After having been convinced of the simpler boons of HTML5 by James, last night I finally understood the more long-term boons by reading this piece by Jeffrey Zeldman. Shortly after, I converted this blog from XHTML to HTML51. It involved changing:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" dir="ltr" lang="en-US"> <head profile="http://gmpg.org/xfn/1">
<!DOCTYPE html> <html dir="ltr" lang="en-US"> <head>
That was pretty much all there was to it. “That’s it?” and “Sure it’s simpler but is that really all there is to HTML5?” you ask. Of course not, and certainly I’m not using all the semantic new tags like
header yet, but I don’t have to, and that’s part of the point. It’s really all better explained by the simple quote that did it for me:
[HTML5's] introduction of a limited set of additional semantic elements, its instructions on how to handle failure, and its integration of application development tools hold the promise of richer and more consistent user experiences, faster prototyping, and increased human and machine semantics.
Translated into a bit less of an elegant but more readable statement, it basically means that you should use HTML5, because it’s less code to write, which makes it faster to work with. In addition, it handles errors better and in the future when we get to use the new tags, screen-readers will have a much better idea of what it is you’re trying to write.
This house is clear.
- A note on validation: right now not everything validates, but in fairness, the XHTML that came before didn’t validate either, and since I’m redesigning, I won’t fix it now. ↑