Hacker News Reader

Comment by mensetmanusman | Hacker News Reader

If iPhone adoption approaches 90% in the US it would be prudent to force Apple to allow other software to run on the device if the user so desires.

Otherwise we have systemic risk associated with a digital monoculture.

E.g. Imagine how fast a zero day zero click iOS worm could brick every device and cause a huge disruption to daily life.

View on HN

I'm actually a bit more worried about the reverse: software that only works on iPhones.

This is already a problem to some degree with software that only works on iPhone and Android; today I got a PCR test and the hospital apparently only sends the results over WhatsApp. I don't have WhatsApp. They looked at me as if I'm crazy when I asked if they could just email me the results; it was all a bit difficult (although they said they would; we'll see what happens, otherwise I'll just pick it up tomorrow morning before flight I guess).

If you want to use iPhones or WhatsApp or whatever: great! Your device, your life, your choice, I'm not bothered by it. But I don't like it when I'm more or less forced to use the same things, even if I don't want to. It really reminds me of the "you need Windows"-days.

The article says "the device held 90% of smartphone activations in the week after Christmas 2020", but this does not mean that the iPhone has 90% market share.

Apple's US market share is currently 55% of the smartphone market, down from 65% a year ago.

Really? That’s almost too huge to be true. A 10% decrease in one year? Do you have a source on this or an explanation? A cursory Google gives me a current 43% US market share figure.

don't forget market growth.

55%? The last I heard (admittedly a few years ago) Apple was less than 20%. Though, maybe that was worldwide.

90% market share among young people is concerning though. If they never switch that means overall US market share will approach 90% over time.

I'd wager iMessage and Snapchat are big reasons why iPhone is huge with young Americans. Snapchat's Android version is considered to be significantly worse than Apple's. Not sure if that is still the case now, but bad first impressions persist for a long time.

In the rest of the world, Whatsapp and FB Messenger appear to be the dominant messaging apps (outside China of course).

>it would be prudent to force Apple to allow other software to run on the device if the user so desires.

>[...]

>E.g. Imagine how fast a zero day zero click iOS worm could brick every device and cause a huge disruption to daily life.

how would a 0click iOS worm be mitigated by allowing users to run whatever apps they want? You can run whatever apps on windows, but that didn't stop eternalblue from ravaging corporate networks.

edit: it seems like I misinterpreted what "other software" meant. my reply to swiley takes this into account.

They wouldn't be, just like they aren't on Linux, Windows 7, and older OSX.

So not whatever apps you want, but rather whatever operating systems you want? That makes more sense, but I'm skeptical that'd materialize to everything. On android phones with unlocked bootloaders (ie. "allow other software to run on the device if the user so desires"), and the only "other software" that's available are... AOSP forks. There are forks with better mitigations (eg. grapheneos), but they're still by and large, android.

This is mostly because making a different OS is insanely hard, at least if you want to support the many different pieces of hardware your OS might run on. Project Sandcastle is different and making progress[0], likely because Apple mobile hardware uses vastly similar components and suppliers.

0: https://projectsandcastle.org/status#:~:text=WHAT%E2%80%99S%...

Unfortunately, Project Sandcastle hasn’t had any commits in at least two years[0].

0: https://github.com/corellium/projectsandcastle/commit/03db9c... (the latest commit)

I think the "other software" in question here must refer to alternative operating systems, not merely other apps. Though even allowing alternative browsers with their own engines could possibly reduce the monoculture risks associated with that portion of the attack surface.

By running a different OS or firmware on the device allows for a smaller attack surface

While that would change the status quo of "iPhone == iOS", running a different OS is not what most people would do, so it really wouldn't move the needle on how much harm is possible with a potential widely-deployed 0-click exploit.

People generally buy iPhone for the no-nonsense (or low-nonsense) UX provided by iOS compared to Android - as in, they want to purchase iOS, but obviously it's only compatible and available on certain hardware. An extremely, extremely small majority would install an alternative OS on their phone unless that OS is something like "iOS but with built-in hacks for mobile games", or "iOS but paid app store apps can be sideloaded for free".

Who are you imagining would desire, apart from you?

Who would pay more for an iPhone they didn't want, gambling on future regulation forcing Apple to let them run Android [LineageOS, etc] at some point, when they could have bought a cheaper phone running what they wanted in the first place?

I want an iPhone because I like the OS and the integration with my iPad. This doesn’t mean that I want to be prevented from installing a 3rd party YouTube app.

This same fallacy gets used for political parties as well. Just because someone picked an option, doesn’t mean that they approve of every single thing that comes with it. Only that they found overall that the package was preferable.

That's different, we weren't talking about agreeing with everything or third party apps, this was about third party operating systems. The comparison would be like voting Party1 while hoping someone would legislate that they must behave like Party2, when you could have voted Party2 in the first place.

You like the OS and its integration, so it wouldn't be you buying it in the hope that you could replace the OS, right?

Reply on news.ycombinator.com