The following InfoWorld article spread like wildfire throughout the web yesterday — W3C: Hold off on deploying HTML5 in websites. The story was prompted by comments from W3C official Philippe Le Hégaret. The community has been in uproar although I suspect many of Le Hégaret’s quotes were used selectively or taken out of context. Even so, let’s go through each statement made in the article…
The problem we’re facing right now is there is already a lot of excitement for HTML5.
Agreed. There’s also much hype and confusion. To most people, HTML5 means “the future of the web”. You can take a 5 year-old quote about “Web2.0” or “Ajax” and interchange it with “HTML5” today. The mainstream media neither knows or cares about the technical realities and there’s little point trying to explain that CSS3 != HTML5.
It’s a little too early to deploy it because we’re running into interoperability issues, including differences between video on devices.
I don’t think it’s ready for production yet, especially since W3C still will make some changes on APIs. The real problem is can we make [HTML5] work across browsers and at the moment, that is not the case.
Le Hégaret is quoted as saying one browser is not the same as another. Is that news? We’ve struggled with differences since the dawn of the web.
There are 4 major rendering engines: Trident, Gecko, Webkit and Presto. The chances of each matching the other feature-for-feature are nil. Browser changes are a fact of life on the web — it’s not a problem with HTML5.
HTML5 is viewed as a game changer. Companies now can deploy HTML5 in their applications or in intranets where a rendering engine can be controlled, but it is a different story on the open web where interoperability is an issue.
Interoperability always has and always will be an issue. It’s easy to support one browser but the best developers can make a site work on any device, whether it’s Firefox 4.0, a screen reader, mobile device, Lynx or IE1.0. It’s not always worth that effort but HTML5 does not change the process. Many sites use XHTML1.0 or 1.1, yet neither is supported by IE8 and below.
Of course, developers should use common sense. If you’re developing a system where 90% of your audience are IE6 users, don’t expect the HTML5 video tag to work. You’ll have fewer issues with other elements, such as header.
The HTML5 specification itself features support for video and Canvas 2D. But other technologies such as CSS and MathML are considered part of the “open Web platform” along with HTML5, even if they are not covered by the actual specification. SVG is referenced by the HTML5 specification.
This quote is most telling. It’s a long meaningless statement in the middle of the article, but what point is it making? Is Le Hégaret saying CSS and MathML can or cannot be used because they’re separate from HTML5? Or has the reporter simply plucked a random statement from the interview?
We’re not going to retire Flash anytime soon. It will take years before all Web clients support HTML5. IE6 is still being used on the Web today and it is 10 years old.
Over time, however, HTML5 will become the standard for websites and you will see less and less websites using Flash.
Ahh, the good old “HTML5 is a Flash replacement” statement. Was this started by Apple? If not, their marketing department must love it!
I’m not sure why the article needed to mention Flash, but HTML5 is not a Flash replacement! The video tag and newer CSS3 features may reduce reliance on Flash, but that doesn’t mean Flash can’t or shouldn’t be used. Flash will offer the most reliable cross-browser video medium and gaming platform for many years to come.
Digital rights management also is not supported in HTML5. This means some video producers will not deploy their videos in HTML5 without this type of protection.
If we are going to develop a solution for DRM which is open, it would be broken by a hacker within two days. There is a possibility for DRM in HTML5 at some point, however, but it is not in the plan at the moment.
Is this for real? An open standard is, by definition, open. DRM has never been a feature of HTML. If you want to control media distribution, use an alternative such as Flash or Silverlight — like you’re doing now.
HTML5 also lacks authoring tools at the moment.
What about Notepad? It’s true that few WYSIWYG editors produce HTML5 but few produce decent HTML4. The best developers use a text editor or IDE. There may be a few color-coding quirks and validation difficulties, but there’s little to stop you writing HTML5 code today.
Where are we going?
Whereas Web2.0 and Ajax were vague terms, HTML5 refers to a real set of technologies with specifications and version numbers. The InfoWorld article succumbs to the belief that, at some point, HTML5 will be “complete” and ready for production. This is not the case.
Browsers continually evolve and new features are ultimately copied, tweaked or rejected by others. If at least two vendors agree, a single feature may become part of a W3C specification — whether it’s HTML5, CSS3, SVG, MathML or whatever. However, a W3C Recommendation does not automatically mean the facility will appear in other browsers. Fortunately, there are always workarounds, fall-backs, or shims when we need them.
HTML5 is not a destination — it’s the journey. We’re not sure where we’re going, but at least all the browser vendors are on the same bus.
Craig BucklerCraig is a freelance UK web consultant who built his first page for IE2.0 in 1995. Since that time he's been advocating standards, accessibility, and best-practice HTML5 techniques. He's created enterprise specifications, websites and online applications for companies and organisations including the UK Parliament, the European Parliament, the Department of Energy & Climate Change, Microsoft, and more. He's written more than 1,000 articles for SitePoint and you can find him @craigbuckler.
