4 ms·

That argument doesn't work because the Mac App Store is not a popular way to buy software. Most of the apps people still buy Macs for are not in the Mac App Sto

That argument doesn't work because the Mac App Store is not a popular way to buy software. Most of the apps people still buy Macs for are not in the Mac App Store (source: none of the most popular creative apps are in the Mac App Store https://blog.robenkleene.com/2019/08/07/apples-app-stores-have-failed-creative-apps/ https://blog.robenkleene.com/2019/08/07/apples-app-stores-ha...), so it didn't help security (https://www.theverge.com/2019/7/8/20687014/zoom-security-flaw-video-conference-websites-hijack-mac-cameras. https://www.theverge.com/2019/7/8/20687014/zoom-security-fla...). What it did do is harm the Cocoa apps that relied on Apple's stack. Look I understand the prism that you're looking at this through, and I do believe that was the security team at Apple's intention, but I don't think that's the reality of how it has played out. UPDATE: I think the fatal flaw of Apple's security teams approach is that it only would have worked if there were some way to pull mobile use cases back to the Mac, because those use cases are compatible with sandboxing. But that just hasn't happened, the use cases that moved to mobile, e.g., social media, chat, video consumption, aren't coming back to the Mac ever. And the use cases that did stay on the Mac: audio/video editing, 3D, programming are just simply demonstrably not compatible with sandboxing.

zepto 7y ago
The Mac App Store not being popular doesn’t invalidate the argument at all, since Apple had reason to believe that it would be when they made the investment. It’s not clear what you mean with your update - what would it even mean to pull social media apps back to the Mac, and how would that help the situation with tools for thought? It’s not at all clear to me that the use cases that did stay on the Mac are demonstrably not compatible with sandboxing. Can you explain that? Certainly in-process plug-ins aren’t compatible, but those are utterly insecure and Apple continues to improve out of process extension mechanisms.
- robenkleene 7y ago
  Generally agree with your opening statement about the Mac App Store, I cover the incompatibilities with sandboxing here: https://blog.robenkleene.com/2019/08/07/apples-app-stores-have-failed-creative-apps/ https://blog.robenkleene.com/2019/08/07/apples-app-stores-ha... But yeah plugins are a big one. I don't know how you'd pull mobile use cases back to the Mac, that's Apple's strategy, not mine. My overarching point is that Apple seems to be trying to trying to fix what they perceive as macOS's flaws, e.g., security relative to iOS, instead of improving it's strengths, e.g., that it can run categories of apps that haven't had much success on mobile, like the major creative apps. If the users of these apps are given a choice: Use a less effective versions that works in Apple's ecosystem, or switch platforms, we already know what is going to happen, they'll switch platforms, that's what's happened with the markets Apple has already lost like 3D and special effects. UPDATE: Here's the question I have, how does improving the Mac for mobile use cases (e.g., the ones that work well with the new security model, I think we can take this as a given since it's a security model that came from iOS) help Apple sell Mac's if the main reason people buy Macs is to use software that doesn't work with this security model? And if you're still not convinced about the security model being incompatible with the major creative apps, then I'd ask: Why haven't any major creative apps been successful on iOS? Because I attribute that to the security model as well.
  - zepto 7y ago
    I don’t see how it’s Apple’s strategy to pull mobile use cases back to the Mac - that seems to be something you are inferring but I don’t see anything that demonstrates it. Your observations about people switching platforms seem accurate. However the security strategy Apple is adopting seems to me to be nothing to do with mobile, and everything to do with a globally connected world where personal data is valuable and subject to attack. The fact that it originated on iOS has nothing to do with it being mobile, and much more to do with it being a clean slate where they didn’t have to deal with legacy insecure apps. It seems unclear to me why you’d think that major creative apps haven’t been successful on iOS because of the security model when the processing power, storage, input method, and screen size all seem like they have much more influence over this and are only just now becoming viable for these apps. I also note that you are increasing the conflation of ‘major creative apps’ with ‘Tools for thought’.
    - robenkleene 7y ago
      > I don’t see how it’s Apple’s strategy to pull mobile use cases back to the Mac - that seems to be something you are inferring but I don’t see anything that demonstrates it. SwiftUI and Catalyst, Apple's biggest developer technology pushes since Swift, both are explicitly for this purpose. Not to mention Apple themselves porting a bunch of their own apps from iOS to Mac, and then all the previous technologies that have gone from iOS to Mac. And Apple Arcade being mainly about mobile games on the Mac. > However the security strategy Apple is adopting seems to me to be nothing to do with mobile, and everything to do with a globally connected world where personal data is valuable and subject to attack. The apps that people use Macs for listed here like Adobe CS, Sketch, Ableton, etc... don't harvest data the same way mobile apps do. That's not their business model. The security model is about mobile apps like Facebook/Instagram/WhatsApp, etc... whose business model is about harvesting data. But nobody runs apps like that on desktop, therefore the security model is less relevant for the platform. And again, the reason Apple is pushing it is because they are just blanket applying their mobile app strategy to the desktop even though it doesn't make sense there. > I also note that you are increasing the conflation of ‘major creative apps’ with ‘Tools for thought’. Yes, I'm explicitly talking about the major creative apps in this conversation. If you're talking about apps like MindNode, then I agree that an app like that is being served pretty well by Apple's technology stack today, but then there's not much to talk about there :) The point about "processing power, storage, input method, and screen size" is a valid counter hypothesis, personally I disagree, and I think the evidence points to the security model having a devastating impact on these types of apps by how sandboxing impacted the creative app market on macOS. But I can certainly understand a different perspective here.

zepto 7y ago

The Mac App Store not being popular doesn’t invalidate the argument at all, since Apple had reason to believe that it would be when they made the investment. It’s not clear what you mean with your update - what would it even mean to pull social media apps back to the Mac, and how would that help the situation with tools for thought? It’s not at all clear to me that the use cases that did stay on the Mac are demonstrably not compatible with sandboxing. Can you explain that? Certainly in-process plug-ins aren’t compatible, but those are utterly insecure and Apple continues to improve out of process extension mechanisms.

robenkleene 7y ago
Generally agree with your opening statement about the Mac App Store, I cover the incompatibilities with sandboxing here: https://blog.robenkleene.com/2019/08/07/apples-app-stores-have-failed-creative-apps/ https://blog.robenkleene.com/2019/08/07/apples-app-stores-ha... But yeah plugins are a big one. I don't know how you'd pull mobile use cases back to the Mac, that's Apple's strategy, not mine. My overarching point is that Apple seems to be trying to trying to fix what they perceive as macOS's flaws, e.g., security relative to iOS, instead of improving it's strengths, e.g., that it can run categories of apps that haven't had much success on mobile, like the major creative apps. If the users of these apps are given a choice: Use a less effective versions that works in Apple's ecosystem, or switch platforms, we already know what is going to happen, they'll switch platforms, that's what's happened with the markets Apple has already lost like 3D and special effects. UPDATE: Here's the question I have, how does improving the Mac for mobile use cases (e.g., the ones that work well with the new security model, I think we can take this as a given since it's a security model that came from iOS) help Apple sell Mac's if the main reason people buy Macs is to use software that doesn't work with this security model? And if you're still not convinced about the security model being incompatible with the major creative apps, then I'd ask: Why haven't any major creative apps been successful on iOS? Because I attribute that to the security model as well.
- zepto 7y ago
  I don’t see how it’s Apple’s strategy to pull mobile use cases back to the Mac - that seems to be something you are inferring but I don’t see anything that demonstrates it. Your observations about people switching platforms seem accurate. However the security strategy Apple is adopting seems to me to be nothing to do with mobile, and everything to do with a globally connected world where personal data is valuable and subject to attack. The fact that it originated on iOS has nothing to do with it being mobile, and much more to do with it being a clean slate where they didn’t have to deal with legacy insecure apps. It seems unclear to me why you’d think that major creative apps haven’t been successful on iOS because of the security model when the processing power, storage, input method, and screen size all seem like they have much more influence over this and are only just now becoming viable for these apps. I also note that you are increasing the conflation of ‘major creative apps’ with ‘Tools for thought’.
  - robenkleene 7y ago
    > I don’t see how it’s Apple’s strategy to pull mobile use cases back to the Mac - that seems to be something you are inferring but I don’t see anything that demonstrates it. SwiftUI and Catalyst, Apple's biggest developer technology pushes since Swift, both are explicitly for this purpose. Not to mention Apple themselves porting a bunch of their own apps from iOS to Mac, and then all the previous technologies that have gone from iOS to Mac. And Apple Arcade being mainly about mobile games on the Mac. > However the security strategy Apple is adopting seems to me to be nothing to do with mobile, and everything to do with a globally connected world where personal data is valuable and subject to attack. The apps that people use Macs for listed here like Adobe CS, Sketch, Ableton, etc... don't harvest data the same way mobile apps do. That's not their business model. The security model is about mobile apps like Facebook/Instagram/WhatsApp, etc... whose business model is about harvesting data. But nobody runs apps like that on desktop, therefore the security model is less relevant for the platform. And again, the reason Apple is pushing it is because they are just blanket applying their mobile app strategy to the desktop even though it doesn't make sense there. > I also note that you are increasing the conflation of ‘major creative apps’ with ‘Tools for thought’. Yes, I'm explicitly talking about the major creative apps in this conversation. If you're talking about apps like MindNode, then I agree that an app like that is being served pretty well by Apple's technology stack today, but then there's not much to talk about there :) The point about "processing power, storage, input method, and screen size" is a valid counter hypothesis, personally I disagree, and I think the evidence points to the security model having a devastating impact on these types of apps by how sandboxing impacted the creative app market on macOS. But I can certainly understand a different perspective here.