First, the Proposed Recommendation. W3C develops specifications, like HTML 5.1, and when they are “done”, as agreed by the W3C, they are published as a “Recommendation”. Which means what it says – W3C Recommends that the Web use the specification as a standard.
HTML 5.0 was published as a Recommendation a bit over 2 years ago. It was a massive change from HTML 4, published before the 21st Century began. And it was a very big improvement. But not everything was as good as it could be.
A couple of years before the HTML 5 Recommendation was published, a decision was taken to get it done in 2014. Early this year, we explained that we were planning to release HTML 5.1 this year.
There is an implementation report for HTML 5.1 that shows almost all of the the things we added since HTML 5.0 are implemented, and work out there on the Web already. Some things that didn’t work, or did but don’t any more, were removed.
HTML 5.1 certainly isn’t perfect, but we are convinced it is a big improvement over HTML 5.0, and so it should become the latest W3C Recommendation for HTML. That’s why we have asked W3C to make it a Proposed Recommendation. That means it gets a formal review from W3C’s members to advise Tim Berners-Lee whether this should be a W3C Recommendation, before he makes a decision.
Meanwhile, we are already working on a replacement. We believe HTML 5.1 is today the best forward looking, reality-based, HTML specification ever. So our goal with HTML 5.2 is to improve on that.
As well as fixing bugs people find in HTML 5.1, we are working to describe HTML as it really will be in late 2017. By then Custom Elements are likely to be more than just a one-browser project and we will document how they fit in with HTML. We expect improvements in the ability to use HTML for editing content, using e.g.
We have wanted to make the specification of HTML more modular, and easier to read, for a long time. Both of those are difficult, time-consuming jobs. They are both harder to do than people have hoped over the last few years. We have worked on strategies to deal with making HTML more modular, but so far we have only broken out one “module”: ARIA in HTML.
We hope to break out at least one substantial more module in the next year. Whether it happens depends on sufficient participation and commitment from across the community.
We will further improve our testing efforts, and make sure that HTML 5.2 describes things that work, and will be implemented around the Web. We have developed a process for HTML 5.1 that ensures we don’t introduce things that don’t work, and remove things already there that don’t reflect reality.
And we will continue working to a timeline, with the HTML 5.2 specification heading for Recommendation around the end of 2017.
By which time, we will probably also be working on a replacement for it, because the Web seems like it will continue to develop for some time to come…