This 30 minutes session (presented at the Professional Developer Conference) focuses on the “behind the scenes” aspects of HTML5 and the efforts that Microsoft is engaged in with the W3C Working Groups.
“HTML5” is used today as an umbrella term for anything that is related to “new web standards”. This is not correct, since technically HTML5 is a well-defined specification – and when used as an umbrella term “HTML5” covers a lot of other documents with their well-defined names. Although I don’t like this approach, I have to admit it’s quite convenient when we need to refer to the entire universe of web-related standards. The alternative would be to call it “HTML5 and Co.” or “HTML5 and SVG and CSS and Web Apps and – and –” or " HTML5 and everything that relates to it”. Or you can provide any other name you’d like. Personally, for brevity, I will use “HTML5” in quotation marks when I need to refer to the whole set of new standards and plain HTML5 to refer to only that specification.
The important point is not really the name, but the understanding that this is a large effort…and that there are many people and Working Groups working on several equally important specifications. The future of the Web is not just one of these documents; it’s all of them. Actually, it’s most of them cooperating together.
To me, the real beauty of these standards comes when you can apply CSS stylesheets to HTML5 elements using SVG markup serialized with WebApps API. (btw, we will see an example of this in a future post )
Microsoft is committed to “HTML5” as a whole: HTML5, CSS3, SVG, Web Apps, and ECMA Script 262. We are investing a lot of resources (budget, time, but most of all people) in W3C and are participating in several Working Groups (WGs), providing feedbacks across the board. For instance, Paul Cotton (chairman of the W3C HTML WG together with Sam Ruby from IBM and Maciej Stachowiak from Apple) and more than 15 of our own Program Managers discuss these important topics daily.
As a developer, I expect to be able to write my code once and get consistent results across all browsers: same markup, same code. If a browser fails to run that code, there is an interoperability issue that needs to be solved; it can be either a browser bug, or an interoperability challenge that needs to be discussed and fixed in one (or more) of the W3C WGs.
In the recent years, starting with Internet Explorer 8, the Internet Explorer team has been working with W3C to build comprehensive test cases, thousands of them! Our goal has always been the same: interoperable specifications and implementations; solving any misunderstanding or misalignment between implementations before they’re deployed; making sure that different User Agents treat the same markup the same way; letting web developers live the dream of writing once, running everywhere.
Given the nature and complexity of “HTML5” itself (there are thousands of pages just in the HTML5 specification!), testing is not binary. There are many factors involved! For instance, checking if a browser supports the <Canvas> tag doesn’t tell you if that browser implements correctly all of the Canvas features. Writing 100 tests that try to cover all of the “HTML5” standards (including those that are still Working Drafts and subject to change in the future) will not help developers, or browsers, or people writing the specifications. Giving “bonus points” will not prove how good or bad a browser is at supporting standards.
If you would like to help us and W3C move these specifications forward, I invite you to participate in one of the Working Groups. It’s just incredible what an impact you can have there!
There are many very interesting specifications. I’m looking forward to the day when a web app will be able to do “pretty much anything it wants”. In the meantime, it’s our responsibility as a browser provider to make sure that whatever standards we implement are solid. We don’t want to repeat mistakes made in the past; we don’t want to implement unstable specifications that might change in the future (thus breaking your sites). We don’t want to implement specs that don’t make a lot of sense (for example, SVG Fonts compared to CSS3 Fonts) just to score 100% on Acid3. We don’t want to support new APIs, until we nailed and solved any possible security and privacy issues. This is not a rush to support everything as soon as possible. It’s critical to make it right.
HTML5 made lot of progress in recent months, the HTML5 specification expected to go to Last Call (kind of feature complete) in the first 2-3 months of 2011. From there, the spec will move to Candidate Recommendation and there will be a call for implementers.
The other specifications are at different stages; some (for example ECMA Script 262 or SVG) are already Recommendations (with the capital R). Others (for example CSS3 Colors or Web Apps Selectors API) are in Last Call or CR. Others (for example IndexDB or File API) are still in Working Draft or First Public WD.
You! This is a great time to start learning and experimenting with some of the new capabilities. I’m amazed by the HTML5 websites that are already in production today. I hope they will inspire you to build the next amazing web application!
Feedback? Questions? Please let me know!
You don't answer your own questions.
Do you support all of html5? No.
When will html5 be ready? You don't know.
Is html5 interoperable? no. It is meant to be but it won't be in reality.
I'm not sure to understand your point.
What is "all of html5" for you? Why is this important to you?
I don't know when HTML5 will become a Recommendation. Actually, nobody knows. It depends from many factors. If you are interested, you can follow the status report here: dev.w3.org/.../2010-03.html
HTML5 is way more interoperable than any previous specifications. IE team is working hard to create an interoperable implementation. I'm sure all other browsers care about interoperability as well, and I hope we will keep working together on solving any eventual ambiguity. Interoperability is not given by 1 person or 1 team, but is given by many people working together. It's too early to measure the success of HTML5; today I'm very positive about its success. We will need to see in a few years how it will go. I wouldn't judge too early... ;)
I saw your presentation at codebits and was plain awesomeness.You're a great speaker. Congrats.
Thank you Carlos :)