FB, G clearly violated the ToS of the Enterprise Acc (no public distribution of such Apps) and Apple has rightly suspended their enterprise distribution cert.
But Apple has been selective in enforcing this rule. If I recall, for many years Uber's driver App was distributed as an enterprise app. Uber has always claimed that drivers are not employees and so this was in clear violation of the ToS.
Imo, Uber's use case was legit. During early days Uber probably did not wish to have 2 Apps in the App Store to avoid customer confusion. Or maybe they were actively updating the Driver App and did not want to add days of App review holding up every update.
Apple should change their ToS and allow such use cases in some form. At times this would get misused (like FB/G) but opening up the walled garden to enable such "private" apps to be easily distributed can make iOS a more interesting platform. In any case they always have the final kill switch of revoking an Enterprise Cert for malicious use.
Australian banks have a good thing going down under: they are the most profitable in the world in terms of profits as % of GDP. They may not make as much in payments but seem to have plenty of money making options.
Yeah, but they are also tightly regulated by APRA and when the GFC was happening and everyone else was worried about their banking system ours was stronger than ever.
Its not comparable scale. WhatsApp are shifting messages from point A to point B and that's it. Even 100x less servers is still a LOT of servers just or that. They don't store data, just traffic it from one user to another.
There is rightly a consensus that iOS Apps are generally better than Android Apps.
IMO this is mainly because of the iOS SDK/Xcode and not due to the App Store review process.
The review system is geared for consumer focused apps. Business Apps need specialized domain knowledge and complex setups to review. That is why there are so many consulting firms doing this for clients.
The review process currently just drains some resources but may really start hurting the iOS ecosystem when business Apps become the driver for mobile platform choice. It already is in the developing world because for many, the phone is their only computer.
A user may really want to use iOS, but if my key problem is solved by an Android only App - I will need to use Android.
This is one of the reasons why we ended up with Windows dominating.
iOS now has a robust sandboxing and an excellent permission system - so the risk of an App doing harm is low. All apps without in-app purchases should be automatically approved. Approvals can always be revoked later if issues surface.
In my experience this is less and less the case. I think it's largely a function of developer effort. In fact, at this point I consider a number of apps (WhatsApp, Wechat, Zalo, Viber, Twitter, all of Google's apps) to be more usable in their Android incarnations than they are on iOS. This is partly to do with the greater degree of platform integration Android allows and, I suspect, the fact that it's just harder to introduce crash bugs in Java than Obj-C. Text entry on Android is currently vastly superior thanks to Google's very good stock keyboard and the plethora of excellent third party keyboards.
iOS is clearly ahead when it comes to anything having to do with multimedia but for a very large class of apps these days it's either a tossup or a win to Android. iOS 8 will probably close the gap to some degree because it allows devs to hook into the OS to a degree that was never possible before but I also expect Android apps to improve now that people can afford to ignore 2.x and the baseline for cheap phones has been raised by the Android One program.
> it's just harder to introduce crash bugs in Java than Obj-C
I am not convinced. ObjC pointers are a lot safer than C pointers under ARC, and sending messages to nil won't even crash. It's almost more likely to have bloody AutoLayout crash your app :)
ARC is still awfully fiddly. Between all the bridging casts that are often necessary and the interaction with strong/weak/etc. modifiers you can put on properties, it's really easy to crash or leak memory with ARC. It's an improvement I guess, but I'm not sure it's much of one, because it lulls people into thinking they don't need to understand the underlying reference counting.
I agree that it's terrible in theory, but it I've rarely had a memory bug slip into the App Store. Would love to see statistics from HockeyApp or others on reasons for app to crash.
But Apple has been selective in enforcing this rule. If I recall, for many years Uber's driver App was distributed as an enterprise app. Uber has always claimed that drivers are not employees and so this was in clear violation of the ToS.
Imo, Uber's use case was legit. During early days Uber probably did not wish to have 2 Apps in the App Store to avoid customer confusion. Or maybe they were actively updating the Driver App and did not want to add days of App review holding up every update.
Apple should change their ToS and allow such use cases in some form. At times this would get misused (like FB/G) but opening up the walled garden to enable such "private" apps to be easily distributed can make iOS a more interesting platform. In any case they always have the final kill switch of revoking an Enterprise Cert for malicious use.