Faruk At.eş

Opera Confirms WebKit Prefix Usage

Opera, the browser with a small desktop market share but huge mobile market share, particularly in developing and less-opulent countries, has confirmed today that they will start implementing support for the -webkit- prefix.

Their argument, roughly summed up, is that too many websites have started to take advantage of new HTML5 and CSS3-features in a way that only benefits those users with WebKit-based browsers—Chrome, Safari, and every iOS and Android device’s default browser—and that the authors of those websites have done a “lazy”[1] job not including a standards-compliant version of features. Basically, their complaint is that authors have written this:

selector {
    -webkit-transform: rotate(5deg);

And not this:

selector {
    -webkit-transform: rotate(5deg);
    transform: rotate(5deg);

Or even “better”, this:

selector {
    -webkit-transform: rotate(5deg);
    -moz-transform: rotate(5deg);
    -ms-transform: rotate(5deg);
    -o-transform: rotate(5deg);
    transform: rotate(5deg);

Understandably, authors are disinclined to write out a property five times and have to update it similarly many times for every change. This is where CSS precompilers come into the picture with their excellent mixin solutions that solve this problem.

For browser vendors that aren’t named Apple or Google—the two companies developing WebKit and their respective browsers using the WebKit engine—it has become a situation of being excluded from the cool, fun party that all the WebKit-users are going to. This prevents those other vendors from being able to offer a top-notch experience to their customers whenever they visit a website that is, as Opera argues, “lazily authored”.

Opera’s precedent is opening the door for Mozilla (Firefox) and Microsoft (IE) to do the same, meaning that for authors—the group already accused of being too lazy to do the right thing—the Web is about to become an ever-more fragmented mess, where the once-safe experimental feature known as -webkit- (and -moz- etc.) now will come to represent a much more fragile, questionable feature or implementation.

What this results in, however, is that “lazy web devs” now have even less incentive to do the right thing and test their work in Opera, because the onus has now fallen to Opera entirely to make sure their browser is doing everything right. Opera must now not only support -webkit- and offer feature-parity as best as possible, and as fast as possible too, it must do so without any bugs or hiccups or glitches in their implementations.

Browsing experiences and implementations being different between actual WebKit browsers themselves, this kind of browser-engine-and-prefix fragmentation is absolutely the last thing I would want to have to deal with as an author of websites.

Regrettably, Opera—and perhaps Microsoft and Mozilla, too—put the blame on authors for being “too lazy to do the right thing”, without seemingly having tried to solve the problem the right way themselves. After all, it isn’t a shortcoming in developers’ willingness or efforts, so much as it has been, all along, a shortcoming in the tools we’re forced to work with. Our tools are incredibly out of date and insufficient to deal with the complexities of today’s World Wide Web.

If Opera, or Microsoft or Mozilla, would have put their full weight behind a set of open source tools that work better than hand-coding websites, tools that produce the right output for CSS, they would have actually been tackling the real source of the problem. Instead, they’re accusing authors of not collectively and massively supporting their browsers when, to be quite blunt, those browsers were too slow and too far behind WebKit at a crucial moment in time.

I’m left feeling that this is just Browser War II, and what grace we’d regained by switching to Feature Detection rather than UA sniffing will be lost once again as a result of these moves.

With some luck, things will just become better for the users of those browsers who will once more pretend to be someone they aren’t, and us web authors aren’t inconvenienced too much. Meanwhile, it is left—again—up to web authors to invent the tools to placate the browser vendors who gave us this mess in the first place.

  1. Not a direct citation of anyone from or representing Opera, but it’s basically the sentiment conveyed. “Lazy” is perhaps the nicest of the possible interpretations.

About me

Faruk Ateş

Faruk Ateş does creative things on the Web, like Modernizr. He lives in San Francisco and writes and speaks about technology, design and business.

Read more about Faruk, or .

Upcoming talks

Here on My own website

Subscribe to this site