It's official: W3C finalises CSS 2.1

Cascading Style Sheets Level 2 Revision 1 (CSS 2.1) Specification (or CSS 2.1 to its friends) has become a real boy, with W3C stamping its seal of approval and making the spec a W3C Recommendation. But in an age of rapidly iterating browsers that are already working hard to win the race regarding CSS3 compatibility, is the W3C now an anachronism? Standards advocates don’t seem to think so.

Eric Meyer, partner and co-founder of An Event Apart says that while the formalisation of CSS 2.1 has been a “long, twisty road”, he’s happy we’ve finally reached its end: “It's important because CSS 2.1 is a specification that describes, as best it can, what's done in practice as opposed to what might be nice in theory”. Meyer argues that it’s a good idea from time to time to document how things work, “both for implementors who come to the game later and authors who want to understand how CSS works”.

Easy! Designs, LLC principal and author of Adaptive Web Design Aaron Gustafson agrees that specs remain relevant in a world of rapid iteration: “At first, I was against the movement of the WHATWG away from a versioned spec because I was concerned the W3C would follow suit. I came to realise, however, that what was happening was a separation that was mutually beneficial: the WHATWG could work on charting the course for the future while the W3C worked on placing the milestones along that course”.

Gustafson argues milestones remain important, for seeing how web languages have evolved and for testing the conformance of browsers and authored documents. “Without them, you’ve no ability to see if you're doing something correctly or not,” he says.

A consistent approach

Web designer Lynne Pope also argues that milestone specifications remain essential to ensure the web doesn’t become fragmented. “Draft specs let us know what is intended and having the proposals out in the open, being able to follow changes, means we don't get any sudden surprises. As we can see with HTML (5, that is) and CSS3, browsers are implementing different features in different ways. Currently, the only way to get close to reasonable cross-browser HTML (5) support is by using JavaScript to compensate for inadequacies”.

Pope says that once standards become a recommendation, everyone knows browser developers have comprehensively tested their code and been reasonably satisfied that they can reach compliance with them: “100 per cent compliance doesn't always happen, but without the browser and application vendors working together to agree a standard cross-browser, design work would become intolerable”.

Looking to the future

With the CSS 2.1 chapter now closed, it’s also certain that the consistency Pope yearns for regarding CSS3 will be a major focus for browser developers, but she also notes that it’s essentially been 13 years since the previous recommendation (CSS 2, in May 1998). However, it’s unlikely that CSS3 will suffer the same fate.

“I’ll predict that CSS 2.1 is last CSS standard to take so many years to develop,” thinks designer, author and speaker Andy Clarke. “Instead, we’ll see CSS3 features implemented in browsers, CSS modules created, worked on and finalised in months, not years”.

Gustafson agrees: “What’s important to realise about the spec process is that a given spec does not become a recommendation until there are at least two interoperable implementations of each feature. With really large specs, it’s difficult to attain that. That's one of the reasons CSS3 is broken into modules. That way each module can move at its own pace as opposed to slowing down the development of future specs.”