If Apple doesn't require the developer to provide source code what stops a nefarious developer from say putting a fake 'log in with google/facebook/apple' option that collects the username and password and then returns a typo error AND a real login option? No one would ever know that the first time the were presented with a login that it was fake, and if the developer stored the information and then sent it hours to weeks later no one would ever be able to tell that it was harvesting data.
What stops it? What you suggested would be plain criminal, so someone would involve the police about this, and then someone would go to jail. Apple can't stop every criminal. Apple most certainly cannot verify the source code for every single app in the store. For my app, I'd say a week of time by an expert is the absolute minimum. Multiply that by a million apps.
In the end, this warning that Apple is giving now in a bet version (and iOS 14 is the only OS in the world doing it), is causing so much trouble to so many people, there's a good chance that it won't be there in a released version.
[automerge]1594629039[/automerge]
So intentionally alter the password on the first login attempt so it fails but capture, encrypt, sit on, and package the data in a user requested action.
There is no way Apple could know they typed the password in right, nor would they have access to external systems to know if the correct data was being sent.
I bet Apple has some Facebook accounts for that purpose, without anything of any value in it, and the password is something like 123456. If the reviewer notices the Facebook login fails, then they know something is up.
[automerge]1594629113[/automerge]
Thank you Apple for exposing allowing this. There I fixed it for you.
If Apple is truly the champion of privacy they'd proactively prevent it from happening rather than just reactively informing you.
Mi7chy, you are getting absolutely daft here. _Every single operating system_ currently used, Windows, Linux, MacOS, Android, iOS up to 13 allows this and has always allowed this.
[automerge]1594629304[/automerge]
It seems like Apple should make an API or whatever that can tell an app what type of content is stored in the clipboard (text, image, etc.) without giving it the content. That way the app knows if it makes sense to offer a paste option without having access to the content.
Yes, that would be useful. There are apps that detect URLs in the clipboard and use them. There are apps that accept plain text but in a very specific format. Yes, it would reduce the number of warnings. But the immediate effect is that functonality will be removed from apps.
[automerge]1594629432[/automerge]
Not sure how something that can basically be a bug can really be litigated. Even that aside, seems like actual ill intent would need to be demonstrated and actual damages.
It's not actually a bug. It is behaviour that is perfectly legal in Windows, MacOS, Linux, Android, and iOS up to 13.0.