Building Better Blocks
#brainful#tutorial
IMPORTANT: Do not modify this block as changes will not be saved. New versions of this block will replace any modifications.
Making the most out of AI is a two-way street. The more helpful you make your data, the more accurate and useful the AI will be.
Keep in mind the following checklist to make the most out of your blocks:
- Short and sweet: Whilst blocks can be as long as you want, it is recommended to keep them as succinct as possible whilst still retaining necessary context for it to be self-sufficient. Break down longer blocks into smaller ones where possible, by shifting related blocks as links and directly continuing blocks as children. The ideal block length is anywhere between 100-1000 words depending on the context and complexity of the content.
- Demonstrate by example: Use examples to illustrate your points. This not only makes it easier for future you to understand, but just as importantly, for AI to understand how to apply the information correctly in practice! So whilst succinctness is valuable, adding a few high quality examples enforces higher accuracy of application. The less assumptions the block makes, the better.
- Focused formatting: Use headings to break down your content into sections. This makes it easier to navigate and understand the structure of your block. Use dividers to separate different parts of your block. The more style annotations you have (bold, underline, italics, highlight), the more context AI has to work with.
- Effective entities: Entities are the main way to organise your blocks. Since brainful has no folders, and essentially stores your data as a graph, entities are the main way to organise your blocks and for AI to understand the key ideas and concepts of the block. Entities associate blocks with a common theme or topic. Create them using the '#' symbol followed by the entity name.
- - Deep links: Link to other blocks within the same space using the [title](block_luid) format. This creates a network of blocks that are interconnected. This is useful for creating a web of knowledge and for AI to understand and traverse the relationships between different blocks.
게시물에 답글