SciPy Proceedings

This is the repository for submitting to and managing the Proceedings for the Annual Scientific Computing with Python Conference.

This repository is a home for authors, reviewers and editors to collaboratively create the proceedings for the conference.

You can find more information about the proceedings' organising principles below.

All communication between authors and reviewers should be civil and respectful. There are no exceptions to this rule. Please see the SciPy2019 Code of Conduct for more info.

You can find the schedule for 2019 below.

Please use @-mentions in issues and pull requests(PRs) to contact the proceedings Co-Chairs.

If you are an Author, please see Instructions for Authors.

If you are a Reviewer, please see Instructions for Reviewers.

If you are an Editor, please see Instructions for Editors.

If you are a Publisher, please see Instructions for Publishers.

Organising Principles: Openness

Overall, the SciPy proceedings are organised to be a fully open proceedings.

We aim to combine the best aspects of open source development, open peer review, and open access publication.

Built by and for Open Source Communities on Open Source Tech

The technologies used for running the conference are themselves developed in the open and built on open source tools.

Open Development:

• with many people contributing code over more than a decade
  • many contributors start as authors submitting to the proceedings
  • provides a natural pathway for new members to join the proceedings committee
• technologies are managed via public, open source GitHub repositories:
  • build system: https://github.com/scipy-conference/scipy_proceedings
  • server: https://github.com/scipy-conference/procbuild

The systems for running the conference are built on top of open source tools:

• build system:
  • LaTeX
  • ReStructured Text (reST)
  • Python: docutils, lxml, pygments, pytest
• server:
  • Flask & waitress
  • pyzmq
  • Docker
  • Python: asyncio

Open Peer Review meets Open Source Code Review

The entire submission and review procedure occurs through public PRs attached to identifiable individuals.

• Authors and reviewers are encouraged to work collaboratively to improve submissions throughout the review process, much like open source code-review.

• Reviews are collaborative, aiming to improve the publication quality. This is possible because the content was already vetted by the program committee.

• Conversations occur attached to people's real GitHub usernames and are open to the public.

  • This allows for a transparent open review process.
  • This holds authors and reviewers accountable and encourages civil communication practices.

Open Access for an Open Community

The papers are published as true Open Access (OA) articles with Creative Commons Attribution (CC By) license.

• There are no article processing charges barring authors from submitting papers.

  • Reviewers and co-chairs volunteer their time.
  • Services with free tiers (like GitHub and Heroku) allow distributing the underlying technologies with minimal cost.

• Papers are openly available at http://conference.scipy.org/proceedings/, with no pay walls barring consumption or author processing charges.

• From 2015 onward, papers have DOIs (making them easily citable) and are also openly available from those DOIs.

The community is involved in the entire process for creating the proceedings, which ensures relevance to the community that created them.

• Papers are submitted by authors who will be presenting talks and posters at the annual SciPy conference. Because we know the content is relevant to the SciPy community, review can focus on improving papers, not vetting them.

• Reviewers are invited by the editors, but community members may volunteer to review papers that interest them. The only barrier to participation is having a GitHub account.

Contacting the Proceedings Co-Chairs

The most effective way to contact the Proceedings Co-Chairs for issues related to this GitHub repository is to use GitHub's issues and "@"-mentioning the Co-Chairs.

In 2019, the Proceedings Co-Chairs are

• Chris Calloway (@cbcunc)
• David Lippa (@dalippa)
• Dillon Niederhut (@deniederhut)
• David Schupe (@stargaser)

Timeline for 2019

In addition to the following list, we break up the deadlines in the respective documents for authors and reviewers.

• April 2: Authors invited to submit full papers
• May 22: 1st Draft for Submission
• May 21–June 25: Open Review Period
• May 22: Reviewers Assigned
• June 11: Initial Complete Review
• June 25: Final Recommendation and Comprehensive Review Deadlines
• July 3: Final Editorial Decisions for Proceedings Contents Deadline
• July 10: Time Window for Publishing Conference Ready Proceedings

Instructions for Authors

Please submit your papers by 23:59 PST of the 1st Draft for Submission Deadline.

Submit your papers as an ReStructured Text file via PR against this repository.

During the Open Review Period authors should work with their reviewers to refine and improve their submission.

Proceedings Co-Chairs have final say in determining whether a paper is to be accepted to the proceedings.

Authors should respond to all the reviewers' comments.

Authors should default to modifying their papers in response to reviewers' comments.

Authors may not agree with the reviewers comments or may not wish to implement the suggested changes. In those cases, the authors and reviewers should attempt to discuss this in the PR's comment sections. It is important to remember in these cases that we expect all communication between authors and reviewers to be civil and respectful.

In the event that authors and reviewers are deadlocked, they should alert the Proceedings Co-Chairs to this situation. As always, the Proceedings Co-Chairs have final say in whether to accept or reject a paper.

Author Deadlines

• April 2: Authors invited to submit full papers
• May 22: 1st Draft for Submission
• May 21–June 25: Open Review Period
• July 3: Final Editorial Decisions for Proceedings Contents Deadline

General Information and Guidelines for Authors:

• Papers are formatted using reStructuredText.
• Example papers are provided in papers/00_bibderwalt and papers/00_vanderwalt.
  • These papers provide examples of how to:
    • Label figures, equations and tables
    • Use math markup
    • Include code snippets
  • 00_bibderwalt shows how to use a bib file for citations.
• For your paper to be found by the build system at http://procbuild.scipy.org your PR needs to have a title that begins with "Paper:". If you do not do this, the co-chairs will change your title on your behalf.
• All citations that have DOIs should include those DOIs in the paper's references section, see mybib.bib.
• All figures and tables should have captions.
• Figures and tables should be positioned inline, close to their explanatory text.
• License conditions on images and figures must be respected (Creative Commons, etc.).
• Code snippets should be formatted to fit inside a single column without overflow.
• Avoid custom LaTeX markup where possible.
• Do not modify any files outside of your paper directory.
• The compiled version of the paper (PDF) should be at most 8 pages, including figures and references.

Author Workflow

Below we outline the steps to submit a paper.

Before you begin, you should have a GitHub account. If we refer to <username> in code examples, you should replace that with your GitHub username.

More generally, angle brackets with a value inside are meant to be replaced with the value that applies to you.

For example, if your GitHub username was mpacer, you would transform

git clone https://github.com/<username>/scipy_proceedings

into:

git clone https://github.com/mpacer/scipy_proceedings

Author workflow steps

1. Get a local copy of the scipy_proceedings repo.
2. Update your local copy of the scipy_proceedings repo.
3. Create a new branch for your paper based off the latest 2019 branch.
   • If you submit multiple papers, you will need a new branch for each.
4. Set up your environment.
5. Write your paper, commit changes, and build your paper
6. Create a PR or push changes to your PR's branch and check your paper on http://procbuild.scipy.org.
   • If you want to alter the build system, do not include it in your submission's PR, create a separate PR against dev (see below for more details).
7. Repeat steps 5 and 6, while also responding to reviewer feedback.

Getting a local copy of the scipy_proceedings repo

• If you do not have a GitHub account, create one.
• Fork the scipy_proceedings repository on GitHub.
• Clone the repo locally
  • git clone https://github.com/<username>/scipy_proceedings
  • cd scipy_proceedings/
• Add the scipy-conference repository as your upstream remote
  • git remote add upstream https://github.com/scipy-conference/scipy_proceedings

If you run git remote -v you should see something like the following:

origin	https://github.com/<username>/scipy_proceedings.git (fetch)
origin	https://github.com/<username>/scipy_proceedings.git (push)
upstream	https://github.com/scipy-conference/scipy_proceedings.git (fetch)
upstream	https://github.com/scipy-conference/scipy_proceedings.git (push)

Getting the latest 2019 branch

• Fetch the latest version of the scipy_proceedings repo
  • git fetch upstream
• Check out the upstream 2019 branch
  • git checkout -b 2019 --track upstream/2019

Creating a new branch based off of 2019

If you are submitting only one paper, you can use the 2019 branch directly.

Otherwise, you will need to create a new branch based on 2019 and set its upstream to origin.

git checkout 2019
git checkout -b <your_branch_name>
git push --set-upstream origin <your_branch_name>

Setting up your environment

• Create a new environment (using your choice of environment manager, e.g., pyenv or conda).
• Install/update the required python libraries (pip install -U -r requirements.txt).
• Install LaTeX and any other non-python dependencies
• Create a new directory papers/<your_directory_name>
  • if you are submitting one paper, we recommend you use <firstname_surname>
  • if you are submitting more than one paper, you will need to use a different directory name for each paper

Write your paper

• Copy an example paper into your directory.
  • You must have only one reST file in the top level of <your_directory_name>.
• As you make changes to your paper, commit those changes in discrete chunks.

Commit your changes

• Commit any changes inside the paper/<your_directory_name>
• When you push your commits to your PR's branch, the paper will be autobuilt
• Do not commit any changes to files outside of your paper directory.

If you want to change the way the build system works, we use a separate submission procedure (see below).

Build your paper

• Run ./make_paper.sh papers/firstname_surname to make a PDF of your paper
• Check the output in output/<your_directory_name>/paper.pdf.
• Check that this output matches what you see on the build server.

Create a paper PR

• Once you are ready to submit your paper, make a pull request on GitHub. Please ensure that you file against the correct branch.
• Create a pull request against our 2019 branch.
• Do not modify any files outside of your paper directory. Create a separate PR for any changes to the build system.

Creating build system PRs

If you want to change the way the build system works, we use a separate submission procedure.

• Create a new branch against dev.
• Make your changes to the build system.
• Do not commit any changes from your paper PR to this new branch.
• Make a separate PR against the dev branch, it will be reviewed separately.

Push to your PR

When you push to your repositories branch it automatically updates the PR. This triggers a new build on the provided build server.

Check your paper's build

We encourage reviewers to review the PDFs built on our build server.

You should regularly check to see if the paper(s) that you build locally match the paper(s) that you see on the server.

If it is not the same, please immediately contact us with a GitHub issue describing the discrepancy. Please include screenshots and an explanation of the differences. For best results, please @-mention the Proceedings Co-Chairs.

Instructions for Reviewers

You will be reviewing authors' pull requests. While authors should have a proper draft of their paper ready for you by 1st Draft Submission deadline.

We ask that you read this set of suggested review criteria before beginning any reviews.

All communication between authors and reviewers should be civil and respectful at all times.

The goal of our review process is to improve the paper that the authors are working on. Our aim is to have you and the author collaborate on making their better by using an iterative process.

While our basic approach is to have you and the author iterate, we ask you to complete an initial review and start that conversation by the Initial Complete Review Deadline.

We ask that by the Final Recommendation Deadline you have a recommendation to either accept or reject the paper at that point and time.

Note: You many recommend changes after the Final Recommendation Deadline. If there are any major changes after the Final Recommendation Deadline you should immediately contact the Proceedings Committee Co-Chairs. As a heuristic, if you think the paper should not be in the proceedings unless the authors make the change in question, then that change should be requested and made before the Final Recommendation Deadline.

Reviewer Deadlines

• May 22: Reviewers Assigned
• June 11: Initial Complete Review
• June 25: Final Recommendation and Comprehensive Review Deadlines

Reviewer Workflow

• Read this set of suggested review criteria
• Click on the Pull Requests Tab and find the papers assigned to you
• After reading the paper, you can start the review conversation however you prefer
  • You can use line comments (on the paper itself) or high-level comments.
• Authors will respond to your comments, possibly via their own comments or by modifying their paper.
• This begins an iterative review process where authors and reviewers can discuss the evolving submission.
• By the Final Recommendation Deadline, we ask that you give two things
  1. A comprehensive review of the paper as it stands. This will act as the final review.
  2. A final recommendation to include the paper in the proceedings or not.
     • When you make the Final Recommendation, please contact the proceedings Co-Chairs in the PR in question.

Review Criteria

A small subcommittee of the SciPy 2017 organizing committee has created this set of suggested review criteria to help guide authors and reviewers alike. Suggestions and amendments to these review criteria are enthusiastically welcomed via discussion or pull request.

Requirements

• Install the requirements in the requirements.txt file: pip install -r requirements.txt
• IEEETran (often packaged as texlive-publishers, or download from CTAN) LaTeX class
• AMSmath LaTeX classes (included in most LaTeX distributions)
• alphaurl (often packaged as texlive-bibtex-extra, or download from CTAN) urlbst BibTeX style

Debian-like distributions:

sudo apt-get install python-docutils texlive-latex-base texlive-publishers \
                     texlive-latex-extra texlive-fonts-recommended \
                     texlive-bibtex-extra

Note you will still need to install docutils with pip even on a Debian system.

Fedora

On Fedora, the package names are slightly different:

su -c `dnf install python-docutils texlive-collection-basic texlive-collection-fontsrecommended texlive-collection-latex texlive-collection-latexrecommended texlive-collection-latexextra texlive-collection-publishers texlive-collection-bibtexextra`

Build Server

There will be a server online building open pull requests at http://procbuild.scipy.org.

Authors: you should check to ensure that your local builds match the papers built on this site. Please create an issue if they do not match.

Reviewers: You should be able to pull a built PDF for review from there.

For organisers

Instructions for Publishers

To information about how to manage the whole proceedings, please see publisher/README.md and publisher/Makefile.

Publisher Deadlines

• April 2: Authors invited to submit full papers
• May 21–June 25: Open Review Period
  • The build server should be maintained throughout the Open Review Period.
• July 10: Time Window for Publishing Conference Ready Proceedings

Instructions for Editors

As reviewers review papers, editors should apply labels to the PR to flag the current state of the review process.

• The labels in question are:
  • needs-more-review if the paper needs further review,
  • pending-comment if the paper is waiting on an authors' response, or
  • unready if the paper is not ready for the proceedings.

Editors should come to a final 'ready', 'unready' decision before the Final Editorial Decisions for Proceedings Contents deadline.

Editor Deadlines

• April 2: Authors invited to submit full papers
• May 21–June 25: Open Review Period
• May 22: Reviewers Assigned
• June 11: Initial Complete Review
  • Editors should verify that reviews have been completed
• July 3: Final Editorial Decisions for Proceedings Contents Deadline