Restructure documentation

[blythed]

@anitaokoh has ideas to restructure the documentation for quicker onboarding, easier understanding and contributions. Please share ideas/ discussions here.

[anitaokoh]

Hey all,
@blythed , @fnikolai , @fazlulkarimweb , @kartik4949 , @guerra2fernando

Below are the images of the documentation structure page and subpages
( sorry I can't get the whole image with the free excalidraw version so I had to break it down into multiple images)

How to read the images

• The red boxes are the main menu and the other color boxes are drop-down menus i.e the green boxes are dropdown menus for the red boxes , the blue are for the green and the yellow are for the blue .
• The Shaded boxes indicates that two or more pages are going to be consolidated. Some of the pages in our current document can be merged with another page due to how short they are or duplicated.

The 7 main menus

• Welcome menu: - This includes an overview of what superduperdb offers and and FAQ
• Get started menu: - This includes the installation options and a descriptive simple code of adding ai to database in few minutes for a quickstart
• Fundamentals menu: - This includes building blocks or modules of superduperdb like the datalayer, component etc
• Integrations menu: All our AI and Data integregations categoreis based on their types (for example, data integration has the sql DB, no-sql db , vector db, python tables, data-warehouses)
• User Guide menu: This includes the "how to" guide ( currently known as the "walkthrough") and the use cases categories
• Developer Guide menu: This includes all the the setup (configure, testing etc) and productionalization
• Reference Menu: This includes changelogs and release notes(when we have one)

Some Observations on the use-cases sub-menu

• The usecases have been shortened : Why?

  • This is because the current"Walkthrough" page or "how-to:" page already touches some of them or we can consolidate them there .e.g vector-search: text, vector-search: image

• The usecases names changed: why?

  • The title, "Building an Image Feature-store in torchvision" is more catchy and descriptive than "image feature" in the current document
  • Adding the DB used in front (i.e "MongoDB: Building an Image Feature-store in torchvision") is useful in three-fold
    • It lets the user know what DB was used.
    • Since the goal is to make sure the database connection setups are similar across the database types (i.e apart from the credentials , connecting to a mongodb should be the same as connecting to a dynamodb for example or any nosql database we integrate with) , we can put a disclaimer that this can be easily changed to any other similar DB .
    • It lets us know which DB is dominating and we can switch it ourselves

What does this structure do?

• It helps for easier navigation and understanding on the user side
• It reduce redundancy and aids consistency on our side

What are your thoughts?
What do you thing about the flow i.e the main menu and the subset menus under it?
What do you thing about the names?
What do you about the pages consolidated?

[fnikolai]

@anitaokoh good work!

A few remarks:

1. About the title, I'd prefer the format Build <what application> using <AI framework> and <database>. For example: Build an Image Feature-store using Torchvision and MongoDB
2. I 'd separate the AI Integrations to AI Frameworks and AI APIs
3. I 'd put the Glossary under Get Started
4. Given the small number of subcategories, perhaps it's better to merge Welcome and Get Started into a single category.

[anitaokoh]

Thank you for the comment @fnikolai

• I really like the idea of point 1 and 4 . This makes alot of sense

• For point 2: The AI api, AI framworks are all AI integrations no? or what did you have in mind?

• One reason I didn't put Glossary in the get started menu is because based on the current content if may be too technical for a get started . i personally don't have a stronger reason for it so I am happy to agree for now. What are thoughts of putting the glossary in the get started section?

[blythed]

Some of the groupings seem rather random for example this one:

[anitaokoh]

Hey @blythed ,

Thanks for the comment.

random in what way? what did you have in mind in that section?

[fnikolai]

Is this category really needed? Most of the things mentioned there can be integrated into other parts of the documentation.