Web Directions

John’s HTML5 article in A List Apart

• By:John
• Tweet: @johnallsopp
• January 7, 2009
• Leave a response

There’s an article by me (John) in today’s A List Apart, on the direction that HTML5 is taking in regards to semantics. I have long thought the approach HTML5 takes to extended the semantics of HTML was problematic in several ways, being not backwards compatible with at the very least IE6, and not extensible, and so essentially not solving the problem that it sets out to solve at all.

There’s some very detailed and thoughtful comments already after the article. So if you are interested in the issues of HTML5 and semantics, please take the time to have a look, and possible add to the conversation.

3 responses to “John’s HTML5 article in A List Apart”:

1. • By:Lachlan Hunt
   • January 9th, 2009
   

   The major compatibility problem with IE already has an available workaround that has been known about for almost a year now.

2. • By:John
   • January 9th, 2009
   

   Thanks Lachlan,

   I did become aware of that during the publication process, but didn’t get a change in in time.

   However, I honestly think the fact that there’s a JavaScript hack for this really means that these elements are backwards compatible in the sense I meant in the article. The core of the focus on backwards compatibility is this — if everyday web developers don’t think that something is backwards compatible, they won’t adopt it, and it will wither on the vine. For those sorts of developers, the likelihood of them even learning about those solutions is low — all they’ll see is that it doesn’t work in IE, and then, end of story, it won’t be adopted.

   But that’s also only one tiny piece of the puzzle. A major piece being how did it come to pass that it was thought that sticking a few random new elements in HTML5 was a good idea? And a heap of issues follow from that — where did the names come from? What structural or semantic model is underpinning these choices? Why not just use DocBook? What problem is it even meant to be solving?

   There’s a very significant debate needed around all this, at least IMO.

3. • By:Richard Fink
   • January 12th, 2009
   

   Caught your article a few days ago. I have not been following HTML 5 at all, and your article was an eye-​​opener that framed the issues very, very, clearly, which is tough to do.
   Nice writing. Clear thinking.
   Got me interested in HTML 5.

Your opinion: