DeadLetter is an open-source machine learning platform for research, personal and commercial applications.

Here you'll find all the information relevant to understand and deploy a Dead Conversation.

Using either our "drop" & "retrieve" method or PARCEL our programable interface (API).

The Basics

Here we'll guide you through the basics of DeadLetter

-	Understanding DeadCL
-	Inputs & Outputs
-	DeadCL Syntax
-	DeadCL MBS  (Meaning Based Syntax)
-	DeadLetter Terminology
-	Deploying DeadLetter

The Components

DeadCL is comprised of the following components;

-	Author Key
-	DeadLetter Libraries
-	DeadLetter Conversations
-	DeadLetter Arrays

Writing Conversations

Applications on DeadLetter are called Conversations, here we'll discuss how you write one

-	Setting Authorship
-	Structuring & Building Flow
-	Discovering Outputs
-	Changing Default URL(s)

Principles

Once you understand a handful of basic principles writing, maintaining & interpreting Conversations is second Nature.

-	Detecting an object
-	Substituting an object
-	Pulling input(s) & Link(s)
-	Posting output(s)
-	Output addresses

Drop & Retrieve Method (DRM)
Principles

By default deadCL is configured to simply read from
a remote URI & post to a 'local' URI.

-	DRM Overview
-	What is 'dropping'
-	What is 'retrieving'

Parcel V4
Principles

Parcel introduces the basic functions of an API - that remains entirely Author programable.

-	DeadLetter Parcel basics
-	Configurable Helper for Author Drops (CHAD)
-	Parcel for messaging (Arnold)
-	Parcel for Conversation inputs & outputs

Tutorials

To test your knowledge of DeadCL we've created the following tutorials;