COAR Notify

These are the sources for the Notify documentation at https://coar-notify.net (was previously at https://notify.coar-repositories.org)

This GitHub repository is also intended to be a "one-stop shop" for all Notify-related development discussions, issues, and documentation.

• If you have a question or feature suggestion, or simply want to join the (mostly technical) discussion(s) about COAR Notify then please start with the discussions
• If you have a definite issue (e.g. a bug or documentation error) then please raise an issue.

We try to serve all technical documentation for COAR Notify on the website at https://coar-notify.net/. This includes:

• COAR Notify Protocol Specification
  • Notification Patterns
  • Vocabulary
• A catalogue of implementation systems & platforms
• Documented workflows using COAR Notify patterns
• Implementation Guide

About COAR Notify

The COAR Notify Initiative is developing and accelerating community adoption of a standard, interoperable, and decentralised approach (using Linked Data Notifications) to link research outputs hosted in the distributed network of repositories with resources from external services, such as overlay-journals and open peer review services.

Funding acknowledgement

COAR is extremely grateful for the generous funding support from Arcadia, a charitable foundation that works to protect nature, preserve cultural heritage and promote open access to knowledge. COAR has been awarded a US$4 million grant from Arcadia, which will go towards the COAR Notify Project. The funded project began on July 1st, 2022 and will last for four years.

Running this website locally

These sources are designed to be run or compiled with Hugo (except for the search index - see below) See installing Hugo

The 'root' of the Hugo sources is the top-level folder webroot in this Git repository. All Hugo commands should be run with that as the working directory.

There are two ways to use Hugo to compile the sources into a "static" website:

Option1: Using built-in web-server to compile & serve the pages in memory

This is the recommended way for general development and testing. No pages are written to disk - it is just compiled and served from memory:

Run this command:

hugo server -e development --disableFastRender=true --renderToMemory=true

The website will now be available at: http://localhost:1313 I recommend --disableFastRender=true because the "live update" feature can be occasionally confusing in terms of browser caching etc. Better to make changes and then reload the browser page as you would normally do.

Advantages

• Quick and easy
• Ideal for most development

Disadvantages

• No support for the Javascript search index operation or the Javascript search function on the site
• No support for HTTPS (but this should not be an issue)

Option 2: Compiling into HTML files on the local file-system

To do this, you need to use the following command:

-e production -d <STATIC_WEBSITE_DESTINATION_FOLDER_PATH> --cleanDestinationDir

<STATIC_WEBSITE_DESTINATION_FOLDER_PATH> should be replaced with the path to a local folder where you want the static HTML pages etc. to be saved. This folder then becomes the 'root' of a compiled website. You can then either access this directly on the file-system or via a separately configured web-server (e.g. Apache).

Advantages

• You can inspect the resulting HTML etc. - occasionally useful for debugging/understanding
• You can generate the search index with pagefind
• You can use the search function on the website (if serving from a web-server)

Disadvantages

• Less convenient than using the built-in web-server.

Search index

The compiled website supports a Javascript search engine (using pagefind) which generates and makes use of a locally compiled search index. However, this search index is not compiled by Hugo, and must be compiled separately, after the Hugo compilation has been run. To compile the search index, run this command (assuming you have installed a compiled binary of pagefind):

pagefind --site <STATIC_WEBSITE_DESTINATION_FOLDER_PATH>

This will add a folder: <STATIC_WEBSITE_DESTINATION_FOLDER_PATH>/pagefind which contains everything that the search function needs (Javascript, CSS, search index etc.)