The WHATWG Blog

This Week in HTML 5 – Episode 9

October 14th, 2008 by Mark Pilgrim, Google in Weekly Review

Welcome back to "This Week in HTML 5," where I'll try to summarize the major activity in the ongoing standards process in the WHATWG and W3C HTML Working Group.

Most of the changes in the spec this week revolve around the <textarea> element.

• r2305 covers editing a <textarea>.
• r2309 defines the rows and cols attributes.
• r2310 defines the wrap attribute. (Long supported by Netscape and still supported in Internet Explorer, the wrap attribute has never been standardized until now.)
• r2311 defines the maxlength attribute.
• r2312 defines the required attribute, also new in HTML 5.
• r2313 removes support for the accept attribute, which has always been problematic and its (limited) potential has never been implemented. This only affects the <textarea> element; <input type=file> elements still have an accept attribute that controls what types of files may be uploaded.

Shelley Powers pointed out that I haven't mentioned the issue of distributed extensibility yet. (The clearest description of the issue is Sam Ruby's message from last year, which spawned a long discussion.) The short version: XHTML (served with the proper MIME type, application/xhtml+xml) supports embedding foreign data in arbitrary namespaces, including SVG and MathML. None of these technologies (XHTML, SVG, or MathML) have had much success on the public web. Despite Chris Wilson's assertion that "we cannot definitively say why XHTML has not been successful on the Web," I think it's pretty clear that Internet Explorer's complete lack of support for the application/xhtml+xml MIME type has something to do with it. (Chris is the project lead on Internet Explorer 8.)

Still, it is true that XHTML does support distributed extensibility, and many people believe that the web would be richer if SVG and MathML (and other as-yet-unknown technologies) could be embedded and rendered in HTML pages. The key phrase here is "as-yet-unknown technologies." In that light, the recent SVG-in-HTML proposal (which I mentioned several weeks ago) is beside the point. The point of distributed extensibility is that it does not require approval from a standards body. "Let a thousand flowers bloom" and all that, where by "flowers," I mean "namespaces." This is an unresolved issue.

Other interesting changes this week:

• r2314 ensures that the required attribute only applies to form controls whose value can change.
• r2316 defines the name attribute for form controls.
• r2317 defines the disabled attribute for form controls.
• r2320 defines all the different ways that a form control can fail to satisfy its constraints. For example, an <input maxlength=20> element with a 21-character value.
• r2322 defines exactly how form data should be encoded before being submitted to the server. I've previously mentioned character encoding in this series; this revision marks the first time that an HTML specification has acknowledged the existence of <input type=hidden name=_charset_> method of specifying the character encoding of submitted form data.
• r2319 removes support for data templates and repetition templates. These were inventions in the original Web Forms 2 specification, but they were never picked up by any major browser.

Around the web:

• Anne van Kesteren gave an interview on the state of several bleeding edge web standards.
• Simon Pieters ponders what to do about nested <h1> elements.
• In response to Olivier Gendrin, Anne van Kesteren points out that the CSSOM standard defines a window.media attribute. Unlike CSS media types, window.media would be directly queryable from Javascript.

Tune in next week for another exciting episode of "This Week in HTML 5."

This Week in HTML 5 – Episode 8 ↔ WHATWG meetups