04 Call Digest - Githubissues

Hello @nmfs-openscapes/2022-noaa-afsc-fall-cohort !

Cohort Call 4 on open communities and coding strategies for future us brought out such a richness of contributions to our shared notes doc! There was real generosity in responses to the reflection prompt on how you would spend 2 hours a week to prioritize learning / implementing something (if you had the extra time): time to connect with and help colleagues as well as time to learn and organize. You’ve all been reflecting on how you would spend this time, and sharing strategies for protecting that time, advocating for it, and sharing it with others (coworking). Similarly, in the roll call prompt of something / someone that helped you in the last week or month, there was SO MUCH GRATITUDE expressed for specific support from colleagues!

We’d love to hear about your Pathways in Call 5, whether you’ve thought it through as part of a team, or you’ve done your own thing. If you’d like to present (3 mins talking + 2 for Q&A; no obligation to do this), please sign up in the Call 5 agenda. There are examples in the PathwaysExamples Google Folder. “Am I doing Openscapes right?” Yes! if you’re reflecting, asking questions, noticing, connecting, planning, and doing in a way that works for you now. This is foundational work.

PSA: Eli Holmes has created a NMFS OpenSci coworking calendar in which people can share coworking sessions or learning sessions that anyone in NMFS can join. To add an event, Create event > add guests “NMFS OpenSci Co-working” > Save and it will appear in the calendar.

adding to NMFS OpenSci coworking Google Cal

We’re all thankful for getting to work and learn alongside you,

Stef, Julie, Megsie, Em, Josh

Digest: Cohort Call 04 [ 2022-noaa-afsc-fall ]

Cohort Google Drive folder - contains agendas, video recordings, slides, pathways folder

nmfs-openscapes.github.io/2022-noaa-afsc-fall - cohort webpage

Goals: We discussed open communities and coding strategies for future us.

Task: Prepare your Pathways presentation. Have a Seaside Chat; prepare to present your Pathway work-in-progress on our final call

Slide Decks:

Open communities (slides)
Coding strategies (slides)

A few lines from shared notes in the Agenda doc, excerpted from all 4 NMFS Cohorts

Incremental progress. If you had 2 hours a week to prioritize learning/implementing something you’ve seen through Openscapes, how would you spend those 2 hours?

Building really good onboarding and guidance documentation for working with confidential data so that we can move forward with more momentum towards collaborative project repositories and use of github for version control and collaborative planning +1
Break it into (4) 30 min sessions to discuss the status of ongoing deliverables and issues within specific projects with coworkers within GitHub
I think I would spend the time really focusing on becoming more familiar with GitHub and RMarkdown/Quarto. +1+1
I would get materials together (that other people have already created) for other people in my division to learn some of the tools we learned in Openscapes and then set aside time for people to come and set things up together, and also for me to show how I implemented some of these practices +1
Using GitHub (desktop, maybe) to put files in one location for collaborators to update and edit in real-time. Multiple copies of files are so much work to manage! (back to what Rod Towell said the other week - we spend 50-80% of our time tidying data, or something like that. UGH!) +1
90+ minutes transferring code & data/updating code to connect fully to online/cloud resources and program management; 15-30 minutes learning or refining a new trick/tool to more elegantly meet a need identified in the course of my hackery or working with a colleague/mentee to advance our collective skills.
I’d go through my code for my current project and ask myself, “Could I give this to someone and have them understand what I’m doing?” If the answer is no (very likely), I would start working on including more comments and organizing things in a way that would help it all make more sense to someone new. This might also include sending it to a colleague to get their feedback about it and what did/didn’t make sense. Another part would be to keep the version control updated in github as well. Maybe look through Tidy Tuesday blogs. +1+1♥️
I would spend time to reach out to my program members to get them started using GitHub and increase collaboration among us. Also, to have data and old code into one place so that the bus factor won’t affect us in the future. +1+1
Figuring out how to minimize reliance on local, large files needed for running models, and improving documentation of progress/lessons learned on ongoing analyses/projects.
Tidying datasets so they can be shared with others. +1+1
I would continue watching Eli Holme’s tutorials. +1 +1+1+1+1
Dedicate time to improving coaching and team coordination skills (less tools and more people skills) +1
Learning RMarkdown/Quarto to get rid of email chains of document versions +1
IDEAS OF HOW TO ASK SUPERVISORS FOR SUPPORT TO PROTECT LEARNING TIME
- Provide us with spaces where we can teach other people in our research groups and/or share skills in existing forums for communication like staff meetings (still workshopping this one) +1
- Build “learning”/”teaching”/”supporting/elevating peers” into performance plans.
- This is tough for contractors/affiliates but I think we are all getting creative in finding “workshops” to attend (not training :) and interactive/working meetings with coworkers. +1
Strategies for creating time / advocating for time
- Setting up 1:1 coworking sessions with a colleague - we’re accountable to each other and it’s motivating to know someone else is there and we can unmute to say hi and checkin +1+1
- Google Keep to track many “to do”s on different projects; pomodoro method, recognizing that “slowing down to speed up” really helps (e.g., taking time to organize my notes, to do lists, etc will clear my brain and help me better focus) +1+1
- Making it part of team culture and weekly or monthly meetings as a team +1+1
- The seaside chats and coworking times have been good excuses to make time to check in with team members about existing projects.

Open communities. What communities are you already a part of? What would you like to learn with your research group? How does this fit with your Pathway?

Step 1 for Openscapes is having that psych safety: feeling comfortable asking a question and accepting help from someone so you don’t spend time reinventing. So a good conversation to have
getting to know colleagues, learn what they’re working on, who I can go to for help
how to be involved in open communities in some way but keep the work-life balance?
Paste error messages into google! Mindset: someone has encountered this before, no matter how in-the-weeds
https://owi.usgs.gov/R/index.html - a cool example of a government org managing their r packages
Topics at quarterly meetings tend to be California Central Valley-related, but talks/skills could be applied elsewhere and is a very helpful and beginner-friendly space: https://iep.ca.gov/Science-Synthesis-Service/Project-Work-Teams/Data-Science
How do I ask questions? Resources for R, but the principles apply broadly: https://openscapes.github.io/series/how-do-i/ask-questions.html

Coding strategies for future us. How do these coding strategies resonate with you / in your work?

Lots of valid points that will absolutely make things more manageable when sharing code across multiple users. Plays into the data hygiene points that were made last week by the folks in Alaska +1
Naming conventions and workflow organization slides are helpful for project management, in addition to coding applications +1+1
The idea of a “project-oriented-workflow” along with the practical setup (RStudio projects that are GitHub repos) really gave me a sense of control over my coding. Before that, I had my nested folders on my computer and R code was nested within those folders, and I could never run that code easily because I was never in the right filepath, etc. Now, I feel so much more agency with where my code is, how it’s running, what it needs. +1
Jenny Bryan’s 3 “naming things” rules is a great idea +1+1+1
Restarting R often makes sense. I have shot myself in the foot many times by not realizing I had an R library loaded. I also really need to get more consistent about my naming strategies - it’s not uncommon for me to have files and variables with a mix of dashes, underscores, and camel case in them! +1+1 +1
Makes total sense but I need more practice on setting relative paths . It’s also helpful to have ALL parameters as well as paths declared at the top of the script.+1
I haven’t done much (any?) coding, so this was a bit over my head, but I’m sure I’ll get into coding at some point, so these tips will be great to have!
Super guilty of the setwd method - Ive never had to share code before, and now that I am working more collaboratively this makes total sense!
- No need to feel guilty!!! It’s been working for you. And, it can evolve! +1
For a friendly and fact-filled intro to R & RStudio setup including here package, and how to organize your files, I love https://rladiessydney.org/courses/ryouwithme/ . short videos plus text +1
I wish I knew a lot of this when I was doing my PhD lol (+3)+100
Creating a package to accompany data analysis in a publication has forced me to be really careful and intentional with my code. +1 +1
- https://github.com/CrumpLab/vertical
- Sharing and organizing research products as R packages
I have spent so many hours debugging from variables in my working environment “infecting” my code. +1+1
I really appreciate learning the complications with the clear environment and setwd. I want to build skills and tools that are transferable, both to future me and others.
the idea of we’re all really smart, if you have trouble getting a script to run it’s because you’re not a good scientist. Not failure - better communication. Culture shift

Coworking (optional)

Wed Nov 23 - AFSC coworking, 10:00 -11:00 am PT.

It’s ok to use Coworking as your Seaside Chat time.

nmfs-openscapes / 2022-noaa-afsc-fall