The other day we caught up with Todd Anglin, VP of HTML5 Web and mobile tools for Telerik, when we covered Telerik's newly available and excellent guide to architecting and building enterprise mobile apps. We also spent some time discussing the recently "finalized" World Wide Web Consortium (W3C (News - Alert)) HTML5 standard, and Todd had some very interesting thoughts to share on it, where HTML5 is headed next, what is likely to happen over the next year, and what WHATWG adds to the HTML5 standards process from its "living standard" perspective.
TMCnet: What exactly does it mean that W3C completed its HTML5 definition - do we now at long last have an officially finalized HTML5 standard in hand?
Todd Anglin: The words "completed" or "finished" or "finalized" rarely mean "done" in software, and HTML5 is no exception. Instead, what we typically mean is that a specific version or release is finished, and additional improvements will be delivered in the inevitable next iteration.
In this case, the W3C, consistent with their earlier pledge to stick to "versioned" standard specs rather than adopting "living standards" that are in a constant state of flux, has put a cap on what will be part of HTML v5 so that the spec can move through the standards process. It does not mean the work on evolving HTML is complete or that the W3C considers the spec process done. Quite the opposite is true in fact. Simultaneous to "completing" HTML5, the W3C has already begun work on HTML 5.1 to capture the next collection of changes to HTML that didn't make the HTML 5.0 cut-off.
TMCnet: Does the Web Hypertext Application Technology Working Group (far better known as WHATWG) now simply serve as a "drag" to moving quickly ahead on HTML5 development or will there be interesting additions via WHATWG that developers should wait for or at least plan for?
Think of WHATWG as Chrome (constantly evolving) and the W3C as Internet Explorer (evolving in slower, yet more easily definable steps). Both have a place and serve different needs and audiences. What's not captured in the W3C's definition of HTML5 today but scoped out through WHATWG will have an opportunity to be part of the next versioned update they publish.
TMCnet: Is there anything in the W3C's completed definition that comes as a surprise? Or does it merely reflect what most platform vendors (such as Telerik's Kendo UI and Icenium) and developers are already putting to use?
Todd Anglin: There are no surprises. The move to make HTML5 "complete" is largely a formality. It doesn't change the rate of innovation happening in the browsers. More than anything, it simply gives the industry (and the browsers) a way of more easily defining what "HTML5" means at a specific point in time - in this case through the finished 5.0 standard.
Chrome supports much more of HTML5 than IE 10. Does that mean IE 10 is not "HTML5 ready?" Today it's a nuanced judgment call. The W3C's versioning will clearly help the industry limit the scope of the "HTML5" umbrella term and more accurately start to define "HTML5" ready browsers, and then HTML 5.1, 5.2, and so on.
TMCnet: Are there any "issues" still left unsolved or unresolved on the table now that we have this first finalized HTML5 definition?
Todd Anglin: Of course! For HTML (and related Web standards) to reach the point that they are capable of handling all software requirements tackled by "traditional" app languages today (.NET (News - Alert), Java, ObjectiveC, etc.), there are many additional capabilities that must be "standardized."
And this work is happening, in typical fashion, far faster than the spec standardization process. In order for Google, Microsoft (News - Alert), Mozilla, and Intel (among others) to make HTML5 an integral part of their respective platforms (Chrome OS, Windows 8, Firefox OS, Tizen), they are already working on features well beyond the scope of the W3C's HTML 5.0. As this work matures and everyone involved in the standards process agrees on the "right" way to do these more advanced things, we'll see these innovations captured and in turn well-defined (by the W3C, at least) in HTML 5.1 and future iterations.
It's important to remember that Web standards, while important, don't drive the Web platform's evolution. Browser authors have complete control - and it isn't necessarily well understood by everyone in the tech world that they do in fact own this control completely. As we learned in the original browser wars, what ships wins. Standards play the role of the United Nations, trying to keep open the lines of communication between the major players (Google, Apple, Microsoft, Mozilla (News - Alert)) so we don't have another browser war, with competing implementations of the same feature set. That hurts the entire Web platform. The truly good news for now at least, is that everyone is committed to playing by the UN's (make that the W3C's) "rules."
TMCnet: Recently there have been folks - notably the CTO of Plantronics (News - Alert), among others - that have suggested that WebRTC protocol should become part of the official HTML5 standard. Any thoughts on this? Good idea? Bad idea?
Todd Anglin: In truth, that is a red herring. It doesn't matter if it's part of the "official" HTML5 standard or not. What matters is that the WebRTC protocol is standardized and that the browser authors - again it’s the browser authors that have the control - implement that standard. Whether that happens via the core HTML5 standard or as an independent standard is a bit irrelevant, other than perhaps simplifying communication - in other words, if saying for example that a browser is "HTML 5.1" compliant guarantees WebRTC support that is always a good thing.
And really, WebRTC is a good example of where there is already broad agreement and implementation of an HTML5 feature not formalized in the current W3C snapshot. Chrome, Firefox, and rapidly evolving browsers offer relatively consistent support for WebRTC today. The biggest missing player is Internet Explorer, but that is a common refrain for many of HTML5's most advanced features!
All in all 2013 should be yet another extremely interesting year for HTML5.
We certainly agree with that!
Edited by Brooke Neuman