Presentational HTML - Semantically Useful or Wastefully Deprecated?

27 May 2008

I recently started a Twitter discussion with Chris Coyier of CSS-Tricks on the relevance and usefulness of the  tag. I’ve had this same debate with co-workers on why one would still use  over just using a  or  with an appropriate <class> or <id> to define the “smallness” of the tag. The one thing we both noticed is that no one uses the <big> tag anymore, so why should anyone use the  tag? This seems to be only the tip of the iceberg when it comes to determining which element to use for an intended result. Let’s look at a few more instances of controversy in the realm of standards and semantics on the web.

Strong or Bold? How Do You Like Your Emphasis?

Another debate is why we now use  and  for bold and italics over just using  and . Is there a true reason for why we as web designers have made this transition? None of these tags are deprecated (according to the W3C), so why have these presentational tags gone by the wayside?

I suppose it becomes a matter of semantics, which leads to the current buzz term in web design - the semantic web. The W3C does state that even though these tags are not deprecated, their use is discouraged in favor of style sheets. My good friend, Eric Filson, takes this approach and never uses any presentational markup for text (no , , , or  ) but rather wraps the element in a  and applies the appropriate CSS (such as font-weight: bold or font-style: italic) instead.

An old forum thread on Webmaster World dives into this struggle with some helpful clarification:

When setting a bold or italic font for purposes of typography only, use  and . If there’s a word or phrase that you want to emphasize as you would while speaking, then - and only then - use  or .

I’m fairly certain however that many designers don’t take this approach. When you look at industry standard editors like Dreamweaver or Expression Web, there are exclusive selectable options defaulted to “use  and  instead of  and ” - so there’s an obvious push from the top down as well to essentially deprecate those tags for designers to begin with.

When an `<address>` Isn’t an Address

Microformats adopters have given some focused attention to the rarely used <address> tag lately since the markup of adr relies heavily on adding address information to a page. One would assume that using the one tag in the HTML spec that seems perfect, <address>, seems obvious. In fact, <address> is almost never the right choice. The adr-examples wiki sets the story straight:

The <address> element is typically very confusing for web authors and is best avoided.

The <address> element does NOT mean what you might think it means (unless you are one of the few folks who has thoroughly read the HTML 4.01 recommendation.

The <address> element does NOT mean “address”.

The hcard-faq also goes into further explanation to say:

The poorly named <address> element really means <contact-info-for-this-web-page>. The HTML4 definition of the ADDRESS element says it is used “to supply contact information for a document or a major part of a document such as a form.” Therefore, <address> should be used ONLY IF it represents the contact information for the page or major part thereof. In short, DO NOT use <address> to markup physical/street/mailing addresses. Only use it to markup the contact information for the page.

So the moral of the story here is apparently we shouldn’t be using <address> anymore because it’s just too confusing. But if you’re going to use it, it’s basically achieving the same thing <meta name="author"> is for, but in a block of text somewhere within the body of your page. What a piece of junk <address> turned out to be!

The Future Of The Web

So with all of that said - what do your markups look like? Will you roll along with whatever works for you, or will you jump through hoops to markup your code semantically and get on the “Web 3.0” bandwagon? Are your sites going to be ready for the bleeding edge of microformats, RDF, OWL, SWRL, SPARQL, GRDDL and WPT? Will any of this matter to paying clients? (strong emphasis!)

Come to think of it - maybe I should have used <acronym> on all of those…

Share and Enjoy:

Tags: acronym, address, adr, bold, buzz, Chris Coyier, co-workers, css, CSS-Tricks, Design, emphasis, Eric Filson, font, GRDDL, html 4.01, HTML spec, markup, microformats, OWL, presentational, semantic web, semantics, SPARQL, SWRL, tags, The Web, twitter, w3c, web design, webdesign, wiki, WPT

Filed under: Articles, Design, Opinions, The Web

3 Responses for "Presentational HTML - Semantically Useful or Wastefully Deprecated?"

Chris Coyier May 27th, 2008 at 10:08 am 1
The reasoning behind replacing is semantics. It’s better to tags that generically describe the intent of the enclosed content than what it should look like. Emphasis might look like italic on one site, where on another it could mean 120% font size and bright red. If it’s the latter, the tag no longer makes sense… It’s kind of like how class names like “BigLeft” are bad, since later that same class could be needed to apply a small and right formatting.

That being said… I still find myself using the small tag fairly often. Mostly because it’s another INLINE element that does what it says on the box. Sometimes you need an extra hook within a p tag, and don’t want to use strong or em, and small fits the bill nicely.
Travis Turner May 27th, 2008 at 10:50 am 2
I have to admit, I’ve almost completely abandoned inline styles except when quickly mashing something together, or for initial look and feel before I go back and move everything into a stylesheet.

I suppose, from an accessibility standpoint, this is probably frowned upon because it hinders a screen-reader’s ability to place emphasis on certain parts of the text or what have you. But to be completely honest, most of my styling seems to be for the presentation layer and NOT for semantics.
Marcin July 15th, 2008 at 10:08 am 3
nice portfolio - many good looking webs.. gratulations