With Mozilla and Canonical getting involved in the phone space things are starting to get interesting again. I'm glad to see more alternatives to the Android/iPhone paradigm that we have been living in for the past few years. If we could get more devices that are OS agnostic and allow users to choose what they would like on purchase would go a long way to expanding the mobile space for the better.
Anything that breaks the programming language monoculture per platform is welcome in my book. That includes things like Mono and platforms like Canonical.
I'm wondering; both android and ios restrict apps from using the javascript vms (v8 and nitro respectively) because of security concerns (being able to generate code and mark their pages executable). The embedded webkit implementations use JSKit interpreter as far as I know. (Correct me if I'm wong)
How does FF OS handle this? I believe if there is absolutely no native code execution involved, it should not be a security concern. Is that so? Is it currently possible to have fast js execution in UIWebView iOS apps?
I'd argue it's to keep developers locked in to Objective-C and iOS. Facebook tried going the HTML5 route in the native iOS app, but everything was too slow (it was comical how much faster facebook.com was in Safari vs running the app) until they ultimately switched back to Objective-C.
It's ironic that Objective-C was an albatross on the neck of Mac OS X, a platform with small marketshare and an uncommon application framework. Now, Objective-C is a key advantage for Apple's vendor lock-in keeping iOS developers from (easily) porting to Android.
I am probably the only person here but I do not think it has much to do with keeping alternative browsers off the platform or preventing people from using UIWebViews.
They already have a clause for third party browsers in the submission/review guidelines so there is no need for the existence of a technical reason to enforce that.
I truly think the real reason is security. The problem with allowing rwx pages in an iOS process via WebKit is that they can be used by the native app to load rogue code in.
This is a really difficult problem to solve and I honestly think Apple chose the easy way out: disallow it all.
I do think a solution is coming though. Already in iOS6 a bunch of ViewControllers are running out of process. See the link below. I think this will lead to out of process, and fully JITted WebKit in a future release.
Many times people think Apple is making some weird technical decision on purpose to annoy developers it is really just a pragmatic decision to get a product out of the door on time. This happens all the time and everywhere. At Google, Mozilla, Apple, Blackberry and Canonical.
Firefox OS does use Spidermonkey's JIT compiler to run apps, but it runs each web app in a separate low-privilege process, to mitigate damage that a compromised app might do:
Security is not only dereferencing an invalid pointer or executing data that is not meant to be. IOS way of doing things is just a choice out of many and have nothing to do with other OSes. It is all about offering and enforcing a highly controlled and predictable development environment.
Nothing prevents exploits that utilize existing web platform which is the basis of FFOS exclusively.
Having said that, main goal appears to be not bringing web to the mobile, but freeing mobile into the web so eventually browsers can have a de facto standards for hardware capabilites. Yet another mobile sandbox is not a goal, as I see it.
FFOS cannot be compared to any other OS for mobile until now, in terms of goals and politics.
Given that it's proven to be incredibly difficult to make any money as an Android manufacturer I'm not surprised to see Sony taking a different tack. Question is whether it was a careful considered move or a desperate bid for something new.
Yes, this is a point often overlooked when people say that Nokia going with Windows Phone was a huge mistake. The Android market really isn't that great a place to be in, and there's no guarantee that Nokia would have made an impact there.
I'm not saying that WP was definitely the right choice, but it's not a simple calculation, especially when you factor in the money MS is giving Nokia.
Android wasn't the only choice, though. I still wonder how would they have fared with Meego, especially if they had finished the Android compatibility layer (to ease porting).
A good example is Amazon's Kindle Fire tablets. Amazon has a curated Android experience (and even avoids using the term "Android"). Of course, Amazon has lots of media content to pair with their Kindle Fire ecosystem that Nokia would not have.
I think Nokia could have built a good-enough content ecosystem. Nokia had a decent music service in many countries. I've even bought from them (for a song that wasn't available in the US stores I normally use). Finding a partner for books wouldn't have been hard. Kobo would probably have jumped at it, among others. Video would have been hardest, but Google was also struggling with video back then so it wouldn't have been a relative disadvantage.
I think it would've been. At the time Samsung wasn't that popular. They were still "Nokia" the largest smartphone maker. Switching to Android would've meant Nokia would've gotten to keep its position, and be #1 now instead of Samsung. Instead they went with an OS that still has only 2% market share after more than 2 years, and Nokia is like the 10th largest smartphone maker or something now.
Every time people imply that Nokia's choice was binary, I want to scream. It wasn't. Nokia was a big company with a solid cash stockpile. They had the resources to try more than one thing. That they didn't is the real management failure.
That's a good point about Nokia's choices not being binary, though part of their earlier problems were related to Nokia exploring too many options: Symbian, S40, Maemo, and Meego.
That's a pretty bold fact. Samsung apparently generated $60 billion dollars in revenue from Android devices last year, about a third of their total revenue.
Well, my point was that it's incredibly difficult to be an Android manufacturer that isn't Samsung, given they utterly dominate the market. It's very crowded.
Strangely, people have been saying what your comment implied, that no-one could make any money on Android devices, for quite a while. In fact I'd go so far as to say it was considered uncontroversial by many of the people saying it.
And now when faced with the elephant in the room that is Samsung, they are forced to retreat to, well only Samsung can make profits, but no-one else can. But if they couldn't predict Samsung, why would we listen to what they've got to say about the future now?
(Just out of curiousity, did you correctly predict that Samsung would make so much money from Android?)
You could say that it's incredibly difficult to be outside of the top 5 smartphone manufacturers. But isn't that true of the laptop market or the server market as well?
