September 20, 2019
Anna Sackmann, Science Data & Engineering Librarian
Scholarly
Workflows
Data Management
Citation Management
LaTeX with Overleaf
Data Analysis
Practices in
Reproducibility
Break down data management planning into three stages.
Before During After
1. Before you begin your research:
Describe the design.
Describe the data.
Data type: numerical, image, video, simulations, etc.
Data format: .txt, .csv, .jpeg, .tif, proprietary
How much and how often?
Who are you sharing the data with?
2. During your research
Follow the 3-2-1 rule.
3
2
1
Create 3 copies of your data.
On at least 2 types of storage.
With 1 copy offsite.
Storage options at UC Berkeley
File Naming: 3 Rules
1. Pick a system that will work for you and your collaborators.
2. Document this system in a readme.txt file.
3. Create a system that is comprehensive and achievable.
Examples:
When was it produced? Who produced it?
What is it for? What is it about?
What factors should be clear at-a-glance?
ISO 8601: 20191011
date_datatype_location_researcher: 20191011_AQI_94619_Sackmann
3. After your research
datadryad.org
zenodo.org
Activity:
Read the case study and select the applicable data planning attributes.
Make note of questions or anything you think should be changed.
http://bit.ly/TOMODEC
Questions:
Research Data Management Program
researchdata@berkeley.edu
http://bit.ly/TOMOslides
By asackmann