Add 'technical difficulty' summary notes

docs/talks/assets/20250904 - technical-difficulties-summary.md

@@ -0,0 +1,33 @@
+# 2025-09-04 Technical Difficulties Round Table - Summary
+
+- No general sense on how to optimise Python code
+- "Don't use for loops" - but not really knowing what the alternative would be
+    - numba decorators
+    - Olivier: Has worked with numba before
+- Submitting batch jobs
+    - How to estimate how much time it would take to run your code?
+        - Just be conservative and overestimate (it would only be used in scheduling)
+    - Not really clear how it works... Submitting jobs
+    - VScode integration to SLURM is unclear
+    - We can have a lunch talk on SLURM (and make sure that Michael K. is there so that he can hear our thoughts on SLURM and Lorenz).
+
+- JupyterHub on Lorenz?
+    - Not really worth it since people can easily just use VScode
+
+- if you open up a folder on Lorenz, how can you open other folders?
+    - Cmd+O: Open any file on your computer in vscode (regardless of in project)
+    - Open multiple folders in project workspace: https://stackoverflow.com/a/45125674
+- Don't know how to use AI coding tools in vscode, is it paid?
+    - GitHub Pro is available for all academics for free
+    - Saves time copy pasting things. Sometimes its not always accurate.
+- More advanced git: How can we resolve merge conflicts?
+    - There's a software carpentry course for git, with a section on collaboration ([link](https://swcarpentry.github.io/git-novice/08-collab.html)).
+- Xarray and performance isn't really clear...
+    - How to choose chunk size?
+    - Why is dask (e.g., doing a spatial average for each file) slower than doing a for loop through each file?
+
+Final takeaways:
+
+- VScode running in an interactive session would be useful on an IMAU level (and not just a Parcels level)
+- We should have a lunch talk on SLURM
+- An xarray talk series would be useful (also looping in performance discussions)

docs/talks/index.md

@@ -23,8 +23,6 @@ or, for upcoming talks,
 
 _Titles are placeholder._
 
-- 2025-09-04 Technical Difficulties Round Table
-    - **Abstract:** In our scientific workflows, we all encounter technical difficulties. How can I easily transfer data from a server to my computer? How can I easily iterate using a notebook on an HPC? How can I share data in a cross-institution collaboration? We'll use this session as a round table (or maybe "support group") where we can share our frustrations. Let's share our workflows, and voice the difficulties that we encounter so that we can learn from each other. This session will not be recorded so that people can share openly - an anonymised summary will be uploaded after discussion.
 - 2025-09-18 TBD
 - 2025-10-02 TBD
 - 2025-10-16 "Open code from a journal editor’s perspective" by Erik van Sebille [(:material-github: Discussion)](https://github.com/UU-IMAU/python-for-lunch/issues/18)
@@ -38,11 +36,6 @@ _Titles are placeholder._
 
 ## Prior
 
-- **2025-06-12 | "A Roadmap to the Pangeo Ecosystem" by Nick Hodgskin**
-
-    - [:octicons-video-16: Recording](https://youtu.be/iPh3KxxcAjA), [:material-presentation-play: Slideshow](./assets/20250612%20-%20Pangeo%20Roadmap/index.html), [:material-github: Discussion](https://github.com/UU-IMAU/python-for-lunch/issues/34)
-    - **Abstract:** We've all heard the name Pangeo - at a conference, or as a mention from the documentation of one of our tools - but what is it? Why are many climate scientists excited about it? It this talk we explore the Pangeo community and the ecosystem of packages it provides, going over its core mission and why it's needed in the next era of climate science where working with "big data" is becoming the norm. In this talk will give you a roadmap of the different packages and resources available in the Pangeo ecosystem - so that you can identify useful tools to investigate for your workflows and use-cases.
-
 - **2025-02-25 | "Git and GitHub: A time machine for every research project" by Nick Hodgskin**
 
     - [:octicons-video-16: Recording](https://youtu.be/QfojYVT01t0), [:material-presentation-play: Slideshow](./assets/20250220%20-%20Git%20and%20GitHub/index.html), [:material-github: Discussion](https://github.com/UU-IMAU/python-for-lunch/issues/11)
@@ -56,7 +49,7 @@ _Titles are placeholder._
 
 - **2025-03-27 | "Python Tidbits: Small Python tips, tricks, and packages you wish you knew about yesterday" by Nick Hodgskin**
 
-    - [:octicons-video-16: Recording](https://youtu.be/X-F0fFFW-wo),[:material-file-document: Notebook](docs/talks/python-tidbits.ipynb), [:material-download: Notebook Download](https://github.com/UU-IMAU/python-for-lunch/blob/main/docs/talks/python-tidbits.ipynb), [:material-github: Discussion](https://github.com/UU-IMAU/python-for-lunch/issues/21)
+    - [:octicons-video-16: Recording](https://youtu.be/X-F0fFFW-wo),[:material-file-document: Notebook](./python-tidbits.ipynb), [:material-download: Notebook Download](https://github.com/UU-IMAU/python-for-lunch/blob/main/docs/talks/python-tidbits.ipynb), [:material-github: Discussion](https://github.com/UU-IMAU/python-for-lunch/issues/21)
     - **Abstract:** Sometimes the most impactful, timesaving features of a programming language are small tips, tricks, and patterns that you pick up along the way. When you finally come across them, you're left thinking "if only I knew about this earlier!". In this talk, we'll go through many of these lesser-known Python tricks - most of which are built directly into the Python language or standard library – to supercharge your Python skills. This talk will be fast paced covering a wide range of features, focusing on "tidbits" that can be explained in a few minutes each.
 
 - **2025-04-17 | "Conda Environments: A tool for reproducible, collaborative, and portable research" by Nick Hodgskin**
@@ -69,4 +62,14 @@ _Titles are placeholder._
     - [:octicons-video-16: Recording](https://youtu.be/YRGjb4fEidg), [:material-presentation-play: Slideshow](./assets/20250515%20-%20VSCode/slides.pdf), [:material-github: Discussion](https://github.com/UU-IMAU/python-for-lunch/issues/29)
     - **Abstract:** As scientists, our projects span multiple languages, diverse config/data files, notebooks, local scripting and cloud or cluster work. In this hands-on workshop you’ll discover how Visual Studio Code brings all these workflows under one roof. We’ll kick off with a quick tour of VS Code’s core features (command palette, terminal, workspace management), then dive into essential settings and extensions. You’ll pick up time-saving tips & shortcuts (multi-cursor edits, snippets, Zen Mode), get a peek at remote development over SSH, and watch GitHub Copilot assist your coding with suggestions. Finally, as a real-world bonus, I’ll show you how I turned VS Code into my local LaTeX editor. By the end, you’ll have a personalized toolkit of settings, extensions, and workflows to supercharge your VS Code experience. Bring your laptop if you want to follow along!
 
+- **2025-06-12 | "A Roadmap to the Pangeo Ecosystem" by Nick Hodgskin**
+
+    - [:octicons-video-16: Recording](https://youtu.be/iPh3KxxcAjA), [:material-presentation-play: Slideshow](./assets/20250612%20-%20Pangeo%20Roadmap/index.html), [:material-github: Discussion](https://github.com/UU-IMAU/python-for-lunch/issues/34)
+    - **Abstract:** We've all heard the name Pangeo - at a conference, or as a mention from the documentation of one of our tools - but what is it? Why are many climate scientists excited about it? It this talk we explore the Pangeo community and the ecosystem of packages it provides, going over its core mission and why it's needed in the next era of climate science where working with "big data" is becoming the norm. In this talk will give you a roadmap of the different packages and resources available in the Pangeo ecosystem - so that you can identify useful tools to investigate for your workflows and use-cases.
+
+- **2025-09-04 | "Technical Difficulties Round Table"**
+
+    - [:material-file-document: Summary Notes](./assets/20250904%20-%20technical-difficulties-summary.md)
+    - **Abstract:** In our scientific workflows, we all encounter technical difficulties. How can I easily transfer data from a server to my computer? How can I easily iterate using a notebook on an HPC? How can I share data in a cross-institution collaboration? We'll use this session as a round table (or maybe "support group") where we can share our frustrations. Let's share our workflows, and voice the difficulties that we encounter so that we can learn from each other. This session will not be recorded so that people can share openly - an anonymised summary will be uploaded after discussion.
+
 - [Archived talks](https://github.com/UU-IMAU/Python-for-lunch-Notebooks/blob/main/archived)