Starting in June, the W3C HTML WG began using Bugzilla for tracking of detailed specification issues. The following is a summary of changes for the week of June 15th, 2008.
The following bugs were CLOSED:
priority request: reply to outline e-mail
Ian Hickson opened and closed his own bug but it originated via a message to public-html from James Graham back in March 2008. James was primarily concerned about this language:
Otherwise, if the element being entered has a rank equal to or greater than the heading of the current section, then create a new section and append it to the outline of the current outline element.
Ian responded to this message on June 17th, 2008:
The first “otherwise” section was too aggressive, which, as you point out, prevented the second “otherwise” clause from affecting the bits it was supposed to.
There was some additional discussion surrounding the outline section as Geoffrey Sneddon was writing a test implementation algorithm. Ian responded to Geoffrey with both information and acknowledgement of some specification changes. There was also some discussion between Geoffrey, Ian, James and Philip Taylor via IRC between June 12th - June 19th.
The editor’s draft of section 4.3.10.1 includes the corrected language.
Newly introduced void elements are not mentioned in “Writing HTML Documents” section
Lachlan Hunt noted that the list of void elements was incomplete and missing:
command, event-source, source. The draft of section 9.1.2 has been updated by Ian Hickson.content sniffing should be allowed on link elements with a relationship type of icon
Philip Taylor did a bit of research on existing usage of content-types with favicons and found many inconsistencies. Philip suggested that
<link rel=icon>resources should be treated the same as those sourced from an<img>element. Philip originally brought this issue to the attention of the editor via IRC on June 18th.Ian Hickson updated the Link type “icon” section with additional language that user agents must expect the resource to be an image and that they should follow the Content-Type sniffing rules for images.
HTMLMediaElement.currentLoop readonly?
Simon Pieters asked if current
Spec says on the HTMLMediaElement interface:
attribute unsigned long currentLoop;Shouldn’t currentLoop be readonly? If not, what happens on setting?
The current working draft has this text:
When any of the start, loopStart, loopEnd, end, and playCount DOM attributes change value (either through content attribute mutations reflecting into the DOM attribute, or direct mutations of the DOM attribute), the user agent must apply the following steps:
Ian Hickson updated the Offsets into the media resource section with:
When any of the start, loopStart, loopEnd, end, playCount, and currentLoop DOM attributes change value (either through content attribute mutations reflecting into the DOM attribute, if applicable, or through direct mutations of the DOM attribute), the user agent must apply the following steps:
Parsing should be specified for future updates
Robert Burns proposed a solution for handling unknown elements. Ian Hickson and Simon Pieters noted complications with existing content and implementations. Philip Taylor followed up with several examples from real-world data. Robert acknowledge Philip’s data as useful but believed it represented a minority of existing web pages. Later, Lachlan Hunt noted that Robert’s proposal would complicate matters from an author’s perspective. Ian also cited an unwillingness from browser vendors to make significant changes to their parsing logic.
Robert disagreed with Ian and asked for a chair to step in with guidance. Co-Chair Michael(tm) Smith closed the bug (WONTFIX) based on the editor’s technical assessment and the lack of a clear business value proposition for browser vendors to implement the proposal. He suggested that Robert make an appeal to his fellow working group members and, in particular, browser vendors on public-html and related mailing list.
If you are looking to help out, the following bugs are marked as NEEDINFO:
- MediaElement features are needed on OBJECT element
- Authors need more control over handling of embedded resources
- ID value types too restrictive and inflexible for some use cases
- Authors need more control over handling of linked resources
Please email any corrections to this summary to Shawn Medero: soypunk@gmail.com.