JOSS Paper Format

Submitted articles must use Markdown and must provide a metadata section at the beginning of the article. Format metadata using YAML, a human-friendly data serialization language (The Official YAML Web Site, 2022). The information provided is included in the title and sidebar of the generated PDF.

What should my paper contain?

Important

Begin your paper with a summary of the high-level functionality of your software for a non-specialist reader. Avoid jargon in this section.

JOSS welcomes submissions from broadly diverse research areas. For this reason, we require that authors include in the paper some sentences that explain the software functionality and domain of use to a non-specialist reader. We also require that authors explain the research applications of the software. The paper should be between 250-1000 words. Authors submitting papers significantly longer than 1000 words may be asked to reduce the length of their paper.

Your paper should include:

A list of the authors of the software and their affiliations, using the correct format (see the example below).
A summary describing the high-level functionality and purpose of the software for a diverse, non-specialist audience.
A Statement of need section that clearly illustrates the research purpose of the software and places it in the context of related work.
A list of key references, including to other software addressing related needs. Note that the references should include full names of venues, e.g., journals and conferences, not abbreviations only understood in the context of a specific discipline.
Mention (if applicable) a representative set of past or ongoing research projects using the software and recent scholarly publications enabled by it.
Acknowledgement of any financial support.

As this short list shows, JOSS papers are only expected to contain a limited set of metadata (see example below), a Statement of need, Summary, Acknowledgements, and References sections. You can look at an example accepted paper. Given this format, a “full length” paper is not permitted, and software documentation such as API (Application Programming Interface) functionality should not be in the paper and instead should be outlined in the software documentation.

Important

Your paper will be reviewed by two or more reviewers in a public GitHub issue. Take a look at the review checklist and review criteria to better understand how your submission will be reviewed.

Article metadata

Names

Providing an author name is straight-forward: just set the name attribute. However, sometimes more control over the name is required.

Name parts

There are many ways to describe the parts of names; we support the following:

given names,
surname,
dropping particle,
non-dropping particle,
and suffix.

We use a heuristic to parse names into these components. This parsing may produce the wrong result, in which case it is necessary to provide the relevant parts explicitly.

The respective field names are

given-names (aliases: given, first, firstname)
surname (aliases: family)
suffix

The full display name will be constructed from these parts, unless the name attribute is given as well.

Particles

It’s usually enough to place particles like “van”, “von”, “della”, etc. at the end of the given name or at the beginning of the surname, depending on the details of how the name is used.

dropping-particle
non-dropping-particle

Literal names

The automatic construction of the full name from parts is geared towards common Western names. It may therefore be necessary sometimes to provide the display name explicitly. This is possible by setting the literal field, e.g., literal: Tachibana Taki. This feature should only be used as a last resort.

Example

authors:
  - name: John Doe
    affiliation: '1'

  - given-names: Ludwig
    dropping-particle: van
    surname: Beethoven
    affiliation: '3'

  # not recommended, but common aliases can be used for name parts.
  - given: Louis
    non-dropping-particle: de
    family: Broglie
    affiliation: '4'

The name parts can also be collected under the author’s name:

authors:
  - name:
      given-names: Kari
      surname: Nordmann

Internal references

The goal of Open Journals is to provide authors with a seamless and pleasant writing experience. Since Markdown has no default mechanism to handle document internal references, known as “cross-references”, Open Journals supports a limited set of LaTex commands. In brief, elements that were marked with \label and can be referenced with \ref and \autoref.

![View of coastal dunes in a nature reserve on Sylt, an island in
the North Sea. Sylt (Danish: *Slid*) is Germany's northernmost
island.](sylt.jpg){#sylt width="100%"}

Tables and figures

Tables and figures can be referenced if they are given a label in the caption. In pure Markdown, this can be done by adding an empty span []{label="floatlabel"} to the caption. LaTeX syntax is supported as well: \label{floatlabel}.

Link to a float element, i.e., a table or figure, with \ref{identifier} or \autoref{identifier}, where identifier must be defined in the float’s caption. The former command results in just the float’s number, while the latter inserts the type and number of the referenced float. E.g., in this document \autoref{proglangs} yields “\autoref{proglangs}”, while \ref{proglangs} gives “\ref{proglangs}”.

: Comparison of programming languages used in the publishing tool.

| Language | Typing          | Garbage Collected | Evaluation | Created |
|----------|:---------------:|:-----------------:|------------|---------|
| Haskell  | static, strong  | yes               | non-strict | 1990    |
| Lua      | dynamic, strong | yes               | strict     | 1993    |
| C        | static, weak    | no                | strict     | 1972    |

Equations

Cross-references to equations work similarly to those for floating elements. The difference is that, since captions are not supported for equations, the label must be included in the equation:

$$a^n + b^n = c^n \label{fermat}$$

Referencing, however, is identical, with \autoref{eq:fermat} resulting in “\autoref{eq:fermat}”.

$$a^n + b^n = c^n \label{eq:fermat}$$

Authors who do not wish to include the label directly in the formula can use a Markdown span to add the label:

[$$a^n + b^n = c^n$$]{label="eq:fermat"}

Markdown

Markdown is a lightweight markup language used frequently in software development and online environments. Based on email conventions, it was developed in 2004 by John Gruber and Aaron Swartz.

Inline markup

The markup in Markdown should be semantic, not presentations. The table below has some basic examples.

Markup	Markdown example	Rendered output
emphasis	`this`	this
strong emphasis	`that`	that
strikeout	`~~not this~~`	~~not this~~
subscript	`H~2~O`	H₂O
superscript	`Ca^2+^`	Ca²⁺
underline	`[underline]{.ul}`	underline
small caps	`[Small Caps]{.sc}`	Small Caps
inline code	`return 23`	`return 23`

Links

Link syntax is [link description](targetURL). E.g., this link to the Journal of Open Source Software is written as
[Journal of Open Source Software](https://joss.theoj.org/).

Open Journal publications are not limited by the constraints of print publications. We encourage authors to use hyperlinks for websites and other external resources. However, the standard scientific practice of citing the relevant publications should be followed regardless.

Grid Tables

Grid tables are made up of special characters which form the rows and columns, and also change table and style variables.

Complex information can be conveyed by using the following features not found in other table styles:

spanning columns
adding footers
using intra-cellular body elements
creating multi-row headers

Grid table syntax uses the characters “-”, “=”, “|”, and “+” to represent the table outline:

Hyphens (-) separate horizontal rows.
Equals signs (=) produce a header when used to create the row under the header text.
Equals signs (=) create a footer when used to enclose the last row of the table.
Vertical bars (|) separate columns and also adjusts the depth of a row.
Plus signs (+) indicates a juncture between horizontal and parallel lines.

Note: Inserting a colon (:) at the boundaries of the separator line after the header will change text alignment. If there is no header, insert colons into the top line.

Sample grid table:

+-------------------+------------+----------+----------+
| Header 1          | Header 2   | Header 3 | Header 4 |
|                   |            |          |          |
+:=================:+:==========:+:========:+:========:+
| row 1, column 1   | column 2   | column 3 | column 4 |
+-------------------+------------+----------+----------+
| row 2             | cells span columns               |
+-------------------+------------+---------------------+
| row 3             | cells      | - body              |
+-------------------+ span rows  | - elements          |
| row 4             |            | - here              |
+===================+============+=====================+
| Footer                                               |
+===================+============+=====================+

Figures and Images

To create a figure, a captioned image must appear by itself in a paragraph. The Markdown syntax for a figure is a link, preceded by an exclamation point (!) and a description.
Example:
![This description will be the figure caption](path/to/image.png)

In order to create a figure rather than an image, there must be a description included and the figure syntax must be the only element in the paragraph, i.e., it must be surrounded by blank lines.

Images that are larger than the text area are scaled to fit the page. You can give images an explicit height and/or width, e.g. when adding an image as part of a paragraph. The Markdown ![Nyan cat](nyan-cat.png){height="9pt"} includes the image saved as nyan-cat.png while scaling it to a height of 9 pt.

Citations

Bibliographic data should be collected in a file paper.bib; it should be formatted in the BibLaTeX format, although plain BibTeX is acceptable as well. All major citation managers offer to export these formats.

Cite a bibliography entry by referencing its identifier: [@upper1974] will create the reference “(Upper 1974)”. Omit the brackets when referring to the author as part of a sentence: “For a case study on writers block, see Upper (1974).” Please refer to the pandoc manual for additional features, including page locators, prefixes, suffixes, and suppression of author names in citations.

The full citation will display as

Upper, D. 1974. “The Unsuccessful Self-Treatment of a Case of “Writer’s Block”.” Journal of Applied Behavior Analysis 7 (3): 497. https://doi.org/10.1901/jaba.1974.7-497a.

Mathematical Formulæ

Mark equations and other math content with dollar signs ($). Use a single dollar sign ($) for math that will appear directly within the text. Use two dollar signs ($$) when the formula is to be presented centered and on a separate line, in “display” style. The formula itself must be given using TeX syntax.

To give some examples: When discussing a variable x or a short formula like

\[\sin \frac{\pi}{2}\]

we would write $x$ and

$\sin \frac{\pi}{2}$

respectively. However, for more complex formulæ, display style is more appropriate. Writing

$$\int_{-\infty}^{+\infty} e^{-x^2} \, dx = \sqrt{\pi}$$

will give us

$$\int_{-\infty}^{+\infty} e^{-x^2} , dx = \sqrt{\pi}$$

Footnotes

Syntax for footnotes centers around the “caret” character ^. The symbol is also used as a delimiter for superscript text and thereby mirrors the superscript numbers used to mark a footnote in the final text.

Articles are published under a Creative Commons license[^1].
Software should use an OSI-approved license.

[^1]: An open license that allows reuse.

The above example results in the following output:

Articles are published under a Creative Commons license []. Software should use an OSI-approved license.

Footnotes

An open license that allows reuse.

Note: numbers do not have to be sequential, they will be reordered automatically in the publishing step. In fact, the identifier of a note can be any sequence of characters, like [^marker], but may not contain whitespace characters.

Blocks

The larger components of a document are called “blocks”.

Headings

Headings are added with # followed by a space, where each additional # demotes the heading to a level lower in the hierarchy:

# Section

## Subsection

### Subsubsection

Please start headings on the first level. The maximum supported level is 5, but paper authors are encouraged to limit themselves to headings of the first two or three levels.

Deeper nesting

Fourth- and fifth-level subsections – like this one and the following heading – are supported by the system; however, their use is discouraged. Use lists instead of fourth- and fifth-level headings.

Lists

Bullet lists and numbered lists, a.k.a. enumerations, offer an additional method to present sequential and hierarchical information.

- apples
- citrus fruits
  - lemons
  - oranges

apples
citrus fruits
- lemons
- oranges

Enumerations start with the number of the first item. Using the the first two laws of thermodynamics as example,

0. If two systems are each in thermal equilibrium with a third, they are
   also in thermal equilibrium with each other.
1. In a process without transfer of matter, the change in internal
   energy, $\Delta U$, of a thermodynamic system is equal to the energy
   gained as heat, $Q$, less the thermodynamic work, $W$, done by the
   system on its surroundings. $$\Delta U = Q - W$$

Rendered:

If two systems are each in thermal equilibrium with a third, they are also in thermal equilibrium with each other.
In a process without transfer of matter, the change in internal energy, $\Delta U$, of a thermodynamic system is equal to the energy gained as heat, $Q$, less the thermodynamic work, $W$, done by the system on its surroundings. $$\Delta U = Q - W$$

Checking that your paper compiles

JOSS uses Pandoc to compile papers from their Markdown form into a PDF. There are a few different ways you can test that your paper is going to compile properly for JOSS:

GitHub Action

If you’re using GitHub for your repository, you can use the Open Journals GitHub Action to automatically compile your paper each time you update your repository.

The PDF is available via the Actions tab in your project and click on the latest workflow run. The zip archive file (including the paper.pdf) is listed in the run’s Artifacts section.

Docker

If you have Docker installed on your local machine, you can use the same Docker Image to compile a draft of your paper locally. In the example below, the paper.md file is in the paper directory. Upon successful execution of the command, the paper.pdf file will be created in the same location as the paper.md file:

docker run --rm \
    --volume $PWD/paper:/data \
    --user $(id -u):$(id -g) \
    --env JOURNAL=joss \
    openjournals/inara

Locally

The materials for the inara container image above are themselves open source and available in its own repository. You can clone that repository and run the inara script locally with make after installing the necessary dependencies, which can be inferred from the Dockerfile.

Behind the scenes

Readers may wonder about the reasons behind some of the choices made for paper writing. Most often, the decisions were driven by radical pragmatism. For example, Markdown is not only nearly ubiquitous in the realms of software, but it can also be converted into many different output formats. The archiving standard for scientific articles is JATS, and the most popular publishing format is PDF. Open Journals has built its pipeline based on pandoc, a universal document converter that can produce both of these publishing formats as well as many more.

A common method for PDF generation is to go via LaTeX. However, support for tagging – a requirement for accessible PDFs – is not readily available for LaTeX. The current method used ConTeXt, to produce tagged PDF/A-3.