It would be great if the instructions could be able to be repurposed into various products from a Single Source of Knowledge. This way:

• there will be a standard name (in all languages) for each of the components of AUMI
• there will be a way to expand on details from tutorials and general workflows
• there will be standardized formats that can be used for making booklets or posters as well as online media
• a diversity of presentations (internationalized, audio, and accessible styles ) can be encouraged and audited.

A "Block" will be the basic unit of information. Blocks consist of

• Metadata
• • BLOCK-ID
• • UI names (localized)
• • UI description (localized)
• • tag list
• • credits
• • last updated date (which is a stand in for a version)
• actual textural content
• one or more related images, possibly organized as slides, animated gifs, or short video)
• inline, links related Blocks (that can be harvested for an index)
• internal notes and comments

and Blocks can be organized through:

• a hierarchical table of contents,
• the result of a tag search
• the result of a full text search
• an Index for the printed versions
• possibly a WIKI, but the instructions will be close to that anyway, the big difference being that several WIKI "pages" will be in a single webpage or document.

Blocks themselves are JSON files. They should be in a directory named BLOCK-ID holding:

• the block file BLOCK-ID-block.json
• text is coded in markdown. Some internal features of Markdown may be exploited, such as extracting headers to be short descriptions in the Table of Contents
• "media" directory with image files, audio files, movie files, and any other media, each starting with BLOCK-ID

The INTERNAL-ID list will give the official names. They are concatenated codes that roughly follow the hierarchy of the main Instructions. For example

• "INTRO"
• "INTRO-HARDWARE"
• "INTRO-SOFTWARE"
• "INTRO-STAGING"
• "INST-USER"
• "INST-ORDER"
• "INST-LOOPS"
• "INST-MORE"
• "LEGAL" etc. These may be turned into 2 or 3 letter codes ( INT, INS )or something. The full list of blocks, by BLOCK-ID is on the Instructions Catalog page.

• Blocks concentrate on one aspect of a feature that are gathered into Subjects
• Subjects are gathered into Features
• Features are gathered into Topics
• and all the Topics combined are the Document.

Example:

• LOOKS are a Topic
• • LAYOUT is a Feature
• • • OVERVIEW is a Subject
• • • TYPES is a Subject
• • • • GRID is a Block : LOOKS-LAYOUT-TYPES-GRID is the BLOCK-ID

The thing corresponding to the current info is in a json file called website.json . since the web pages are built out of markdown. md has no concept of "including" or "framing", so that is a kind of extension. Good old links are still in md link format. but included external files need a new syntax. I'd suggest (this is a block's content json):

## Name
description
***
<<BLOCK-ID>>
<<BLOCK-ID>>
etc.

The *** is also a signal for the end of the summary.