You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am creating this issue to communicate a use case that I believe is important. I believe the use case is already covered; however, I wanted to bring it to everyone's attention.
When creating a book whose audience is AI (LLM), the desired output is markdown (not html).
Said another way, I find myself producing a book twice: once for humans (html) and once for AI (markdown).
Why would you want a markdown target when the source is already in markdown? Because you get the benefit of the pre/post-processors. These processors can ensure you source is as short and concise as is humanly possible while the target is fully expanded and complete.
I have read comments from time to time stating "the only value to md as a target is debugging output". I want to bring the AI use case to everyone's attention.
Proposed Solution
I hope this helps!
Notes
No response
The text was updated successfully, but these errors were encountered:
Problem
I am creating this issue to communicate a use case that I believe is important. I believe the use case is already covered; however, I wanted to bring it to everyone's attention.
When creating a book whose audience is AI (LLM), the desired output is markdown (not html).
Said another way, I find myself producing a book twice: once for humans (html) and once for AI (markdown).
Why would you want a markdown target when the source is already in markdown? Because you get the benefit of the pre/post-processors. These processors can ensure you source is as short and concise as is humanly possible while the target is fully expanded and complete.
I have read comments from time to time stating "the only value to md as a target is debugging output". I want to bring the AI use case to everyone's attention.
Proposed Solution
I hope this helps!
Notes
No response
The text was updated successfully, but these errors were encountered: