> If that can happen, then that is a security flaw of the device and should be fixed, no?
Yes. It should be fixed.
Guess what? Nobody has ever in human history figured out how to make software that is free of security flaws. That’s why these things are only part of the solution.
> I mean, reading a text could install a virus on iPhones last year, yet I didn't see people screaming that all texts should have to go through Apple to make sure they aren't a virus
If you think application software is like a text message then you don’t know enough to comment on this issue.
> If you think application software is like a text message then you don’t know enough to comment on this issue.
If you think that was an example of what I think software is like, then you probably shouldn't comment on it either. I'm just commenting on the fact that, like you said, there's always going to be security issues and using them as the scapegoat for "why" this can't happen is a bad argument.
What's the difference between running an app ad-hoc right now (via XCode) vs allowing a user to download and install any app they want. The answer is, there isn't any except for artificial limitations put in place by apple.
Anything an app that is installed could do would be the same between the store, ad-hoc builds, or installed from a download. The security is in the platform, not the App Store. How many times have we seen something slip through Apple's review process? Fortnite got a full CC flow UI through, many apps back in the day would slip WiFi tethering into their app and get through. All it would take is a simple API call with a boolean to enable or disable the "security issue."
You could always do checks to see what API calls they are using and use that as a metric of "security" but that can also be done device-side. "Hey, this app is using a sketchy API that shouldn't be public, we're going to block it."
> like you said, there's always going to be security issues and using them as the scapegoat for "why" this can't happen is a bad argument.
No it isn’t. If you agree that there are always going to be security issues, then there is no reason why customers shouldn’t be able to choose a product where human moderation is used to mitigate them.
You haven’t made any case as to why customers should be denied this option.
>If you agree that there are always going to be security issues, then there is no reason why customers shouldn’t be able to choose a product where human moderation is used to mitigate them.
First of all, that is one of the worst arguments I've ever heard. A iMessage can install a virus, but you don't see people asking for human moderation of all of their messages on the remote chance that one message contains a virus.
I'll also just copy/paste my reply here since you decided to not reply to it.
What's the difference between running an app ad-hoc right now (via XCode) vs allowing a user to download and install any app they want. The answer is, there isn't any except for artificial limitations put in place by apple.
Anything an app that is installed could do would be the same between the store, ad-hoc builds, or installed from a download. The security is in the platform, not the App Store. How many times have we seen something slip through Apple's review process? Fortnite got a full CC flow UI through, many apps back in the day would slip WiFi tethering into their app and get through. All it would take is a simple API call with a boolean to enable or disable the "security issue."
> You haven’t made any case as to why customers should be denied this option.
Xbox Game Pass, done. Any time Apple bans someone from using a new technology, it harms the consumer and limits user choice.
> A iMessage can install a virus, but you don't see people asking for human moderation of all of their messages on the remote chance that one message contains a virus.
You keep making this silly silly argument.
A text message is not like a software application.
If you really thing the two are comparable, you honestly don’t know anything about computer technology.
You also clearly don’t understand security if you think a platform can be free of security flaws.
As for things slipping through app review. Obviously that happens. That proves that the platform isn’t secure, and that App review isn’t perfect.
If app review wasn’t there, even more things would ‘slip through’. What matters is not how many things slip through, but how many are stopped.
The platform security isn’t perfect, and nor is app review. Together they are much better than one would be alone. It isn’t hard to understand this if you want to.
> A text message is not like a software application.
> If you really thing the two are comparable, you honestly don’t know anything about computer technology.
Ya know, I'm not going to attack you like you seem so set on doing to me. But for reference, I'm a lead SWE working on mobile devices.
As for how a text is comparable. All I'm saying is you claim the platform would be vulnerable with sideload functionality, and what I'm saying is the platform is already vulnerable to the attacks you think this would open the door to. Letting users sideload apps won't make it any more or less vulnerable.
> The platform security isn’t perfect, and nor is app review. Together they are much better than one would be alone. It isn’t hard to understand this if you want to.
I'll say it again, but louder for those in the back. The security is built in to the platform, not app review. If the security was only through the appstore you would see hundreds of viruses in the store a day, controlled by a server side flag. Just like what Uber did to run code differently if the app was being used by reviewers. [1] Why do you think apps need permission to access your sensitive data? That's not the app politely asking while having the ability to get it anyway, that's the _platform_ security.
Now, mind commenting on the part you've ignored twice now? I'll quote it, verbatim, again for you.
"What's the difference between running an app ad-hoc right now (via XCode) vs allowing a user to download and install any app they want. The answer is, there isn't any, except for artificial limitations put in place by apple."
How is the platform so vulnerable and insecure currently? You claim this functionality will cause viruses and adware to run rampant, yet people can do this exact thing right now, with artificial limits (3 max, 7 days max) put in place by apple.
> Ya know, I'm not going to attack you like you seem so set on doing to me. But for reference, I'm a lead SWE working on mobile devices.
Then I assume you understand the difference between text and an executable and they they pose very different security risks.
If so, then we have to assume you are simply arguing in bad faith by pretending they are analogous.
> I'll say it again, but louder for those in the back. The security is built in to the platform, not app review.
This statement is completely false. Security is in both. You surely know that.
> If the security was only through the appstore you would see hundreds of viruses in the store a day, controlled by a server side flag.
No, because they could still be removed after detection.
In any case, this is a dishonest argument. Nobody is claiming that it’s only in the App Store. The claim is that the App Store is a way to catch issues that the platform security alone cannot.
Yes. It should be fixed.
Guess what? Nobody has ever in human history figured out how to make software that is free of security flaws. That’s why these things are only part of the solution.
> I mean, reading a text could install a virus on iPhones last year, yet I didn't see people screaming that all texts should have to go through Apple to make sure they aren't a virus
If you think application software is like a text message then you don’t know enough to comment on this issue.