Society for the Definitive Abolition of Tag Abuse

Questions about Tag Abuse

Tag Abuse is not necessarily a pejorative term, merely descriptive.

What is Tag Abuse?

Tag Abuse is the term which describes the wilful or accidental use of an element type or character entity in an SGML or XML document for a purpose for which it was not designed, in such a way as to be misleading or inaccurate.

A minor example (abusing HTML):

<p>The new number for your FooBank 
   ATM card is <code>123456</code>.</p>

The code element in HTML is for computer code (programming fragments), not for coded numbers like access keys or passwords. There is a special element for sample literal sequences of characters called samp.

What does it matter?

Often it's relatively harmless and all it demonstrates is that the author is unable to apply accurate markup because it doesn't exist in the DTD or Schema, or because they are insufficiently trained.

But poor quality markup leads to poor quality documents: searching fails or becomes misleading; conversions don't produce the expected output; printer and browser formatting become unreliable, and in extreme cases, data may become corrupted or users may end up with wrong information, and in mission-critical systems that can lead to more serious damage.

It also contribute to perpetuating bad practices, faulty beliefs, or urban myths. Tag Abuse demonstrates that the DTD or Schema doesn't properly match the type of document being written, or that the author doesn't know how to use it.

Why does Tag Abuse happen?

Three main reasons for Tag Abuse have been identified so far:

• The document author has misunderstood or is unaware of the intention of the document type designer in respect of the the elements provided, possibly because of inadequate documentation by the designer or insufficient experience in the field by the document author;
• The author has not read the documentation, or simply couldn't be bothered to find out how to encode the text properly;
• The author has come across an important item of information which needs encoding in a manner which will enable it to be distinguished later, but the document type designer has not provided an element for this. The author therefore chooses the element closest to the required intent, even if it is suboptimal.

One aspect which shares features of all three, and is one of the most prevalent, is the abuse of HTML markup for its visual appearance in a particular browser. Some appropriate remedies are addressed here.