It is frankly nothing short of amazing that Apple ships things like homomorphic encryption, and differential privacy, and client-side vectorization, and encrypted vectors, at the scale that they inhabit... and they still get a bad report card back from consumers about privacy.
Comparing Apple and Google, or Apple and Microsoft, it seems to me that Apple's track record on these issues is actually not as bad as public opinion might suggest. Meta doesn't even make the list for comparison, and neither does Amazon.
It makes me wonder if picking privacy as a strategy is workable in the first place. People trust TLS; people use banking apps on their phone now without thinking. I remember in 2008 or so when people still didn't quite trust SSL.
I'm not sure if Apple will be able to bridge the gap here, though, if _all_ of their competition simply chooses not to ship those features. Do customers know the difference? Do they... care? In theory they want their data to be private, yes. But if they are not willing to educate themselves to perform their counterparty obligation in fulfilling "informed consent," there may be no action Apple could take to avoid catching the same bad rap everyone else does.
Consumers seem to think well of their pro-privacy standpoints, even if the devil is in the details with regards to how effective it might be.
The people giving “poor reports” are often the hardcore tech users (or LARPers) who grew up in a different era and mindset and are slowly being shoved out.
How did they do it? The idea for homomorphic encryption has been around a long time, but as far as I am aware the time+space requirements have found to be tremendous, thus rendering any serious implementation of it impractible.
If they managed to do it, they should open-source the code. If they made a research breakthrough, they should publish. Doing either of those things would give me the peace of mind that I need.
Fantastic links. They should be at the top of this thread. This stuff looks really impressive to me. I'm not an expert.
I take it that it is implicitly assumed that code running on the server (which I cannot spot the source for) doesn't need to be available, to be trusted? (That there's no way you could come up with an attack through cryptanalysis, involving what the server does, since the security of the data coming from the client is all that matters?)
> It is frankly nothing short of amazing that Apple ships things like homomorphic encryption, and differential privacy, and client-side vectorization, and encrypted vectors, at the scale that they inhabit... and they still get a bad report card back from consumers about privacy.
Personally, I don't shy away from criticizing Google, but that's not the point. Apple makes big claims about their privacy practices that neither Google nor Microsoft make, it would be bizarre to hold Google or Microsoft to claims and standards that Apple set for themselves.
> Comparing Apple and Google, or Apple and Microsoft, it seems to me that Apple's track record on these issues is actually not as bad as public opinion might suggest. Meta doesn't even make the list for comparison, and neither does Amazon.
> It makes me wonder if picking privacy as a strategy is workable in the first place. People trust TLS; people use banking apps on their phone now without thinking. I remember in 2008 or so when people still didn't quite trust SSL.
> I'm not sure if Apple will be able to bridge the gap here, though, if _all_ of their competition simply chooses not to ship those features. Do customers know the difference? Do they... care? In theory they want their data to be private, yes. But if they are not willing to educate themselves to perform their counterparty obligation in fulfilling "informed consent," there may be no action Apple could take to avoid catching the same bad rap everyone else does.
I've said this elsewhere, but what I really dislike about Apple's strategy regarding privacy is that they treat privacy as a purely technological problem that could be solved if only we had better technology, but they ignore that a huge component of why users have been subjected to so many flagrant privacy violations is because they have zero visibility and zero real control over their computing experiences. Apple would very much like to retain their iron grip on what users are allowed to do on their platforms, as they make a ton of money off of their control of the platforms in various ways, so they have a huge incentive to make sure we're all arguing about whether or not Apple is better or worse than Google or Microsoft. Because sure, I do believe that if we held Google to the same privacy standards that Apple currently has, it would probably kill Google. However, if Apple and Google were both forced to give more transparency and control to the users somehow, they'd both be in a lot of trouble.
Despite all of this effort, I think that user trust in the privacy of cloud computing and pushing data out to the internet will only ever go down, because attacks against user privacy and security will only ever continue to get more and more sophisticated as long as there are resourceful people who have a reason to perform said attacks. And there certainly always will be those resourceful people, including in many cases our own governments, unfortunately.
Not impressed. I'd be much more impressed if we could run software like Little Snitch on iOS, or install Firefox. Or even just side load apps without pay $100/year.
(Note: a Safari webview with a Firefox logo on it does not count.)
> Or even just side load apps without pay $100/year
Then use an Android. Apple has taken a stance on this since it was formed. If you don't know by now that Apple is actively hostile toward hobbyists with their approach to computing, I can't help you. That is why there are alternatives from Google, Microsoft, Amazon, and so many others.
To others (me), this is a feature. I would rather the $100/yr fee exist for several reasons.
> You can; or, at least, the APIs are available for this, and have been for some time.
Why is there no citation on this one? iOS has a (fairly limited) VPN API but it is woefully insufficient to make an application-level firewall and I don't think any exist.
That requires an entitlement that A.) is only available in the EU for apps distributed in the EU and B.) as far as I know, has never been granted to anyone. Mozilla for example does not have it.
If it weren't for EU regulations we would probably still not even be able to change the default browser in iOS, so realistically it appears regulating Apple works great. It's hard to argue against regulating Apple while also defending it by showing the fruit bore from regulating them.
> Then use an Android.
I own both iOS and Android devices for what it's worth. Anyway...
> Apple has taken a stance on this since it was formed.
Since... it was formed? Really? I don't remember the Apple II requiring a $100/year developer license to "side load" software.
> If you don't know by now that Apple is actively hostile toward hobbyists with their approach to computing, I can't help you.
This is a frankly insane thing to say to someone who is in middle of criticizing Apple for this exact hostility. Of course framing it as a developer problem is a convenient way to ignore that developers are users and users are developers, and that these restrictions also have negative impacts on even users who aren't developers.
> That is why there are alternatives from Google, Microsoft, Amazon, and so many others.
Apple is the same company that ran a tirade about how Android was a stolen product, so I find it amusing that in just ten years it has become the crutch to defend any and all bad practices Apple holds dear. Don't like it? Just use something else!
Of course, I do use other devices, but I can't magically not be impacted by Apple's persistent brain-death. For years I have had to deal with the utter stupidity that is the fact that Apple refused to support patent-unencumbered video formats like WebM, and devices that can't play WebM natively and in Safari are still plentiful in the wild. So I can't just pretend Apple doesn't exist.
Not that it matters: I am free to criticize Apple however I want, even if Apple fans do not like it. Even just because I feel like it, but even moreso as a paying customer of Apple, and a developer who has dealt with the impact of Apple being involved in the ecosystem. (Thank you Apple, for refusing to support SPIR-V in WebGPU, we love having an additional standard for something for basically no reason.)
> To others (me), this is a feature. I would rather the $100/yr fee exist for several reasons.
I know. I don't think highly of this position, but I am well aware of it.
Then so is Chrome, or, where do you draw the line? WebKit bad, Gecko good? Why? Blink is a fork of WebKit. Ultimately who cares which JIT-enabled browser engine runs your JS?
Clearly none was needed since you know what I was referring to. I disagree that the APIs are not sufficient, since I've used them in enterprise contexts and found them to be comprehensive. The same APIs are available via consumer means. There are tons of VPN and filtering apps for iOS.
From [0]:
- Change the system’s Wi-Fi configuration
- ...
- Create and manage VPN configurations, using the built-in VPN protocols (Personal VPN) or a custom VPN protocol
- Create and manage network relay configurations
- Implement an on-device content filter
- Create and manage system-wide DNS configurations
- ...
And continues with:
- Configure your VPN to include and exclude some network traffic
- "... built-in proxying for TCP and UDP traffic over HTTP/3 and HTTP/2..."
- Use the Network Extension framework to allow or deny network connections
- ...
> That's Safari in a trenchcoat.
Only if you consider the core of Firefox to be Gecko, and not the entire product experience created around Gecko, which is merely an engine.
The security surface area of a JIT-enabled browser engine is significant and complex (see: Chrome). Apple arguably keeps phones safer by maintaining this restriction. Isn't that what you want in the first place?
> That requires an entitlement that A...
See note about Apple's active hostility toward hobbyists. This is considered a feature from their perspective, and reasonable minds could differ about whether they are right, but it is a choice nonetheless.
> I own both iOS and Android devices for what it's worth
So do I. I don't understand why this is mysterious to you, then.
> Since... it was formed? Really?
Yes, really. "Since it was formed" as in, Apple has always taken a stance (after well documented disagreement between the Steves) to build a "walled garden." Whether that wall is a $100/yr fee, or special screws in a tower case, or what not, they have consistently implemented that opinion with action since their inception.
Apple II aside, macOS is actively hostile toward 3rd party software. See: Notary, signing, and so on.
> This is a frankly insane thing to say to someone who is in middle of criticizing Apple for this exact hostility
Being hostile toward hobbyists is a feature to them, not a bug, and it is orthogonal to privacy at best (at worst, in their view, openness is harmful to privacy). If we are arguing about privacy, then we might agree on this point, but for different reasons.
> [Apple is the same company that ran a tirade]... [Don't like it? Just use something else!]
Yes, that is your right. I don't claim to defend everything Apple produces, but, focusing again on the topic at hand (privacy/security), I personally think they do a better job than most. Their choices to get there can be argued over, of course, which is what we are doing now. I see their choices are part-and-parcel of a larger cohesive strategy; apparently you do not?
> So I can't just pretend Apple doesn't exist
Sure, I never suggested you could, merely that other devices will be naturally better for people who want to use them as enthusiasts or hobbyists outside the Blessed Apple Path(tm).
> I know. I don't think highly of this position, but I am well aware of it.
¯\_(ツ)_/¯ I think highly of your position to argue with it in good faith, so I'm sad to hear that.
So Apple kneecaps Meta's cross-app advertising, something that literally makes them no direct revenue to implement, and protects users (it famously reduced Facebook cross-app analytics traffic to a significant degree), and you think this is business as usual?
Then you should reconsider my comment at the top of this thread, because it is 100% speaking to this exact phenomenon.
Ask App Not To Track was literally just the ability for the user to choose for themselves whether tracking is allowed. That’s a user agency improvement.
Comparing Apple and Google, or Apple and Microsoft, it seems to me that Apple's track record on these issues is actually not as bad as public opinion might suggest. Meta doesn't even make the list for comparison, and neither does Amazon.
It makes me wonder if picking privacy as a strategy is workable in the first place. People trust TLS; people use banking apps on their phone now without thinking. I remember in 2008 or so when people still didn't quite trust SSL.
I'm not sure if Apple will be able to bridge the gap here, though, if _all_ of their competition simply chooses not to ship those features. Do customers know the difference? Do they... care? In theory they want their data to be private, yes. But if they are not willing to educate themselves to perform their counterparty obligation in fulfilling "informed consent," there may be no action Apple could take to avoid catching the same bad rap everyone else does.