The WHATWG Blog

(X)HTML 5 will have the only usable implementation of ruby markup

August 14th, 2007 by Lars Gunther in Browsers, Elements

One of the misconceptions about the (X)HTML 5 effort is that there will be no noticeable benefits compared to HTML 4 or XHTML 1.0. Here is one thing to prove this wrong. (X)HTML 5 will have the only practically usable implementation of ruby markup. While this maybe won't make too much of a difference in Europe and America, it sure is good news to a pair of billion people in Asia - and a few thousand theologians and historians.

Ian Hickson, the editor of the HTML 5 specification, said on the WHAT WG mailing list just recently: "I have in fact already begun looking at exactly what the parsing and semantic requirements for <ruby> will have to be. It should be added to the spec in the coming weeks." Currently the only browser that natively supports ruby is Internet Explorer, at least somewhat. This should be great news for Microsoft. For once they will actually have a lead compared to Gecko based browsers, Opera, Konqueror or Safari in supporting a part of the HTML 5 spec.

Keith Bowes, in a recent comment on Molly's blog, says:

Personally, I don’t see the point of HTML5. HTML 4 was a big improvement over HTML 3.2: better internationalization, better support for style sheets, more structure and less presentation, some of the more questionable things were removed or put in the dust bin of de facto obsolescence, etc. But I really don’t see where HTML5 is better enough.

A pair of billion Chinese or Japanese speaking people will beg to differ - and a few thousand theologians. Ruby is a big improvement for them. But wait, I may hear. Ruby is not new. It has been part of XHTML 1.1 for many years now. To which I retort: My point exactly! XHTML 1.1 is namely useless! Here is what its spec says: Content must be served with an XML MIME-type, and we all know that one specific browser with a huge market share does not support true XHTML!

If I want to use ruby markup I can (a) serve XHTML 1.1 as text/html, against all rules and have all IE users benefit, or (b) serve it as it should to users of Gecko based browsers, that can support ruby through an extension, or to all CSS savvy browsers using dedicated CSS-rules. But I can't do both!

Since (X)HTML 5 will have two serializations, this problem will cease to exist. I may use ruby markup and still send my content as text/html.

But there is one more lesson to be learned from this. I am a strong advocate of accessibility - after all I am a theologian first and web developer second - but I see that some of my friends have mistrusted HTML 5 because of accessibility features not yet documented in the spec. Ask questions! I did for ruby. If I had gotten a not so positive response I would have argued my case strongly. But please withhold negative judgments for a while. The process is not over yet. The war is not lost.

Finally, how does ruby benefit theologians? As I see things the best way to produce interlinear reproductions of ancient texts - the kind that we grapple with all the time - is to use ruby markup, where the relationship between the ancient word and the translated word is made explicit also within the markup.

Nowe elementy w HTML 5 (Polish) ↔ HTML contre XHTML (French)