Introducing WebAPI
hacks.mozilla.org
hacks.mozilla.org
Apple's strategy for their phone platform runs deeper then just what is in the open source Webkit codebase.
I used to think/hope open platforms would ultimately win in the market but iOS and the Apple App Store has kind of proved otherwise.
If people want to run web apps on iOS devices Apples is more than happy to sell them for that purpose.
The most closed thing I can think of is when they only enabled their faster JS engine for the browser itself and not for apps. The next major release changed that, and it is not related to WebKit itself.
Apple is actively (Frequent commits) working on fixing this problem. The project is called Webkit2, and features a Javascript interpreter that runs in a separate process, and thus is able to safely execute Javascript on iOS. This model, btw., is inspired by the multi-process architecture found in other major browsers.
For one, i ditched the kindle app ever since i discovered read.amazon.com .
3 to 6 months is a fairly short timeline for getting all browser vendors and the W3C on board, not to mention getting stable and secure implementations into the wild. At least they're doing all the legwork by putting together the initial specification drafts and implementation prototypes. These additions would be truly incredible if accepted. I know they like to lump every standard under HTML5 these days so I wonder if this is intended to be an addition to that or if WebAPI is completely orthogonal to HTML5.
As far as the specs go, most of the things they list have W3C APIs at reasonably advanced stages. I keep track (somewhat) of the Camera APIs, and they are at a point where Opera has an experimental implementation on Android, and Ericsson has a Webkit version with some support.
The implementation is more tricky. They can fix Firefox on Android of course, but in theory they could pay people to work on Webkit to help out iOS users (and the default Android browser). I'm not sure what Mozilla sees as their vision: is it to create the best web browser on the planet, or is it to make the web platform as good as it could be?
I don't see how they could help Microsoft out with their implementation though.
> Mozilla's mission is to promote openness, innovation and opportunity on the web.
http://blog.chromium.org/2011/08/connecting-web-apps-with-we...
2) This doesn't compete with WebIntents, which is about application selection. WebAPI is about device integration which is compatible with WebIntents. You could, for example, want to make a call and be prompted with several Phone app choices (thanks to WebIntents) and then dial the number and make the call (thank to the Telephony API which is part of WebAPI).
[1]http://www.readwriteweb.com/archives/chrome_and_firefox_work...
On a separate note, should someone be starting up a OMA push proxy for SSE? I have a feeling that between Chrome getting an Android port, the Moz/Chrome WebRTC and Moz's Web APIs, that Server Side Events are going to become "a thing" and OMA push proxy would be a useful service.
I guess I may be thinking about it incorrectly. If this is for web apps to implement a common standard of functionality across Contact "apps", then this makes sense. Otherwise, why are we falling back to "native" when it comes to Contacts, especially when modern contacts exist as part of an online profile - Google Contacts, Facebook Friends and Twitter accounts I follow compose my entire "Contact List".
Also remember that this isn't just part of B2G. This is functionality that will be rolled into Fennec. So when you open your contacts web app in Fennec on Android you should have the same contacts that the native contacts app has.
To put it in perspective, Mozilla has no mobile platform and is crying for relevance.
Apple -> iOS/WebKit Google -> Android/WebKit Mozilla -> ???/Fennec
By getting desktop browsers to implement telephony and other collaboration APIs, I'm sure their hope is to get mobile browsers to likewise expose the precious voice and video APIs of the mobile handsets. And perhaps by doing this, lines between browsers on mobile devices are blurred to where Fennec has a role.
Frankly, your tone is trollish.
Both WebAPI and WebRTC will produce awesome collaboration apps. Wins all around.
But how, exactly, do the two fit together? Is WebRTC a low-level implementation of protocols, codecs, and transports whereas WebAPI provide the friendly developer-oriented APIs?
Will WebAPI provide hooks for standard widgets for dialer, SMS, contacts, etc?
They're not related. WebRTC is a video chat/conferencing project.
> Will WebAPI provide hooks for standard widgets for dialer, SMS, contacts, etc?
Yes, that's its precise purpose.
I'm not seeing any indication that standard widgets are even in their goals. For instance, their Telephony API [1] lists functions that an application implementing its own UI might call, but nothing that indicates that those are UI widgets or expose UI-like functionalities.