Foundational Open Science Skills (FOSS) Lesson 2: Project Management & Intro to CyVerse

Date: 2023-02-02
Today's Lead Instructor: Tyson, Jeff
Today's Helpers: Michele, Jason, Tina
Course Website: https://cyverse-learning-materials.github.io/foss
Zoom Link: https://arizona.zoom.us/j/87386173667

Instant Feedback: (please complete before you leave class) Complete Form

Happy Groundhog Day!

Agenda

Warm-up (5 minutes):

Questions & Comments about Open Science left over from last week?

Introduction to Classical Project Management (30 minutes)

Discussion Q/A (20 minute Breakout)

Breakout Discussion

Now we will do a breakout discussion section to talk about overall project management.

Question "Give an example of a poorly managed project you were involved in (no real names please). What contributed to this feeling?"

Who has been involved in a poorly managed science project? += yes; -=no

-++++++

Why was it poorly managed?
Bad leadership? -+++
Unclear mission or goal? -+++++
Roles are murky? ++++++++
Other contributing factors? ++
- No defined tasks, time frame or role for team members
- Poor communication+++
- Logistics mismanaged +
- Somebody left the project midway and we couldn't reproduce their +
- Lack of respect, there was racism involved, professor used bad words, made others feel less or uncomfortable
- professor threatened students that if we did not complete the task he would fire us.+++++
- professor did not let us see the grant/budget to plan our experiments accordingly. Professor said it was none of our business. We are students and we should know our place.
- Professor relied on donations from his friends in industry to get materials for me to complete the work. Did not plan the budget well. Professor kept saying there was no money for purchasing materials. Professor expected me to pay for lab materials out of my own pocket. Professor did not let me see the budget and said that if I did not purchase the materials then I have no passion and I should leave the project (it was for my PhD).
- Difficulties with transfer of knowledge
- Project onboarded new technology without people who could operate it
- Lost sight of main project goals+

Question "Why do you think effective project management is important to Open Science?"

• $$ tax payer funded accountability

• Better products for open science

• Better team morale!

• DOCUMENTATION!!

• Realizing that project administration also is important and takes time/effort. +1
  Realistic setting of expectations for new team members
  Setting the culture for good record-keeping- maintaining lab notebooks, well-commented code, etc.
  Ensures project deliverables are aligned with project goals and objectives
  Maximizes resource allocation and utilization
  Define the tasks very clear for all the members of the research group +1
  Timeline accountability

Question "What are some limitations to you, your lab/group, or your domain?"

too many hats+++
Time to participate in different initiatives+
Results/oriented people
Team members involved in multiple projects+
Enough funds to keep everyone engaged and valued+

What hats do you wear?
Community Manager+++
I have an experience working as a community manager that was
Data Science Educator+++++
Data Scientist +
Data Analyst +
Data Engineer
Data Steward
Developer Advocate++
DevOps Engineer
Research Applications Manager (RAM)
Research Software Engineer+++++
Molecular biologist
Administrator
Research scientist (but have to be the data collector, data manager, software developer, data scientist, Data Science Educator, and many)

(Comment: hats often get shared more in smaller teams/organizations and/or you have to wear more hats) +

Breakout Room 1:

Breakout Room 2:

BioBreak (10 minutes)

• List of open data hosting repositories

Introduction to Research Objects (15 minutes)

Breakout (5 minutes)

Who has created a research object? ++++++

Where on the web does the object live?
GitHub +++
Gitlab ++
GitHub/ private server +++
HPC
ORNL DAAC
scholarworks
OSF

Publications / preprints bioarxiv
Data: Zenodo.org

CyVerse
Box folder +
Google Cloud +

Do you think someone could reproduce your research by accessing your research object?

Certinaly not++

The data is private- so this might be tough

In what case(s) might a research object not work for your research?

What would a research object look like for your research?

Tyson: I would create a GitHub repository with a folder structure for my /docs MkDocs markdown, /code (.py, .sh), /src (software) with a README.md and a LICENSE. I would keep my data somewhere else (CyVerse or use scripts to download the data on the fly).
Uploading trained models for reproducibility, while keeping the data private

What are your software requirements for creating a research object?

Tyson: GitHub+, Python (pip,conda)+, open source GIS (gdal, pdal, grass-gis, QGIS) -ex: https://github.com/tyson-swetnam/emsi

Luca: Python ('conda'), GitHub, Blender (as python module), Jupyter Notebook
Shashank: Python / PyTorch-geometric
R/RStudio+, RNotebook +, KnitR
SAS
STATA
PyTorch / Tensorflow / Cuda +
GCS
Python Virtual Environment

Breakout Room 1:

Breakout Room 2:

Introduction to CyVerse (30 minutes)

https://user.cyverse.org - User Portal (Account Creation; use academic email if you have one; if you have an ORCID, include it!)

https://learning.cyverse.org/what_is_cyverse/

Homework Assignment

1. Questions to ponder:

• How do the various elements of Project Management, both classical and RO, support or enable Open Science?
• How would you use the OS pillars to organize your research, your lab, or your governance doc?

2. Work through Self-Paced CyVerse Intro: https://cyverse-learning-materials.github.io/cyverse_mooc/
3. Read the Learning Objectives for Lesson/Week 3: https://foss.cyverse.org/03_managing_data/