Learning to Love HTML5


It seems that new resources and articles for teaching and promoting HTML5 are popping up almost daily. We’ve been given HTML5 templates in the form of the HTML5 boilerplate1 and HTML5 Reset2 (although they both go beyond just HTML5 stuff). We’ve got a plethora of books3 to choose from that cover HTML5 and its related technologies. We’ve got shivs4, galleries5, and a physician6 to help heal your HTML5 maladies. And don’t forget the official spec7.

From my own vantage point — aside from a few disputes8 about what the term “HTML5″ should and shouldn’t mean — the web design and development community has for the most part embraced all the new technologies and semantics with a positive attitude.

HTML5 for Web designers book9
Flickr Photo by Jeremy Keith10

While it’s certainly true that HTML5 has the potential to change the web for the better, the reality is that these kinds of major changes can be difficult to grasp and embrace. I’m personally in the process of gaining a better understanding of the subtleties of HTML5’s various new features, so I thought I would discuss some things associated with HTML5 that appear to be somewhat confusing, and maybe this will help us all understand certain aspects of the language a little better, enabling us to use the new features in the most practical and appropriate manner possible.

The Good (and Easy) Parts

The good stuff in HTML5 has been discussed pretty solidly in a number of sources including books by Bruce Lawson11, Jeremy Keith12, and Mark Pilgrim13, to name a few. The benefits gained from using HTML5 include improved semantics, reduced redundancies, and inclusion of new features that minimize the need for complex scripting to achieve standard tasks (like input validation in forms, for example).

I think those are all commendable improvements in the evolution of the web’s markup language. Some of the improvements, however, are a little confusing, and do seem to be a bit revolutionary, as opposed to evolutionary, the latter of which is one of the design principles14 on which HTML5 is based. Let’s look at a few examples, so we can see how flexible and valuable some of the new elements really are — once we get past some of the confusion.

An <article> Isn’t Just an Article

Among the additions to the semantic elements are the new <section> and <article> tags, which will replace certain instances of semantically meaningless <div> tags that we’re all accustomed to in XHTML. The problem arises when we try to decipher how these tags should be used.

Someone new to the language would probably assume that an <article> element would represent a single article like a blog post. But this is not always the case.

Let’s consider a blog post as an example, which is the same example used in the spec15. Naturally, we would think a blog post marked up in HTML5 would look something like this:

<h1>Title of Post</h1>
<p>Content of post...</p>

<p>Content of post...</p>

	<p>Comment by: Comment Author</p>
	<p>Comment #1 goes here...</p>
	</section> <section> <p>Comment by: Comment Author</p>
	<p>Comment #2 goes here...</p>
	</section> <section> <p>Comment by: Comment Author</p>
	<p>Comment #3 goes here...</p>

For brevity, I’ve left out some of the other HTML5 tags that might go into such an example. In this example, the <article> tags wrap the entire article, then the “section” below it wraps all the comments, each of which is in its own “section” element.

It would not be invalid or wrong to structure a blog post like this. But according to the way <article> is described in the spec, the <article> element should wrap the entire article and the comments. Additionally, each comment itself could be wrapped in <article> tags that are nested within the main <article> tag.

Below is a screen grab from the spec, with <article> tags indicated:

Articles nested in an article
Article tags can be nested inside article tags — a concept that seems confusing at first glance.

So, an <article> element can have other <article> elements nested inside it, thus complicating how we naturally view the word “article”. Bruce Lawson, co-author of Introducing HTML516, attempts to clear up the confusion in this interview17:

“Think of <article> not in terms of print, like “newspaper article” but as a discrete entity like “article of clothing” that is complete in itself, but can also mix with other articles to make a wider ensemble.”

— Bruce Lawson

So keep in mind that you can nest <article> elements and an <article> element can contain more than just article content. Bruce’s explanation above is very good and is the kind of HTML5 education that’s needed to help us understand how these new elements can be used.

Section or Article?

Probably one of the most confusing things to figure out when creating an HTML5 layout is whether or not to use <article> or <section>. As I write this sentence, I can honestly say I don’t know the difference without actually looking up what the spec says or referencing one of my HTML5 books. But slowly it’s becoming more clear. I think Jeremy Keith defines <article> best on page 67 of HTML5 for Web Designers:

“The article element is [a] specialized kind of section. Use it for self-contained related content… Ask yourself if you would syndicate the content in an RSS or Atom feed. If the content still makes sense in that context, then article is probably the right element to use.”

— Jeremy Keith, HTML5 for Web Designers

Keith’s explanation helps a lot, but then he goes on to explain that the difference between <article> and <section> is quite small, and it’s up to each developer to decide how these elements should be used. And adding to the confusion is the fact that you can have multiple articles within sections and multiple sections within articles.

As a result, you might wonder why we have both. The main difference is that the <article> element is designed for syndication, whereas the <section> element is designed for document structure and portability. This simple way to view the differences certainly helps make the two new elements a little more distinct. The important thing to keep in mind here is that, despite our initial confusion, these changes, when more widely adopted, are going to help developers and content creators to improve the way they work and the way content is shared.

Headers and Footers (Plural!)

Two other elements introduced in HTML5 are the <header> and <footer> elements. On the surface, these seem pretty straightforward. For years we’ve marking up our website headers and footers with <div id="header">, <div id="footer"> or similar. This is great for DOM manipulation and styling, because we can target these elements directly. But they mean nothing semantically.

“The div element has no defined semantics, and the id attribute has no defined semantics. (User agents are not allowed to infer any meaning from the value of the id attribute.)”

Mark Pilgrim, Dive Into HTML518

HTML5’s introduction of <header> and <footer> elements is the perfect way to remedy this problem of semantics, especially for such often-used elements. But these elements are not as straightforward as they seem. Technically speaking, if every website in the world added one <header> and one <footer> to each of their pages, this would be perfectly valid HTML5. But these new elements are not just limited to use as a “website header” and “website footer”.

A header is designed to mark up introductory or navigational aids, and a footer is designed to contain information about the containing element. For example, if you used the footer element as the footer for a full web page, then in that case copyright, policy links, and related content might be appropriate for it to hold. A header on the same page might contain a logo and navigation bar.

But the same page might also include multiple <section> elements. Each of those sections is permitted to contain its own header and/or footer element. Keith sums up the purpose of these elements well:

“A header will usually appear at the top of a document or section, but it doesn’t have to. It is defined by its content… rather than its position.”

“Like the header element, footer sounds like it’s a description of position, but as with header, this isn’t the case.”

— Jeremy Keith, HTML5 for Web Designers

And the spec adds to Keith’s clarification by noting:19

“The header element is not sectioning content; it doesn’t introduce a new section.”

The header element in the HTML5 specification20

These explanations help dispel any false assumptions we might have about these new elements, so we can understand how these elements can be used. Really, this method of dividing pages into portable and syndicatible content is just adding semantics to what content creators and developers have been doing for years.

Headings Down A Different Path

Prior to HTML5, heading tags (<h1> through <h6>) were pretty easy to understand. Over the years, some best practices21 have been adopted in order to improve semantics, SEO, and accessibility. Generally, we’ve become accustomed to including a single <h1> element on each page, with the other heading elements following sequentially without gaps (although sometimes it would be necessary to reverse the order).

With the introduction of HTML5, to use the new structural elements we need to rethink the way we view the structure of our pages.

Here are some things to note about the changes in heading/document structure in HTML5

  • Instead of a single <h1> element per page, HTML5 best practice encourages up to one <h1> for each <section> element (or other section defined by some other means)
  • Although we’re permitted to start a section with an <h2> (or lower-ranked) element, it’s strongly encouraged22 to start each <section> with an <h1> element to help sections become portable
  • Document nodes are created by sections, not headings (unlike previous versions of HTML)
  • An <hgroup> element is used to group related heading elements that you want to act as a single heading for a defined or implied section; <hgroup> is not used on every set of headings, only those that act as a single unit outside of adjacent content
  • To see if you’re structuring your document correctly, you can use the HTML5 Outliner23
  • Despite the above points, whatever heading/document structure you used in HTML4 or XHTML will still be valid HTML5

So, although the old way we structure pages does not amount to invalid HTML5, our view of what constitutes “best practice” document structure is changing for the better.

Block or Inline? Neither! (Sort of…)

For layout and styling purposes, CSS developers are accustomed to HTML elements (for styling and layout purposes) being defined under one of two categories: Block elements and inline elements (although you could divide those two into further categories24). This understanding simplified our expectations of an element’s display on any given page, making it easier (once we grasp the difference between the two) to style and manoeuvre the elements.

HTML5 evolves this concept to include multiple categories, none of which is block or inline. Well, theoretically, block and inline elements still exist, but they do so under different labels. Now the different categories of elements include:

I certainly welcome this kind of improvement to more appropriately categorize elements, and I think developers will adapt well to these changes, but it is important that we promote proper nomenclature to ensure minimal confusion over how these elements will display by default. Of all the areas discussed in this article, however, I think this one is the easiest to grasp and accept.


While this summarizes some of what I’ve learned in my study of HTML5, a far better way for anyone to learn about these new features to the markup is to pick up a book on the topic. I highly recommend one of those mentioned in the article, or you can read Mark Pilgrim’s book30 online.

These new elements and concepts don’t have to be confusing. We can take the time to study them carefully, avoiding confusion and dispelling myths. This will help us enjoy the benefits of these new elements as soon as possible, and will help developers and content creators pave the way towards a more meaningful web — a web that, to paraphrase Jeremy Keith, ‘wouldn’t exist without markup’.

You may be interested in the following related posts:


  1. 1 http://html5boilerplate.com/
  2. 2 http://html5reset.org/
  3. 3 http://www.webdesignerdepot.com/2010/05/html5-and-css3-books-to-watch-for-in-2010/
  4. 4 http://code.google.com/p/html5shiv/
  5. 5 http://html5gallery.com/
  6. 6 http://html5doctor.com/
  7. 7 http://dev.w3.org/html5/spec-author-view/spec.html
  8. 8 http://jeffcroft.com/blog/2010/aug/02/term-html5/
  9. 9 http://www.flickr.com/photos/adactio/4764451727/
  10. 10 http://www.flickr.com/photos/adactio/4764451727/
  11. 11 http://introducinghtml5.com/
  12. 12 http://books.alistapart.com/products/html5-for-web-designers
  13. 13 http://diveintohtml5.org/
  14. 14 http://www.w3.org/TR/html-design-principles/#evolution-not-revolution
  15. 15 http://www.whatwg.org/specs/web-apps/current-work/multipage/sections.html#the-article-element
  16. 16 http://www.peachpit.com/store/product.aspx?isbn=0321687299
  17. 17 http://www.peachpit.com/articles/article.aspx?p=1629150
  18. 18 http://diveintohtml5.org/semantics.html#header-element
  19. 19 http://www.whatwg.org/specs/web-apps/current-work/multipage/sections.html#the-header-element
  20. 20 http://www.whatwg.org/specs/web-apps/current-work/multipage/sections.html#the-header-element
  21. 21 http://www.456bereastreet.com/archive/200911/headings_and_document_structure_conclusions/
  22. 22 http://www.whatwg.org/specs/web-apps/current-work/multipage/sections.html#headings-and-sections
  23. 23 http://gsnedders.html5.org/outliner/
  24. 24 http://reference.sitepoint.com/css/formattingcontext
  25. 25 http://www.whatwg.org/specs/web-apps/current-work/multipage/grouping-content.html
  26. 26 http://www.whatwg.org/specs/web-apps/current-work/multipage/text-level-semantics.html
  27. 27 http://www.whatwg.org/specs/web-apps/current-work/multipage/content-models.html#sectioning-content-0
  28. 28 http://www.whatwg.org/specs/web-apps/current-work/multipage/forms.html#forms
  29. 29 http://www.whatwg.org/specs/web-apps/current-work/multipage/embedded-content-1.html#embedded-content-1
  30. 30 http://diveintohtml5.org/
  31. 31 http://www.smashingmagazine.com/2010/09/23/html5-the-facts-and-the-myths/
  32. 32 http://www.smashingmagazine.com/2010/05/18/html5-and-flash-why-its-not-a-war-and-why-flash-wont-die/
  33. 33 http://www.smashingmagazine.com/2009/08/04/designing-a-html-5-layout-from-scratch/
  34. 34 http://www.smashingmagazine.com/2009/07/16/html5-and-the-future-of-the-web/

↑ Back to top Tweet itShare on Facebook

Louis Lazaris is a freelance web developer and author based in Toronto, Canada. He blogs about front-end code on Impressive Webs and curates Web Tools Weekly, a weekly newsletter for front-end developers.

  1. 1

    Adding new tags to improve the document outline and further the semantics of the document structure, so it becomes more machine readable and portable is a very good thing.

    “Streamlining” would not be the verb I’d use when you don’t use these tags, because it adds to your mark-up.

    You will have a complete spec when it gets released, there is uncertainty with its users now, because the very tag definitions are still hazy. Only after HTML5 becomes a standards, can a webdev know what and how to use these new tags to the letter.

    Since these new tags are mostly to mark-up document outline, and the older tags are used to mark-up content: or these new tags give meaning to document structure, the old tags give meaning to the content – you will always operate on a more abstract level. What defines document structure is very personal to the document author. What defines document content isn’t. That is why these new tags are kept somewhat ambiguous: to be free to outline your document and give meaning to its structure.

    Also as soon as WordPress HTML5 comes out, even these document outline methods will be heavily standardized.

  2. 52

    Red herring when Google (the lifeblood of nearly every commercial website) says not to make excessive use of multiple h1 headings on a page.

    You’ll have a hard time really defending putting more than two h1 headings on page. W3C also says you can use tables to arrange data, but coding a navigation inside a table (perfectly allowed) or putting a form inside a table (perfectly allowed) isn’t something you’d want to find widespread on the internet, count as an argument for using tables for layout, or teach to a new webdeveloper.

    Speaking of the devil, Matt Cutts, Engineer at Google: http://www.youtube.com/watch?v=GIn5qJKU8VM More than one H1 on a page: good or bad?

  3. 103

    Nice writeup, not read the whole book but I did notice it in the iBooks Store and downloaded a Sample.

    Thanks for the share!

  4. 154

    Ow, and for future SEO impact, look at what Google is doing right now with semantics on content blocks, such as a heading followed by a list: http://www.seobythesea.com/?p=3824 (Google Defines Semantic Closeness as a Ranking Signal). And imagine what it can do with the semantically improved document outline of HTML5.

  5. 205

    nice article man…..hope more an more

  6. 256

    Which i am also looking forward to but with HTML5 looking like it is being held up for a few years i can’t see my company taking it on till it’s ready.

  7. 307

    testing the form

  8. 358

    Related to this- I fail to see how “section” is any more or less semantic than “division,” which “div” is short for. When thought of as “an article of clothing,” the same can be said of “article.”

  9. 409

    Nice one! I have been waiting HTML 5 for a long time, it’s good to see that things moving the right way! :)

  10. 460

    Straydog Branding

    November 12, 2010 8:30 am

    This book nevertheless is a must read for anyone new to HTML5 and wanting to get a taste of what it is, what it offers, and how it can be used.

  11. 511

    I just put down Introducing HTML5 by Bruce Lawson to read this article and it just retold everything I read so far. And that’s just pretty cool.

  12. 562

    Interesting post. I have added it to the “HTML5 for Enterprise and Mobile applications” page (http://www.facebook.com/pages/Html5-for-Enterprise-and-Mobile-applications/175012219177552?v=wall)

  13. 613

    I think that while the section and article tags may have similar uses, the content contained within each on respectively can be very different. After looking at the spec description it makes sense for article to me used for content such as this, whereas section would not be used for content “posts” such as this article, instead finding its way into situations like upcoming events and announcements. Also, it makes sense to have articles within sections, as in a feed previewer which is the section and the previews are the articles, and where sections are withing articles, such as chapters within a larger article.

    I understand the more than one h1 tag sounds like too many, but even listening to the guy from Google talk about using them, be leaves room for the tags, while still saying “not too many.” I take this to suggest that a new h1 is appropriate for the beginning of every parent article or section.

  14. 664

    HTML5 is just a big failure in the design industry. no one will use it. unless ie6 dissappear for ever, but thats no going to happen in a lonnnnnnng time.

  15. 715

    semantics tag was a bad idea, imho.

  16. 766

    This is a great article. Thanks for simplifying the basics of html5.

  17. 817

    This book nevertheless is a must read for anyone new to HTML5 and wanting to get a taste of what it is, what it offers, and how it can be used.

  18. 868

    Great post Louis, I think this highlights really well the learning curve that developers embracing HTML5 have to go through. The brief rule that we have picked up on are that articles can be syndicated, sections really should have a heading and that divs are fine if the other two do not fit, but these are usually just for styling purposes, in our experience. The books and sites you have referenced (as well as the spec) have been a great resource and I encourage other developers to read the spec too, although not always the easiest to digest, why not?

  19. 919

    i agree you 100%, ie6 is the black sheep.

  20. 970

    I am learning to love HTML5, but it’s a big change over for me.

    Thanks for posting.

    -Jessica Boblooch

  21. 1021

    What I love the most about HTML5 is the great support it has in terms of SEO, just look at the ‘hgroup’ tag that wraps multiple Heading tags and make it one but still keep the hierarchy of the heading tags… if you go into more detail you will also notice that the ’em’, ‘i’, ‘b’, ‘strong’, etc. have different characteristically when it comes to SEO (machine/robot reading) then they use to have.

    What’s also great is the video and audio embedding capabilities (it’s a shame that mp4 isn’t supported by most browsers due to legalities that might be happening in the future), but on a positive not, Firefox 4 will be adding Open GL support which will make it possible to run games engines through a browser, now that’s impressive.

    I’m hoping that in the not so distant future HTML5 will KILL Flash on the web, in my opinion open source will always be a million times better then a paid service, and the support will always be better.

  22. 1072
  23. 1123

    Great article!
    It helped me to better understand some of the new elements HTML5 offers us.


↑ Back to top