Software

Examine the role of meta tags

Are meta tags are worth the time and effort it takes to add them to Web pages? Take a look at the role of meta tags.

This article originally appeared as a Design and Usability Tactics e-newsletter.

By Michael Meadhra

If you mention meta tags to most Web builders, they immediately think of creating keyword lists for search engines to use when indexing Web pages. That's a natural response to years of conventional wisdom, which held that the keyword meta tag and its brethren were essential ingredients in any search engine optimization effort.

In their quest for higher search engine rankings, some unscrupulous Webmasters would load meta tags with high-ranking keywords that were unrelated to the content of the Web pages on which they appeared. Because of this abuse of the keyword meta tag, major search engines no longer use those keywords as a significant factor in their rankings. Now many Web builders wonder if meta tags are worth the time and effort it takes to add them to Web pages.

A closer look at meta tags

Metadata is data that describes other data. In the case of a Web page, the metadata describes the contents and characteristics of the Web page. Meta tags provide a convenient convention for storing that descriptive data in tag attributes. Meta tags are embedded between the head tags of an HTML document, so they're accessible to search engine robots but aren't displayed in the Web browser window.

Most meta tags have the following syntax:

<meta name="meta tag name" content="meta tag content" />

For example, the following meta tag declares that the author of the page is John Doe:

<meta name="author" content="John Doe" />

Some meta tags are equivalent to HTTP headers and include the http-equiv attribute in place of the name attribute. For example, the Content-Type meta tag specifies the document format and character set.

<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />

Web builders tend to forget about the http-equiv meta tags—probably because they seldom deal with those tags directly. Programs such as Dreamweaver automatically insert the Content-Type tag when you create a new page.

Why bother with meta tags?

Strictly speaking, meta tags are optional. Your Web pages will display in a browser without any meta tags, so why bother with meta tags if they don't boost search engine rankings?

One meta tag you need to continue using is the description meta tag. Some major search engines use the description meta tag to display a short description of your page in the search results. Support for the description meta tag isn't universal, but it's common enough to be worth the time it takes to type in a sentence or two describing your page.

The keyword meta tag isn't dead yet either. Although major search engines no longer use it, your internal site search can make good use of the information in the keyword tag. Even if you don't have a site search currently enabled, you might consider entering keywords now while the page content is fresh in your mind, instead of waiting until after you add the search feature.

Common meta tags and their uses

There's a lot more to meta tags than just the keyword and description tags. In fact, there's a long list of established meta tags available for use in your Web pages. Some meta tags provide more specific information about your page content while others control browser functions and robots. For example, consider this listing of some of the most common meta tags:

  • abstract—provides a brief summary of the contents of your Web page
  • author—offers the page author's name or e-mail address
  • copyright—indicates a copyright statement or the address of a page containing copyright information
  • description—provides a one- or two-sentence description of the page
  • expires—gives the date on which the page content is considered obsolete
  • keywords—offers a comma-delimited list of keywords and phrases that describe the page content
  • generator—indicates what HTML editor was used to generate the page (this is usually inserted automatically by that program; you should delete it)
  • refresh—instructs the browser to automatically reload the page after the specified number of seconds and optionally lists an alternate URL to load (this isn't recommended for page redirects)
  • revisit—indicates how often search engines and offline readers should return to check for new content on the page
  • robots—controls search engine spiders and crawlers, telling them whether to index the page and follow its links

Just because the search engines abandoned the best-known meta tag doesn't mean that Web builders should ignore meta tags and the valuable functions they can perform, including enhancing your local site search.

Dublin Core meta tags

Most Web builders have heard of the Dublin Core meta tags, but few of them actually know what they are.

Defining the Dublin Core meta tags

Like other meta tags, the Dublin Core meta tags contain metadata that describes the Web page content. What sets the Dublin Core meta tags apart is that they're part of a standardized list of meta tags that first received official sanction at a workshop in Dublin, OH in March 1995.

The Dublin Core meta tags are part of the larger Semantic Web initiative, which strives to give data "well-defined meaning." The idea behind the Semantic Web is that computers need to know about the meaning of the Web page content in order to effectively index, sort, retrieve, manipulate, and present it to humans in a usable form. The Dublin Core meta tags are a standardized set of tools for describing Web content in a machine-readable form.

Listing the Dublin Core meta tags

The Dublin Core metadata consists of elements, element refinements, and encoding schemes. Although the full list is much longer, the following 15 elements comprise the "core" of the Dublin Core.

  • title—the formal name for the resource
  • creator—the entity (e.g., person, company, or institution) primarily responsible for creating the content
  • subject—keywords, phrases, or classification codes describing the content
  • description—an abstract, table of contents, or other description of the content
  • publisher—the entity (e.g., person, company, or institution) responsible for making the content available
  • contributor—an entity that contributes content
  • date—the creation or publication date of the resource, normally given in YYYY-MM-DD format
  • type—terms describing the category, function, or genre of the content
  • format—the physical or digital manifestation of the resource; in other words, the media type or its dimensions
  • identifier—an unambiguous reference to the resource; it's usually its Web address, but it might also be an International Standard Book Number (ISBN) or other standardized identifier
  • source—a reference to a resource from which the current resource was derived
  • language—the language used for the content; normally expressed as standard two- or three-letter language tags (e.g., en or eng for English)
  • relation—a related resource
  • coverage—extent or scope of the resource, a geographic location, or time period
  • rights—information about rights held over the resource, which include copyright and other real and intellectual property rights

Note: These definitions are paraphrased from DCMI Metadata Terms.

The syntax for a Dublin Core meta tag is the same as for any meta tag except that the name attribute includes "dc:" or "dcterms:" preceding the element name.

<meta name="dc:element" content="meta tag content" />

For example, the following is the Dublin Core meta tag for the title element of a page entitled "Web Design and Usability Tactics":

<meta name="dc:title" content="Web Design and Usability Tactics" />

What makes the Dublin Core meta tags different?

At first glance, the Dublin Core meta tags may not appear significantly different from their regular meta tag counterparts. After all, the dc:subject meta tag normally contains keywords, much like the regular keyword meta tag.

The difference is that the Dublin Core meta tags are characterized by metadata that conforms to a much more rigorous standard. The accepted practice for Dublin Core metadata is to use standard formats for dates and identifiers and to select most other entries from an appropriate "controlled vocabulary," which is a predefined list that is maintained by a recognized standards organization in a particular field.

For example, the dc:subject meta tag normally contains categories selected from the U.S. Library of Congress Subject Headings or keywords selected from another controlled vocabulary. Contrast that with regular keyword meta tag content, which is often an ad hoc list of terms written by marketing specialists and intended more for optimizing page ranking on search engines than accurately describing page content.

Because Dublin Core metadata content is supposed to conform to recognized standards, it's much more reliable, and therefore more useful, than regular metadata. It's also much harder for the average Web builder to use because it requires in-depth knowledge of the Web page content in context to the appropriate controlled vocabulary.

As a result, developing Dublin Core metadata is usually a job for archivists, librarians, and other specialists. At the very least, it usually requires consultation with subject matter experts to categorize page content properly.

For links for more information about the Dublin Core meta tags and the Semantic Web, visit our discussion forum.

Michael Meadhra has been working in the IT field since the earliest days of the Web. His book credits span some three dozen titles, including How to Do Everything with Dreamweaver MX.

Editor's Picks

Free Newsletters, In your Inbox