A node is a single, well-defined piece of functionality.

An input node sits at the start of a flow. It has one or more output ports.

An output node sits at the end of a flow. It has one input port.

A function node or query node sits in the middle of a flow. It has one input port and one or more output ports.

It is defined in an html file and a js file. The js file contains the runtime functionality. The html file contains the editor definition.

A config node contains configuration information that can be used by nodes. It does not get wired into a flow.

A js/html file pair may contain multiple nodes. This is called a node set. Any error loading the js/html files will cause all nodes in a node set to be unavailable.

An npm node module may contain multiple js/html file pairs - each identified in the package.json file. Hence, an npm module may contain multiple node sets, each containing one or more nodes.

A node pack is a collection of similarly-themed nodes that a user might want to install together. It is structurally the same as a node module, but acts as a meta-package that pulls in multiple specific node modules using the standard package dependency mechanism.

A flow is a collection of connected nodes and their associated config nodes.