar question! Some overlap with other subjects. Track conversations via email.
  • And how do you leverage that against what you thinkg you should teach if different?
    • Sometimes regular statistical methods can answer questions which are ignored/missed by people who know the AI methods.
  • Any tips on getting people engaged in helping/instructing?
    • MB: we didn't get to this in the discussion, so I will add my two cents. In my team, there are very obviously people who enjoy doing training and those who don't. So, first point of order is to find people who like teaching/instructing. I found this out gradually through just talking face-to-face with team members. Body language made it pretty obvious who wasn't keen on instructing/helping a course. That will give you a solid core you can depend on. Next, you will need to think about having some sort of pipeline of helpers/instructors coming in, because your core of people who like teaching will gradually deplete, and there might just not be enough of them to begin with. I tend to target Early Career RSEs because that makes sense within the context of my team, but more broadly Early Career people tend to be looking for opportunities to develop themselves and haven't formed too many calcified opinions about what they don't want to do. I gently nudge EC-RSEs towards helping on a course or two, and then next towards co-instructing. This has ensured the training numbers in my team have stayed fairly healthy. I still have to do a bit of arm twisting here and there, but much less than if I didn't do any of the above.

Questions/comments for Eirini:

Instructor Training checkout questions

Next meeting

28 November 2022, 16:00 UK time https://hackmd.io/Q0tIfE1qSf6kBvZLxnsOkA

Select a repo