The clinic is getting busy with more HTML5 ailments. This week, we’ll cover questions about <aside>
, blogging platforms, stylesheet links, id
attribute validation, and take a mammoth semantic journey.
Aside question
Lukasz asked:
I wonder if such use of aside tag inside
section
would be correct:<section id="about"> <h1>about me</h1> <p>blablabla</p> <aside> <h2>you can also find me on:</h2> <ul> <li><a href="#">twitter</a></li> ... </ul> </aside> </section>
Thanks in advance for any help and keep up the good work!
Hard to give definitive advice without seeing the whole page, but it seems fine to me. See Mike’s article on <aside>
.
In general, use <aside>
for a section of a page that consists of content that is tangentially related to but separate from the surrounding content. In print, this would be a sidebar, pull-quote, or footnote. On a weblog article, this could be article metadata in the margin or the comments section.
If your social network links are main content, then <aside>
isn’t the element (maybe <section>
, maybe <nav>
). If they’re just related content (i.e., they could be omitted when showing this page in a feed reader), then <aside>
is a good choice.
peace – Oli
Blogging services
Patrick asked:
Do you have any suggestions to people implementing HTML5 on blogging services like Tumblr? The last time I checked, there are problems making HTML5 blogs on these sites:
- Tumblr – inserted iframe breaks HTML5 conformance
- Posterous – no Javascript allowed at the moment (according to their doc)
- Blogger – it doesn’t preserve my HTML5 DOCTYPE
Are there blogging services that don’t mess with the HTML5 markup?
I used to use Tumblr, and you can use HTML5 in your templates. But any new HTML5 elements (e.g., <section>
) in your articles get munged, even
using the “Raw HTML” setting. Unfortunately, I think that managed blogging services will probably be a little slow to adopt HTML5 (= fix the HTML5 errors they/their tools have), and you’ll have to use your own CMS for that level of control.
You can of course build your own theme for something like WordPress, and a little birdie did tell me that the latest version of WordPress has some HTML5 in its default template, so this might carry over into their hosted service. If you can’t wait and don’t have a server, remember that HTML5 is a sliding scale — I was using <div class="section">
with Tumblr for a while.
peace – Oli
Link stylesheet
Steve asked:
Hello, can we get rid of type=”text/css” when declaring a link rel=”stylesheet”?
Thank you and long live the web.
You certainly can do that. Same with including JavaScript:
<link rel="stylesheet" media="screen, projection" href="http://html5doctor.com/css/style.css">
<script src="http://html5doctor.com/js/html5.js"></script>
Thanks, Tom
ID attribute validation
Andrés asked:
Does the id attribute allows to begin with a number or a hyphen, unlike previous specs? Is there any set of character that aren’t allowed, besides space character?
There are three rules for the value of the ID attribute:
- It must be at least one character long,
- it must contain no space characters, and
- it must be unique within the page.
So a hyphen would be valid as an ID value. It validates too. However, you may run into legacy stuff that hasn’t been updated yet (HTML Tidy, WYSIWYG plugins for a CMS, etc), so I’d advise testing within your workflow and in a browser (no idea if IE would barf on that, and it wouldn’t surprise me if it did).
peace – Oli
I’m so confused – article, header, footer, nav?
Benjamin asked:
Hey Doctor(s), I’m so confused right now.
I’ve been trying to implement HTML5 for my new blog and CMS, but having difficulties trying to figure out where advanced content should go. All the demonstrations and examples about HTML5 are over-simplified and just don’t include any real world information. For example, an article typically includes (in no particular order):
- Article
- Title
- Tagline
- Tags
- Avatar
- Content
- Links to child articles (eg. download, docs, demo)
- Link Title
- Link Avatar
- Link Author Name
- Links to recommended articles (related content)
- Link Title
- Link Avatar
- Link Author Name
- Article Author Details
- Author Name
- Author Avatar
- Article Details
- Published Date
- Modified Date
- Share Buttons
So all in all, way more complicated than the typical:
- Article
- Article Title
- Article Content
So say for example, where should all the stuff in the complicated (real-world) example go? Should the avatar go in the header, the article, or the footer? Or maybe even a details element? Should the navigation links to other articles be like: nav > ul > li > article > header > a > h1 With the a containing the link, and the h1 containing the article header? Should the post details (despite we want them display perhaps underneath the title, and before the content) be in a details element? And should that be within the header element or be below the header element? Should they use for example:
<dl> <dt class="published">Published At:</dt> <dd class="published">...</dd> </dl>
or:
<ul> <li class="published"> <label>Published At:</label> <span>...</span> </li> </ul>
or something else? The second example seems to remove the duplication of the css class, but seems to add more complications though. Should those tag and article links in nav elements be inside the header? Or footer? Or just article? How many footers and headers can be in article?
I’m soo confused. Any help would be greatly appreciated? Perhaps the best help out there would be a detailed HTML5 example of the real-world article structure I posted above.
OK now, take a deeeep breath and relax. There, feeling a little better? ;)
That’s a mammoth e-mail, so here are some basics to get you started:
- Start with a basic page before attempting something complex — don’t jump in the deep end.
- Look at websites in HTML5 Gallery and check their code.
- Look at the code for this site. Granted, it’s not perfect, but it’s pretty good, and we have a lot of the content you list.
- Use the most appropriate element for each piece of content, even if it’s not one of the new ones. (Hopefully it ends up being what you used before.) Perhaps our handy flowchart might help for that.
Some more specific answers:
“Should the navigation links to other articles be like: nav > ul > li > article > header > a > h1 With the a containing the link, and the h1 containing the article header?”
The title of an article is not the same as an article, so it’d be wrong to use <article>
or <header>
here. I assume you didn’t use <h1>
in navigation lists before, right?
“Should the post details (despite we want them display perhaps underneath the title, and before the content) be in a details element?”
Possibly, if you wanted that interaction, but likely not as you probably want them visible all the time. See what the spec says about <details>
. Decide on the semantics of each piece of content first. Worry about styling later.
“And should that be within the header element or be below the header element?”
Header or footer seem like good places to put them. <label>
is a caption for a form control, not an element to use whenever you need to label something.
I recommend you start with your HTML4 or XHTML1 knowledge and build on it. HTML5 is an evolution, not a revolution. In fact, I’d recommend you start out making a site in HTML4 with the HTML5 doctype (and validate it as HTML5), then change bits over time as you read each article on HTML5 Doctor and learn more ;-)
peace – Oli
Got a question for us?
That wraps up this round of questions. If you’ve got a query about the HTML5 spec or how to implement it, you can get in touch with us and we’ll do our best to help.
3 Responses on the article “Your Questions #14”
I think it’s important to acknowledge that the @type attribute does have at least 2 legitimate purposes: It can be for fallback in case the server isn’t able to send a proper mime type. It can also be used by the user agent to avoid bothering with resources that are using mime types it wouldn’t be able to support.
@JJ: No, @type doesn’t do this. No browser looks at it if you send wrong MIME type.
I didn’t say wrong mime type. I said improper mime type.
For example, if the server were configured wrong and it was sending a strange mime type like “/html” or just the empty string – something that could not, by its very nature, be a proper mime type. When that happens, the browser falls back to the @type attribute.
Join the discussion.