W3C Blog

The Openweb Platform Weekly Summary from January 21 to January 27 is written by Karl Dubost.

<main> element landed in Firefox

We mentioned last week that <main> element was included in HTML 5.1. It has now landed into Firefox code.

Mobile first and/or Desktop first for img@srcset

A while ago, a bug had been opened because the current syntax in HTML for responsive design is not allowing developpers to match their current practice in CSS with media-queries.

The img@srcset does not afford developers the ability to define the breakpoints for images as either minimum values (mobile first) or maximum values (desktop first) to match the media queries used in their design.

The discussion is still going on.

[css3-ui] defining 'cursor: auto' properly

David Baron thinks that cursor: auto is loosely defined. Indeed the current definition might lead to interoperability issues, the usual "free to implementations". David suggested a better definition.

HTTP 2.0: binary or text only

What should be chosen for representing HTTP/2.0 messages? A text-only format or a binary format. There is a full thread discussing about it on the HTTP working group mailing-list. Mark Nottingham says

a textual-protocol was a nice-to-have, but that it shouldn't be a determining factor in design.

I.e., if you can get everything you need out of a protocol, and make it textual, do so, but if it detracts from the value you get from it, don't let that constrain you.

And Roberto Peon comments:

Text formats are, surprisingly, not easier to debug in my experience.

The full thread is an interesting read and give hindsights on development issues of new formats with the usual trade-offs.

Hardware timestamps for events

In May 2012, a discussion had been started about having the possibility to precisely attach a timestamp to an event. Since then the WebKit project is playing with the idea of having this feature. There are discussions on where Event.highResolutionTimestamp should be added, possibly hr-time.

XML declaration part of the DOM… or not.

With DOM4, it is, currently, not possible to extract the XML declaration. But is it really part of the DOM? Daniel Glazman is pushing for it to be part of the DOM and then have the necessary API to access it. On the other hand Henri Sivonen is advocating the opposite arguing it is a serialization artifact and not really part of the Object Model. Björn Höhrmann noted that

Opera, Firefox, and Webkit […] retain the standalone declaration

HTTP headers statistics

James Snell has started a survey for HTTP headers in the context of HTTP/2.0 compression. He is explaining in details on his blog. Understanding the variability of HTTP headers in between requests will help defining the compression efficiency.

rNews RDFA in BBC News

Jeremy Tarling working at the BBC explained that they were in the process of implementing RDFa in the BBC news page.

CSS reset to default values

There is a new shorthand property, all, which has been defined in CSS for resetting everything to default values. For example,

.foobar { all: default; }

HTML Templates

How many times when creating a Web applications, you wanted a way to create an HTML prototype without affecting the rest of your document. The HTML template specification is here for answering this use case. A first public Working Draft is being proposed.

Styling HTML placeholder attribute

Mounir Lamouri is giving a good overview of current implementations:

- Webkit: ::-webkit-input-placeholder pseudo-element;
- IE10: :-ms-input-placeholder pseudo-class;
- Gecko18-: :-moz-placeholder pseudo-class;
- Gecko19+: ::-moz-placeholder pseudo-element;
- Presto: nothing.

In the long thread, the CSS mailing-list participants are discussing if it should be a pseudo-class, a pseudo-element and it should have a different name, but everyone agrees that it should be formally standardized. A summary of the ideas for styling HTML placeholder attribute has been written on the CSS wiki.

HTTP, Media Queries and CSS for Responsive Web design

Henri Sivonen started a long thread about Media Queries and optimizing what data gets transferred. Basically, a Google engineer is trying to see what could be done at the HTTP level. Henri Sivonen is proposing that it would be better solved at the client level.