HTML5

HTML5 is the standard that was never supposed to be needed.  W3C promoted XML as the new Web standard but, understandably, it failed to take off.  Why?  Too complicated to implement and, hence, too expensive for the majority of organizations with a Web presence - not to mention individuals with home pages.  So, the folks at Opera got busy with preparing a draft version of HTMl5, which was eventually picked up by others and finally accepted by W3C as inevitable.


There's general agreement that HTML5 has an excellent set of features, including the ability to play video without needing third-party software like Flash, and Canvas, its drawing feature. However, the new tags are also thought to be very valuable in enabling a greater definition of page layout. Some of the new tags have been referred to as 'semantic' tags, but I'm afraid this isn't the case.  There is no such thing as a semantic tag in HTML5.


The tags that are said to be "semantic" include <header>, <footer>, <section>, <nav>, <article> and one or two more - some of which are not yet formally accepted (indeed, HTML5 as a whole is not yet formally accepted as a new standard).  Let's look at what "semantic" means; the OED is quite brief in its definitions: "Relating to signification or meaning" and "Also (the study or analysis of) the relationships between linguistic symbols and their meanings".

Now, when we use any of these so-called "semantic" tags, we are saying nothing at all about the meaning associated with the symbols found between the opening and closing tags.  <header> simply indicates that this block of text and/or images is placed at the head of the page; <footer> that such a block goes at the foot of the page; <section> indicates a specific part of the page; and <nav> tells us that navigation elements are to be found here.  In other words, they have nothing to do with meaning and everything to do with page structure. <article> is really a rather odd: it seems that the HTML5 development team have taken the blog as the typical page structure and <article> means a blog entry - since this is the example given in most of the books I've seen on HTML5.  So, an entire paper in Information Research would not be considered to be an "article" - indeed, I'm at a loss to discover how the paper as a whole ought to be tagged in HTML5.

The degree of confusion around the use of the word "semantic" seems to mirror that associated with another computer science adoption - "ontology".  Just as the computer use of the term actually means a classification and has nothing whatsoever to do with the philosophy of being; so "semantic" is being used inappropriately, presumably because the development team couldn't think of anything else.  Indeed, I was rendered speechless by one member of the team with whom I had an e-mail conversation, when he said that he couldn't see the difference between type of information and the meaning of the information.

So, don't expect anything magical from HTML5's semantic tags - they are just another set of layout indicators and no more semantic than the long-standing <table> tag.