This is the place where research scientists, researchers in industry, software developers, and data scientists can communicate there novel solutions for problems in the subsurface. This reproducible journal is code agnostic. It is a place for people to communicate their analysis, tools, and solutions.
We accept submissions that address challenges within subsurface science. This includes all aspects of research from fundamental science to tools to solve exploration problems. The key is that the solutions presented are open and reproducible. To that end submissions are in the form of a Jupyter Notebook.
Submission is through this github repository. We require at a minimum two files, the notebook, and a file that lists the dependencies.
We accept Jupyter Notebooks only, as these are open source and code agnostic. We accept notebooks in any language supported within Jupyter. For example this could be Python, Julia, R, C++, Fortran, Octave, etc.
The notebook serves as the manuscript including figures and tables. Please structure the notebook as follows (we provide templates here):
The second file is a list of dependencies.
conda
please provide an environment.yml
file with the
dependenciesrequirements.txt
file.You do not need prior experience in running a notebook or coding. We will figure it out together.
Create a [github]((https://github.com) account.
Fork the Subsurface repository.
Create a branch (the branch name should be author names separated with dashes)
$ git checkout -b AUTHOR1-AUTHOR2-YEAR
Add your notebook, requirements, and any data required for reproducibility:
$ git commit -a -m "Some comment"
Push to github
$ git push origin AUTHOR1-AUTHOR2-YEAR
Issue a pull request (PR) to Subsurface with title containing author(s) name and follow the template that will appear once you opened the pull request:
**AUTHOR**
Dear @Subsurface/editors,
I request a review for the following notebook:
### Original article
**Title:**
**Author(s):**
**Year:**
### Potential reviewers
<!-- If you know potential reviewers, you can tell us here -->
You can suggest reviewers from editorial board.
Answer questions and requests made in the PR conversation page.
Once the pull request has been made the notebook will be executed and using nbval
it will automatically tested for reproducibility. If this check fails don't sweat, it
is probably due to the dependencies. We can help work it out. Subsequently the gang of
four will look at the notebook and solicit a minimum of two reviews and we will
itterate towards a reproducible article. Once passing review the notebook will be given
a unique DOI (via zenodo perhaps) and a
binder image will be generated. This will result in a
eternally reproducible article.
Jump right into the issues to add your ideas about how a journal should look like. We would like to hear about scope, feature, accessibility, peer review process, technology, licensing, possible journals in kind, sources of funding, and anything our ideas do not cover yet. You can open new issues and add to existing ones.
Please also check out the #journal channel on the Software Underground Slack. As per usual discussion here and in the community falls under our Code of Conduct.