Somewhere in building 50 on Redmond campus, the IE team is wondering why nobody ever said this in their defence when the exact same scenario was playing out with IE (dominant browser, not available on all platforms, glacial standards bodies, browser specific extensions, etc).
This is a very different situation in quite a few ways. For one, IE was not just a big browser, but was almost completely dominant. For another, many of IE's browser-specific features were Windows-only by design, such as ActiveX.
Here, we have an open source rendering engine that clearly delineates browser-specific features (each of which is well-documented). The use of a prefix makes it possible to use the browser-specific features offered by WebKit without sacrificing compatibility with other browsers, something Microsoft actively resisted. For quite a while, it was very difficult, if not impossible, to make sites that looked and felt right both in IE and in other browsers.
Had Microsoft extended standards in a way that played nice with the rest of the community, maybe they would have had someone saying something like this in their defence. Instead, they used browser-specific features as a way of undermining standards.
I don't think Microsoft was acting "not nice" - most of their extensions were submitted to the W3C (and even ActiveX was standardized by X/Open, however pointlessly). However, the major vendor, Netscape, simply was not going to adopt any IEism under any circumstances.
The biggest issue with being frozen on IE6 was not so much the standards but all the terrible bugs.
Keep in mind that at the time, Netscape was even more proprietary than Microsoft. (Layers instead of DOM, JavaScript SS instead of CSS)
And if you want to get technical about it, in theory you could support ActiveX on any platform. The problem was the controls were all Win32 software. There probably would have been some marginal benefit for Netscape to support it on Windows.
NaCl is specific to individual processor architectures by design, much like ActiveX was specific to Windows.
Prefixed WebKit features are certainly not consistently well-documented. CSS animations, transforms, etc. had to be carefully reverse engineered by other browsers.
CSS animations and transforms were released along with formal specs. The Webkit developers read mailing lists, hang out on IRC, etc. If anyone thought these were underspecified or ambiguous, they could easily ask for answers or adjustments to the spec. It’s not like the Webkit folks were trying to hide anything.
You’re right. CSS animations were announced and added to Webkit betas in October 2007, and it took until the beginning of April 2008 for Apple to put up their proposed formal spec. I’m not sure when they first hit a publicly released version of Safari. Still, 6 months is pretty short in browser spec time, and pretty much no one was using these features in the wild until much later.
As others have said, what Microsoft did was arguably worse, but I generally agree: Apple and Google are industry darlings, and because of this are getting away with things people would be crying bloody murder over if Microsoft did them.
I get the distinct sense that a lot of people don’t really care about web standards so much as they care about their favourite company controlling standards. If it were Microsoft who was forging ahead with new features, would the MacBook-wielding, Gmail-using technorati be so enthusiastic?
I’m not even entirely against the new -webkit- stuff, I just hate the hypocrisy from people who claim to be all for web standards, except for when there’s a shiny new feature that only works in Safari, Chrome, and iOS. Microsoft has arguably been doing the best job of respecting web standards since IE9, and they get very little credit for it.
I think you're right. Today I've been reading the exact opposite of what people were saying about IE a decade ago.
I just hate the hypocrisy from people who claim to be all for web standards, except for when there’s a shiny new feature that only works in Safari, Chrome, and iOS.
I'm right there with you. For a while I've been coming to realize that the standards never actually mattered for these folks, its just that "IE6 doesn't conform to standards" made a more convincing argument for firefox than the real reason which was "IE6 is a pain in the ass and I hate it."
Now the people who want the new hotness are realizing that the standards they tied themselves to have become an albatross. If they actually ever gave a shit about standards, they would have been pitching ideas about how to update the standardization process to work with today's implementation realities. They actually would have been doing it along the way, and we would have never ended up where we are now.
I may be taking this too far, but I think there’s more to it than IE. Many in the “hacker” community seem to wrap themselves in the flag of “open”, but only when it’s in their favour.
It’s the people who argue for a decentralized, open-source alternative to Facebook, but embrace Google+. It’s the people who explain how terrible it is to tie your business to a single vendor, but decide to build businesses on iOS. It’s the people who tell you they would never put any personal documents on the cloud, but embrace Dropbox. It’s the people who claim to support open, DRM-free data formats and think the Khan Academy is the future of education, but think iBooks textbooks are great.
Like I said, I don’t have a particularly strong attachment to open source and open data formats — I just hate the way people use openness and standards as a prop. Saying standards aren’t necessary when WebKit has neat new features is (loosely) analogous to only supporting democracy when the party you vote for wins.
It wouldn't have been a problem if they had done so the way that WebKit, Opera and Mozilla developers have done so. By prefixing it clearly lays out that it is a "vendor" specific property implementation.
That implementation can be discussed, modified or accepted and become a standard non-prefix property. It's happened a few times where the WebKit implementation has had to be reversed to match another vendor's approach. This is good.
It has gotten us closer to using CSS3 in a shorter period of time than we have with any W3C standard.
The way that both Microsoft and Netscape did it in the 1990s was not only the wrong way to do it, it was a malicious approach aimed at segmenting the web.
And Microsoft's biggest sin was that they froze IE at version 6 for over 5 years, ensuring that their mishmash of standards and proprietary implementations would weigh on the Internet.
