When you look at a web page in a web browser, you see, at the simplest level, words. These words usually have some style characteristics, such as different font sizes and colors. In most cases a page also displays images. Sometimes there is a form where you can enter (or search) for information, or customize the display of the page to your liking. Often a page contains animated content and content that changes while the rest of the page remains the same.
This article provides an introduction to HTML. If you've ever wondered what goes on behind the scenes in your web browser, this article is the place to start learning.
A brief history of HTML
In the late 1980s, Tim Berners-Lee was working as a physicist at CERN (the European Organization for Nuclear Research). He devised a way for scientists to share documents over the internet. Prior to his invention, communication via the internet was limited to plain text, using technologies such as email, FTP (File Transfer Protocol), and Usenet-based discussion boards. The invention of HTML made use of a model of content stored on a central server that could be transferred and displayed on a local workstation via a browser. It simplified access to content and enabled the display of "rich" content (such as sophisticated text formatting and the display of images).
What is HTML?
HTML is a markup language. It tells the web browser how to display content. HTML separates "content" (words, images, audio, video, and so on) from "presentation" (the definition of the type of content and the instructions for how that type of content should be displayed). HTML uses a pre-defined set of elements to identify content types. Elements contain one or more "tags" that contain or express content. Tags are surrounded by angle brackets, and the "closing" tag (the one that indicates the end of the content) is prefixed by a forward slash.
For example, the paragraph element consists of the start tag "<p>" and the closing tag "</p>". The following example show a paragraph contained within the HTML paragraph element:
<p>My dog ate all the guacamole.</p>
When this content is displayed in a web browser, it looks like this:
My dog ate all the guacamole.
The browser uses the tags as an indicator of how to display the content in the tags.
Elements that contain content can usually also contain other elements. For example, the italic element ("<i>") can be embedded within a paragraph element:
<p>My dog ate <i>all</i> the guacamole.</p>
When displayed, this looks like:
My dog ate all the guacamole.
Some elements do not contain other elements. For example, the image tag ("<img>") simply specifies the file name of the content (an image) as an attribute:
Often a forward slash is placed before the final angle bracket to indicate the end of the tag. This is optional in HTML but required in XHTML (which is an XML schema that implements HTML elements).
The rest of this article goes into greater deatil regarding the concepts introduced in this section. However, if you want to see HTML in action, check out Mozilla Thimble, which is an interactive on-line editor that teaches you how to write HTML markup. Also, see HTML Elements for a list of available elements and a desription of their use.
Elements - the basic building blocks
HTML consists of a set of elements. Elements define the semantic meaning of their content. Elements include everything between two matching element tags, including the tags themselves. For example, the "<p>" element indicates a paragraph; the "<img>" element indicates an image. See the HTML Elements page for a complete list.
Some elements have very precise meaning, as in "this is an image," "this is a heading," or "this is an ordered list." Otheres are less specific, such as "this is a section on the page" or "this is part of the text." Yet others are used for technical reasons, such as "this is identifying information for the page that should not be displayed." Regardless, in one way or another all HTML elements have a semantic value.
Most elements may contain other elements, forming a hierarchic structure. A very simple but complete web page looks like this:
<html> <body> <p>My dog ate all the guacamole.</p> </body> </html>
As you can see, <html> elements surround the rest of the document, and <body> elements surround the page content. This structure is often throught of as a tree with branches (in this case, the <body> and <p> elements) growing from the trunk (<html>). Technically, this hierarchical structure is called the DOM - the Document Object Model.
HTML documents are written in plain text. They can be written in any text editor that allows content to be saved as plain text (although most HTML authors prefer to use a specialized editor that highlights syntax and shows the DOM. Tag names may be written in either upper or lower case. However, the W3C (the global consortium that maintains the HTML standard) recommends using lower case (and XHTML, the XML schema for HTML, requires lower case).
HTML attaches special meaning to anything that starts with the less-than sign ("<") and ends with the greater-than sign (">"). Such markup is called a tag. Here is a simple example:
<p>This is text within a paragraph.</p>
In this example there is a start tag and a closing tag. Closing tags are the same as the start tag but also contain a forward slash immediately after the leading less-than sign. Most elements in HTML are written using both start and closing tags. Start and closing tags should be properly nested, that is closing tags should be written in the opposite order of the start tags. Proper nesting is one rule that must be obeyed in order to write valid code.
This is an an example of valid code:
<em>I <strong>really</strong> mean that</em>.
This is an example of invalid code:
Invalid: <em>I <strong>really</em> mean that</strong>.
Note that in the valid example, the closing tag for the nested element is placed before the closing tag for the element in which it is nested.
Until the adoption of the HTML5 parsing rules, browsers didn't interpret invalid code in the same way and produced different results when facing invalid code. Browsers were forgiving to Web authors, but unfortunately not all in the same way, resulting in almost unpredictable results in case of invalid HTML. These days are over with the latest evolution of browsers, like Internet Explorer 10, Firefox 4, Opera 11.60, Chrome 18 or Safari 5, as they implement the now standard invalid-code-parsing rules. Invalid code results in the same DOM tree on all modern browsers.
Some elements do not contain any text content or any other elements. These are empty elements and need no closing tag. This is an example:
Many people mark up empty elements using a trailing forward slash (which is mandatory in XHTML).
<img src="smileyface.jpg" />
In HTML this slash has no technical functionality and using it is a pure stylistic choice.
The start tag may contain additional information, as in the preceding example. Such information is called an attribute. Attributes usually consist of 2 parts:
- An attribute name.
- An attribute value.
A few attributes can only have one value. They are Boolean attributes and may be shortened by only specifying the attribute name or leaving the attribute value empty. Thus, the following 3 examples have the same meaning:
<input required="required"> <input required=""> <input required>
Attribute values that consist of a single word or number may be written as they are, but as soon as there are two or more strings of characters in the value, it must be written within quotation marks. Both single quotes (') and double quotes (") are allowed. Many developers prefer to always use quotes to make the code less ambiguous to the eye and to avoid mistakes. The following is such a mistake:
<p class=foo bar> (Beware, this probably does not mean what you think it means.)
In this example the value was supposed to be "foo bar" but since there were no quotes the code is interpreted as if it had been written like this:
<p class="foo" bar="">
Named character references
Named character references (often casually called entities) are used to print characters that have a special meaning in HTML. For example, HTML interprets the less-than and greater-than symbols as tag delimiters. When you want to display a greater-than symbol in the text, you can use a named character reference. There are four common named character references one must know:
>denotes the greater than sign (
<denotes the less than sign (
&denotes the ampersand (
"denotes double quote (")
There are many more entities, but these four are the most important because they represent characters that have a special meaning in HTML.
Doctype and comments
In addition to tags, text content and entities, an HTML document must contain a doctype declaration as the first line. In modern HTML this is written like this:
The doctype has a long and intricate history, but for now all you need to know is that this doctype tells the browser to interpret the HTML and CSS code according to W3C standards and not try to pretend that it is Internet Explorer from the 90's. (See quirks mode.)
HTML has a mechanism for embedding comments that are not displayed when the page is rendered in a browser. This is useful for explaining a section of markup, or leaving notes for other people who might work on the page, or for leaving reminders for yourself. HTML comments are enclosed in symbols as follows:
<!-- This is comment text -->
A complete but small document
Putting this together here is a tiny example of an HTML-document. You can copy this code to a text editor, save it as myfirstdoc.html and load it in a browser. Make sure you are saving it using the character encoding UTF-8. Since this document uses no styling it will look very plain, but it is only a small start.
<!DOCTYPE html> <html lang="en"> <head> <meta charset="utf-8" /> <title>A tiny document</title> </head> <body> <h1>Main heading in my document</h1> <!-- Note that it is "h" + "one", not "h" + the letter "l" --> <p>Loook ma, I am coding <abbr title="Hyper Text Markup Language">HTML</abbr>.</p> </body> </html>