Write the Documentation

When writing documentation, place a single sentence on each line. This makes it easy to move content around, and to spot long or fragmented sentences. This will also allow us to assign comments on a single sentence in GitHub for easier review and merging.

Sentences should be concise, easy to read, and written in the present tense. Refer to the reader in the second-person ("you"); this gives our docs a personable tone.

Avoid the use of abbreviations, especially those derived from Latin (for example, "e.g.," "i.e.," and "etc."). You should also avoid certain terms because of their cultural history (for example, "abort," "hang," "master," and "slave"). For more information, see the Linguistic Society of America’s Guidelines for Inclusive Language.

Make sure that there are no trailing spaces at the end of each line. A single space will automatically be added between each sentence when the docs are built; including trailing spaces affects this spacing.

Use commas to separate words and clauses in a simple series of three or more items (for example, "the mouse, the desktop, and the computer"). Always include a comma before the conjunction.

Text formatting conventions

The OpenNMS documentation uses the following conventions:

  • File names and paths are written as `poller-configuration.xml` (rendered as poller-configuration.xml).

  • Bold field names, button and icon names, and words that require strong emphasis using the *Bold* syntax (rendered as Bold).

  • Italicize words that require emphasis using the _Italics_ syntax (rendered as Italics).

  • Write command line commands and configuration file lines as `command` (rendered as command). Note that the contents will not be parsed.

  • You can force line breaks by including + at the end of a line, followed by a line break.

    Forced line break
    This is the first line +
    and this is a forced second line.

Always leave a blank line at the beginning and end of a new .adoc file. Start the content on line 2, setting a relative path to the images directory (if applicable) to allow image rendering on GitHub:

Set relative path to images directory
:imagesdir: relative/path/to/images/dir

Because curly braces ({}) are used for AsciiDoc attributes, everything inside the braces will be treated as an attribute. To include braces without setting attributes, you must escape the opening brace by typing \{. If you do not escape the opening brace, the braces and the text inside them will not be rendered in the final documentation build.

Headings and document structure

Each document begins with headings from level one (the document title) and an assigned ID.

Assign ID and document title
= The Document Title

When writing a document’s title, capitalize the following:

  • The initial letter of the first word in the title.

  • The initial letter in all principal words—that is, words that are not articles ("the," "an," "a"), conjunctions ("and," "but," "or"), or prepositions.

  • All prepositions with five or more letters.

  • Any preposition that is part of a compound verb (for example, "Backing Up Files").

  • Both words in a hyphenated compound (for example, "Year-End Processing"); exceptions are short articles, such as "of-the" in "Run-of-the-Mill."

Subsequent headings should use the following syntax:

Create subheadings
== Subheading one

... content here ...

=== Sub-subheading one

... content here ...

Make sure that headings are brief, specific, and descriptive (for example, "Install server software"). Headings should be capitalized as follows:

  • The initial letter of the first word in the heading.

  • The initial letter in all proper nouns.

In some cases, sections in the document also need to have assigned IDs; this depends on where they fit in the overall structure. If you want to have a link to specific content, that content must have an ID. A missing ID in a mandatory place causes the docs build to fail.

Use the target’s assigned ID to link to other parts of the documentation. To refer to a target’s assigned ID, use the following syntax:

Reference assigned ID

To use your target’s assigned ID in a document, simply write the ID in your text. For example, see <<target-id>>.

If you need to link to another document and want to modify the link text, use the following syntax:

Reference assigned ID with defined link text
<<target-it, link text that fits the context>>

To include an external link, use the following syntax:

Include external link with defined hyperlink text
http://www.opennnms.com[Link text here]

Links should have descriptive hyperlink text associated, and should avoid generic phrases (for example, "Click here"). This helps with search engine optimization.


Vertical lists provide clarity, emphasis, and order in documents, and can improve the visual impact when properly formatted. List items can be sentences, phrases, or single words. Sequential lists are numbered, and non-sequential lists are not.

When writing a list, follow these guidelines:

  • Introduce the list using a complete sentence, followed by a colon (:). Do not treat the list as an extension of this sentence; that is, do not use semicolons or commas to end items, and don’t insert "and" or "or" before the last item.

  • Capitalize the initial letter of the first word of each item in the list.

  • End all items with a period if one or more of the items contains a verb.

  • Use parallel structure for all items in a list.


Tables present structured information, and can improve the visual impact of a document when formatted properly.

When creating a table, follow these guidelines:

  • Capitalize the initial letter of all principal words in column headings—that is, words that are not articles ("the," "an," "a"), conjunctions ("and," "but," "or"), or prepositions.

  • In most cases, do not use end punctuation for column headings, with the exception of ellipses (…​) when the items in the column complete the phrase begun in the heading.

  • For items within a column (with the exception of the heading), use periods for whole sentences only (strings of words that include at least one verb).

In most cases, the "Description" column should appear immediately following the item being described.

Construct tables using the following syntax:

Construct table with three columns
[options="header", cols="1,3,1"]
| Parameter
| Description
| Default

| myFirstParam
| My first long description.
| myDefault

| mySecondParam
| My second long description.
| myDefault

This renders as follows:

Parameter Description Default


My first long description.



My second long description.


For tables that are made up of more than two columns, use a separate line for each cell’s content and include a blank line to separate rows, as in the code sample above.

For content that has required and optional elements, use the following syntax:

Construct table with required and optional elements
[options="header", cols="1,3,1"]
| Parameter
| Description
| Default


| myFirstParam
| My first long description.
| myDefault

| mySecondParam
| My second long description.
| myDefault


| myThirdParam
| My third long description.
| myDefault

Tables should have alt text associated with them; this allows screen readers to provide users with more context for the information being presented. Alt text should succinctly convey the table’s content and function, and should not be redundant. If it would be redundant, omit it.

Code snippets

You can include code snippets, configuration settings, or source code files in documentation. To enable syntax highlighting, provide the appropriate language parameter; this works for source code and configuration settings.

Use explicitly-defined code snippets as sparsely as possible. Code becomes obsolete very quickly, and directing to archaic practices is detrimental for users.

To include code snippets, use the following syntax:

Example 1. Include code snippet
example code here

If there is no suitable syntax highlighter for the language used, simply omit it, as in the previous example. The following syntax highlighters are available:

  • Bash, Console, or Shell

  • Groovy

  • Java

  • Javascript

  • JSON

  • Karaf

  • Properties

  • Python

  • SQL

  • XML

  • YAML or YML


Use admonitions to define specific sections such as Notes, Tips, and Important information. Use them sparingly to draw the reader’s attention to important text that may otherwise be overlooked.

Admonitions can include multiple lines of text by using the forced new line syntax (+, followed by a line break).

Remember to write the admonition type in full caps; it does not render properly otherwise.

There is no easy way to add new admonition types. Do not create your own for inclusion in the OpenNMS documentation.
Include a Note admonition
NOTE: This is a note.

A Note renders as follows:

This is a note.
Include a Tip admonition
TIP: This is a tip.

A Tip renders as follows:

This is a tip.
Include an Important note
IMPORTANT: This is an important hint.

An Important note renders as follows:

This is an important hint.
Include a Caution note
CAUTION: This is a cautionary note.

A Caution note renders as follows:

This is a cautionary note.
Include a Warning
WARNING: This is a warning.

A Warning renders as follows:

This is a warning.


Images may be useful to help explain and visualize complex problems, but they can clutter a document. When considering whether you should add an image to a document, determine whether the image itself is necessary by asking yourself if the reader is already looking at the software in question. Is there a button that is hard to find or a complicated screen that needs explanation? Additionally, consider how much of the image is text-based. Don’t insert images of tables or text that the reader must read to understand the rest of the document.

Minimize the use of screenshots. Include screenshots only to illustrate a concept that may be difficult to understand, or something that is not easy to locate in the UI.

All images share the same namespace. The best practice is to use unique identifiers for all image files. Image files should be in the .png format.

To include an image file in a doc, make sure that it resides in the appropriate ./images directory relative to the document that you are including it in (see the repository file structure section). Where possible, include the source file in the ./images directory as well; this allows other contributors to update it in the future.

Insert an image into a document using the following syntax:

.This is the image caption
image::docs/02_opennms-logo.png[opennms logo]

The image path for all images that you include is relative to the .adoc file where the image is referenced.

Images should have alt text associated with them. This allows screen readers to provide users with information about the image and its role in the document. Alt text should succinctly convey the image’s content and function.


You can use common attributes to automatically render certain text. Some of these include the following:

  • {version} - The current version of the OpenNMS software

  • {page-component-title} - The name of the product (Horizon or Meridian)

For a complete list of attributes, see the repository’s antora.yml file.

AsciiDocs also includes common attributes that can be used:

  • {docdir} - The document’s root directory.

  • {nbsp} - Inserts a non-breaking space.


A separate build of the OpenNMS documentation exists that includes comments. When comments are used, they are rendered with a yellow background.

This build doesn’t run by default, but after a normal build, you can use the make annotated command to create a local build. The resulting "annotated" docs render the full manual as a single page, allowing you to easily search for content.

To write a comment, use the following syntax:

Include a comment
// this is a comment

To write a comment block, use the following syntax:

Include a comment block
The note included here will still be processed, but will not be rendered in the output.
That said, missing includes here still break the build!

Comments are not visible in the normal build, and comment blocks will not be included in the output of any build.