The word Node can be referred back to Telecommunications Networks - a device such as a computer or switch attached to a computer or telecommunications network, or a point in a network topology where lines intersect or terminate. From there, it probably arrived in XML to represent any type of entity in a document e.g. a Comment node, a Text node, an Element node etc. Similarly, the representation of HTML DOM has various entities like tags, text content, comments, attributes etc. - all these are called Nodes.
With respect to HTML DOM, there are 12 different types of Nodes, out of which 5 are deprecated (
). HTMLElement is a Node of type 1 (ELEMENT_NODE).
To construct these Nodes and build the DOM Tree, the browser implementations have Interfaces. The chain of these interfaces start at Object as the root.
Let's take an example to illustrate this.
<html>
<head>
<title>Document Title</title>
</head>
<body>
<h1>Some heading</h1>
<hr />
<p>Paragraph text goes here.</p>
</body>
</html>
In the above DOM structure, <H1>is an instance of HTMLHeadingElement. The properties available on <H1> instance would be inherited in a fashion - HTMLHeadingElement << HTMLElement << Element << Node << EventTarget << Object. For example, the two DOM properties of the <H1> element (1) .childNodes and (2) .children are inherited from Node. Another interesting thing about these two similar sounding properties is the difference between their results. The former will list any type of Node (one text node, in this case) and the later will list Nodeonly of type 1 (ELEMENT_NODE) (none, in this case).
To summarize, every HTMLElement is a Node but not vice versa. The most common type of Node which is not an HTMLElement is TEXT_NODE.