> My high-level take on the past year is that the Blink project has been more focused on next-gen webapps with a heavy focus on the compositor, scheduling, and style subsystems. The WebKit project has been more focused on documents and improving existing pages with faster line layout and style selection (as well as an enormous amount of great work on JSC and bindings).
That was always the case, even when they both used WebKit. Apple didn't include many of the "next-gen webapps" stuff that Google implemented, like IndexedDB. They were always more focused on CSS, for example.
That was a good balance of power for WebKit, IMHO. One company was focused on web apps, the other on native apps, but they both benefited from having a cross-platform renderer. With the big contributors pulling in different directions sometimes, WebKit tended to settle on a happy middle.
Hopefully there's a silver lining to the Blink fork: more differentiated renderers in the wild means that web developers will have to rely more on standards again, instead of going nuts with "-webkit" prefixes and Chrome-only apps.
Ir's true that Blink project leaders have expressed a disinterest in improving web documents relative to web apps. But the WebKit project is very interested in both document-centric and app-centric uses of the app platform. We are frankly a little confused at the idea that there is a conflict. The best platforms include awesome document support. It's true that the Web is a little quirky in this regard - it's built inside-out with the app layer embedded inside the document layer, instead of vice versa. But that is part of what has helped it be more successful.
But actions speak louder than words. Some awesome webapp-focused stuff that's been just recently announced: IndexedDB, WebGL, JavaScript Promises, Media Source Extensions, Web Crypto, a vastly improved JavaScript VM that scales from super quick startup for simple pages to advanced optimizations for complex webapps, and massive optimizations for webapp responsiveness (covering DOM, rendering, layout, style, JS, etc). And that's just the announcements, there is a lot more cool stuff in WebKit nightlies like HTML templates, new ES6 language features, major web developer tools improvements, and more.
With all due respect, IndexedDB was first developed like 5 years ago at this point. WebGL over 3. It's fine that you have different priorities, but you should just admit that you have different priorities instead of saying you want both while letting webapp features lag behind by several years compared to Chrome and Firefox.
We've had support for WebGL internally for years. The decision to ship it was based on sufficiently improving graphics driver security. For most of webgl's existence, Apple engineers have been editors of the spec. For IndexedDB, we originally wanted to keep pushing for WebSQL, and still believe IndexedDB is technically inferior. But we are setting aside our personal opinions in the interests of interoperability.
There are other features we have shipped well before Chrome or Firefox. Everyone is first at something. Eventually the browser engines converge. That's how the browser market works.
I know you guys are first on some things, but as I said those things generally are CSS related. I can't think of any web-app enabling current generation features that Apple engineers are editors of the spec. Not WebRTC, not Manifest, not Full Screen mode.
If I'm wrong, please list these specs. I'm not questioning your abilities, just your priorities. JavaScript Promises are not webapp-enabling so that's irrelevant. Does your IndexedDB allow storing of Blobs? That's important for many classes of web apps (any that need to store large amounts of arbitrary data).
Would you consider Canvas, HTML5 Video, CSS Transitions and Animations, Application Cache, or WebSQL to be web-app-enabling features?[1] We were first to ship all of those. And they sure as heck aren't for documents primarily.
Editing a spec is a separate question from shipping order or general engineering priority. Not sure why you are conflating the two. Sounds like moving the goalposts.
I'd encourage to test IndexedDB in a WebKit nightly or in the Yosemite beta. It's still kind of buggy and not necessarily complete for final, but we'd love to hear about the key blocking issues from web developers.
[1] Yes, it seems like AppCache and WebSQL are both likely to be replaced by newer technologies, but I don't think that's relevant to whether sufficient priority has been placed on the area of webapps.
I bring up editing spec because it is an indicator of priorities. Apple does edit specs, just not the type we're talking about here. As for your list of features you guys did first, that's great and I'm appreciative of the work, but those are all pretty dated at this point. We're talking about today's web, not that of 5+ years ago. Anything in the last 2 years? We can both list off a dozen from Google and Mozilla.
That was always the case, even when they both used WebKit. Apple didn't include many of the "next-gen webapps" stuff that Google implemented, like IndexedDB. They were always more focused on CSS, for example.