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 NumFOCUS Code of Conduct for more info. Attendees at SciPy 2024 are subject to the NumFOCUS Code of Conduct.
You can find the schedule for 2024 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.
If you are Submitting Slides, please see Instructions for Slides.
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.
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
The systems for running the conference are built on top of open source tools, including:
- MyST Markdown (mystmd.org)
- Typst - for fast PDF generation (e.g. SciPy template)
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.
The papers are published as true Open Access (OA) articles with Creative Commons Attribution (CC-BY-4.0) 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) allow distributing the underlying technologies with minimal cost.
-
Papers are openly available at https://proceedings.scipy.org, with no pay walls barring consumption or author processing charges.
-
From 2010 onward, papers have DOIs (making them easily citable) and are also openly available from those DOIs.
-
From 2023 onwards, full HTML is the preferred format in addition to the PDF being available.
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.
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 2024, the Proceedings Co-Chairs are:
- Meghann Agarwal (@mepa)
- Amey Ambade (@ameyxd)
- Chris Calloway (@cbcunc)
- Rowan Cockett (@rowanc1)
- Sanhita Joshi (@sanhitamj)
- Charles Lindsey (@cdlindsey)
- Hongsup Shin (@hongsupshin)
In addition to the following list, we break up the deadlines in the respective documents for authors and reviewers.
- Apr 9: Reviewer invitations sent
- Apr 23: Deadline to respond to offer to be a reviewer
- Apr 26: Authors invited to submit full papers
- May 3: Webinar offered to authors
- Jun 7: Deadline to submit first draft by authors
- Jun 8: Assignment of reviewers to papers
- Jun 8: Open Review Period begins
- Reviewers comment on papers to authors during this period.
- Authors also respond to review comments with improvements to papers during this period.
- Jul 3: Initial complete review
- Reviewers continue to comment on paper improvements during this period.
- Authors also respond to review comments with further improvements to papers during this period.
- Aug 19: Final review deadline
- Authors continue to make revisions in response to final review comments during this period.
- Sept 2: Final author revision deadline
- Sept 2: Open Review Period ends
- Authors put down their pens.
- Reviewers make an up or down decision on publication readiness of papers during this period.
- Sept 9: Final reviewer decision deadline
- Sept 23: Proceedings final sign-off by editors
- The publication process begins after final sign-off.
Please submit your papers by 23:59 PST of the Deadline to submit first draft.
Submit your papers as a MyST Markdown (mystmd.org) or
LaTeX file via PR against this repository.
Please only use LaTeX if you are already familiar with writing papers in LaTeX.
The build process are using the mystmd
CLI in 2024, which allows us to support
a web-first reading experience.
In future years this will allow us to accept notebooks and computational
environments, however, this is not available in 2024.
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.
An excellent webinar entitled "SciPy Proceedings 2024: Quickstart and authoring tutorial" is available on YouTube.
If you have a challenge with any technical aspect of authoring your paper in MyST or LaTeX, please do not hesitate to reach out via your GitHub pull request or issue on this repository. A member of the Proceedings Co-chairs will help you directly or identify a work-around.
- Apr 26: Authors invited to submit full papers
- May 3: Webinar offered to authors
- Jun 7: Deadline to submit first draft by authors
- Reviewers comment on papers to authors during this period.
- Authors also respond to review comments with improvements to papers during this period.
- Sept 2: Final author revision deadline
- Authors put down their pens.
- Papers are formatted using MyST (mystmd.org) or LaTeX (which also uses MyST, please see notes on LaTeX below)
- The paper is written and reviewed using the interactive HTML view (i.e.
myst start
), the PDF is built upon acceptance only - Example papers are provided in
papers/00_myst_template
andpapers/00_tex_template
- These papers provide examples of how to:
- Label figures, equations and tables
- Use math markup
- Include code snippets
- Use a BibTeX files and/or DOIs for citations
- These papers provide examples of how to:
- When creating your pull-request, add a pull-request label of
paper
to trigger the build process. If you do not add this, a proceedings chair member will add it for you. - Authors may include a project or consortium (e.g. The Jupyter Project)
- There must be at least one corresponding author, and this must be a specific person with a valid email address
- Authors of papers from previous SciPys may change their name on their published work by contacting the Proceedings Co-chairs
- 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 close to their explanatory text.
- All abbreviations should be identified in your
myst.yml
(docs) - License conditions on images and figures must be respected (Creative Commons, etc.)
- Images and figures should be reasonably sized and formatted for viewing online; typically less than 1 MB
- Do not modify any files outside of your paper directory
- When using the LaTeX option, please consider:
- SciPy is supporting HTML. LaTeX is not involved in reading or rendering (as of 2024 we use Typst for building PDFs)
- Custom LaTeX macros are not supported and some packages may not be supported
- The compiled version of the paper should be at most 6000 words
including figures but not including references; this is about 8 pages for the published PDF that will be created upon acceptance.
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 you typically clone using the web URL, and your GitHub username was mpacer
, you would transform
git clone <scheme>github.com/<username>/scipy_proceedings.git
into:
git clone https://github.com/mpacer/scipy_proceedings
Note
There is a webinar on YouTube that goes through the author submission process for 2024 submissions using MyST Markdown.
- Get a local copy of the
scipy_proceedings
repo. - Update your local copy of the
scipy_proceedings
repo. - Create a new branch for your paper based off the latest
2024
branch.- If you submit multiple papers, you will need a new branch for each.
- Install MyST Markdown and Node and copy a template.
- Write your paper, commit changes, and build your paper
- Create a PR or push changes to your PR's branch and check your paper.
- If you want to alter other parts of the
scipy_proceedings
repo, do not include it in your submission's PR, create a separate PR againstdev
(see below for more details). - Creating build system PRs is deprecated in 2024. Curvenote is the build system now.
- If you want to alter other parts of the
- Repeat steps 5 and 6, while also responding to reviewer feedback.
- If you do not have a GitHub account, create one.
- Fork the scipy_proceedings repository on GitHub.
- Clone the repo locally
- replace
<scheme>
withgit@
orhttps://
, for example - replace
<username>
with your GitHub username git clone <scheme>github.com/<username>/scipy_proceedings.git
cd scipy_proceedings/
- replace
- Add the
scipy-conference
repository as yourupstream
remotegit remote add upstream <scheme>github.com/scipy-conference/scipy_proceedings
If you run git remote -v
you should see something like the following:
origin <scheme>github.com/<username>/scipy_proceedings.git (fetch)
origin <scheme>github.com/<username>/scipy_proceedings.git (push)
upstream <scheme>github.com/scipy-conference/scipy_proceedings.git (fetch)
upstream <scheme>github.com/scipy-conference/scipy_proceedings.git (push)
- Fetch the latest version of the
scipy_proceedings
repogit fetch upstream
- Check out the upstream
2024
branchgit checkout -b 2024 --track upstream/2024
If you are submitting only one paper, you can use the 2024
branch directly.
Otherwise, you will need to create a new branch based on 2024
and set its
upstream to origin.
git checkout 2024
git checkout -b <your_branch_name>
git push --set-upstream origin <your_branch_name>
- Optional: Create a new environment (using your choice of environment manager, e.g.,
pyenv
orconda
). - Install MyST Markdown from mystmd.org
pip install mystmd
- Install
nodejs
(see options)
- 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
- if you are submitting one paper, we recommend you use
- Copy an example paper into your directory: either
papers/00_myst_template
orpapers/00_tex_template
- Update the
id
in themyst.yml
to byscipy-2024-<your_directory_name>
- Update the
- To have a live preview of your changes:
- Change directories
cd papers/<your_directory_name>
- Run
myst start
and open the web-server provided
- Change directories
- Refer to the syntax in the template papers or online at mystmd.org
- Update the author information and affiliations in
myst.yml
- As you make changes to your paper, commit those changes in discrete chunks
- If you come across any challenges, ask the Proceedings Co-chairs for help via a GitHub issue or comment on your PR
Note: The templates are setup for a single MyST/LaTeX file in the top level of <your_directory_name>
. If you have more than one file run myst init --write-toc
(docs), ensuring that the root
is the main file of your manuscript.
- Commit any changes inside the
papers/<your_directory_name>
- When you push your commits to your PR's branch, the paper will be auto-built in GitHub actions
- Do not commit any changes to files outside of your paper directory
If you want to alter other parts of the scipy_proceedings
repo, we use a separate
submission procedure (see below).
Your paper will be edited and reviewed in HTML, the PDF will only be built on acceptance.
To preview your paper:
- Ensure
mystmd
is installed (guide) - In
papers/<your_directory_name>
runmyst start
- Open the web-server from your console
- Check that this output matches what is built on your 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 the
2024
branch - Do not modify any files outside of your paper directory. Create a separate PR for any changes to the build system.
- Ensure that your PR title begins with
Paper:
. Note: for the first commit in your PR, an editor will add thepaper
label, which will start the GitHub actions.
Creating build system PRs is deprecated in 2024. Curvenote is the build system now.
If you want to change documentation, etc., we use a separate submission procedure.
- Create a new branch against
dev
- Make your changes
- 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
When you push to your repositories branch it automatically run GitHub actions on the PR. Note that this will require authorization for your first commit only. The build process takes about a minute, and then posts or updates a comment on the PR with a link to the build result on Curvenote. The build page has a link to your preview.
The review process will be completed on the HTML, and you can check to see if the paper(s) that you preview locally match the paper(s) that you see online. These will be available in a GitHub comment or through the logs in the GitHub action.
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.
We are interested in working towards full support for publishing computational notebooks as part of the proceedings, and are trialing this part of the submission process for interested authors - please get in touch with the Proceedings Co-Chairs with your interest.
You will be reviewing authors' pull requests. While authors should have a proper draft of their paper ready for you by the Deadline to submit first draft.
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 Reviewer Decision 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 Reviewer Decision deadline. If there are any major changes after the Final Reviewer Decision 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 Reviewer Decision deadline.
- Apr 9: Reviewer invitations sent
- Apr 23: Deadline to respond to offer to be a reviewer
- Jun 8: Assignment of reviewers to papers
- Reviewers comment on papers to authors during this period.
- Authors also respond to review comments with improvements to papers during this period.
- Jul 3: Initial complete review
- Reviewers continue to comment on paper improvements during this period.
- Authors also respond to review comments with further improvements to papers during this period.
- Aug 19: Final review deadline
- Authors continue to make revisions in response to final review comments during this period.
- Sept 2: Final author revision deadline
- Authors put down their pens.
- Reviewers make an up or down decision on publication readiness of papers during this period.
- Sept 9: Final reviewer decision deadline
- Read this set of suggested review criteria
- Click on the Pull Requests Tab and find the papers assigned to you
- A comment at the top of the PR will have a link to the paper to review online
- After reading the paper online, you can start the review conversation however you prefer
- You can use in-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 Reviewer Decision deadline, we ask that you give two things
- A comprehensive review of the paper as it stands. This will act as the final review.
- 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.
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.
- MyST Markdown (
mystmd
) and NodeJS (>18) - GitHub actions for the build process
The build process is completed through GitHub actions on every commit. A comment is posted after the build process completes with a list of checks and a link to the built output on Curvenote.
Authors: you should check to ensure that your local builds match the papers built online. Please create an issue if they do not match.
Reviewers: You should be able to see the built article from the GitHub comment, and review from the preview link.
To information about how to manage the whole proceedings, please see
publisher/README.md
and publisher/Makefile
.
- Apr 26: Authors invited to submit full papers
- The build process is supported by Curvenote (a SciPy sponsor) and it is maintained throughout this period.
- Sept 23: Proceedings final sign-off by editors
- The publication process begins after final sign-off.
As reviewers review papers, editors should apply labels to the PR to flag the
current state of the review process. All paper PRs must have the paper
label before the GitHub action will be triggered. Additionally, as editors and reviewers are assigned, the editors should add the reviewers GitHub handles to the PR summary comment.
Other labels that should be used 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.
- Apr 9: Reviewer invitations sent
- Apr 23: Deadline to respond to offer to be a reviewer
- Apr 26: Authors invited to submit full papers
- Jun 8: Assignment of reviewers to papers
- Reviewers comment on papers to authors during this period.
- Authors also respond to review comments with improvements to papers during this period.
- Jul 3: Initial complete review
- Reviewers continue to comment on paper improvements during this period.
- Authors also respond to review comments with further improvements to papers during this period.
- Editors should verify that reviews have been completed
- Sept 23: Proceedings final sign-off by editors
- The publication process begins after final sign-off.
- Get a local copy of the
scipy_proceedings
repo. - Update your local copy of the
scipy_proceedings
repo. - Create a new branch for your paper based off the latest
2024
branch. - Inside the
presentations
folder, there are directories for:- 3-minute lightning talk slide decks (lightning)
- Posters presented at the poster session (posters)
- 30-minute talk slide decks (slides)
- SciPy tools plenary slide decks (tools)
- Choose the appropriate folder, and make a new directory inside it (it needs a unique name)
- Copy your slide deck or poster into the directory, and add a file called
info.json
with the following fields needed for uploading to Zenodo (using an empty string for author orcid or affiliation if these cannot be provided):
{
"title": "The title of your presentation",
"authors": [
{
"name": "The first author or presenter",
"affiliation": "first author's affiliation",
"orcid": "0000-0000-0000-0000"
},
{
"name": "The second author or presenter",
"affiliation": "second author's affiliation",
"orcid": "0000-0000-0000-0001"
}
],
"description": "1-4 sentences explaining what your presentation is about"
}
You can see examples of submissions in the example
folder in each presentation directory.