Peer Review via Gigantum

Posted by The Gigantum Team on May 18, 2020 5:16:07 PM

Launching an Execution Environment in Gigantum

This post is an overview for reviewers that are using Gigantum to inspect code for a manuscript.

Gigantum is a browser base application that integrates with Jupyter &  RStudio to streamline the creation and sharing of reproducible work in Python & R. 

The overall process for code review in Gigantum is very simple.  

  1. The Editor sends you an email with a link to a Gigantum Project.
  2. You click the link to go to the Project overview in the Gigantum Client. 
  3. You Launch Jupyter Notebook, Jupyter Lab, or RStudio to explore the code interactively (or alternatively peruse the Activity Feed) to verify figures, tables, statistics, etc.
  4. When you finish, communicate your recommendations to the Editor. 

Video Demonstration

The following video shows how easy it is.

Reviewer Experience-1

 

Exploring the Code

Gigantum supports work in Python and R, and the code will generally be available as a collection of Jupyter notebooks or R Markdown files. We set up an interactive Project for review through a web browser, so you don't have to download anything.

To explore the code, just do the following:

  1. On the Project Overview tab, click the large blue launch button.
    • Note: Depending on the Author's code, you can select from Jupyter (Lab or classic Notebook interface) for .ipynb notebook files or RStudio for .Rmd R Markdown files.
  2. The selected environment will open in a new browser tab providing a live execution environment.

If you have experience with the particular environment (Jupyter or RStudio), then this should be straightforward. Just inspect and run the code, and verify that it generates the results reported in the paper. Keep an eye out for any concerns you may have. Feel free to explore and add your own code - this is your personal, anonymous, isolated copy. If you want to EDIT existing code, we strongly recommend making a COPY of the code you wish to edit so you can compare with the author's original code!

If you don't have experience with the particular environment, then you can use the Activity Feed to inspect the code.

Using the Activity Feed

If you have never used Jupyter or RStudio, then running the code may not be comfortable or familiar. In this case, just use the Activity Feed to get an idea of whether the code works or not and what it does.

The Activity Feed is a linear history of what code was executed and what the inputs and outputs were. It is connected to the version history of the code.

To use the Activity Feed just do the following:

  1. Go to the Activity page in the application.
  2. Scroll down through the entries and click to expand the details for a given entry. 
    • Note: The Activity Feed is interactive. You can learn about it here or here.
    • Note: The Activity Feed can partially render table outputs, so to see the entire output you may have to interact with the actual code itself.

A Few Tips

Watch the video above.

You do not need to create an account or login. 

You do not need to download anything. 

You will need to manage your popup blocker.

If you alter the code, you can get back to the original code using the rollback feature.

Contact us anytime at support@gigantum.zendesk.com

Questions?

Learn more about Gigantum in our docs, Medium posts, or on this blog. You can also email us at hello@gigantum.com and follow us on Twitter at @gigantumscience.

 

Topics: Reproducibility, Open Science, Peer Review