Day: 23 June 2020

Wesley Lowery, New York Times:

And so, instead of promising our readers that we will never, on any platform, betray a single personal bias — submitting ourselves to a life sentence of public thoughtlessness — a better pledge would be an assurance that we will devote ourselves to accuracy, that we will diligently seek out the perspectives of those with whom we personally may be inclined to disagree and that we will be just as sure to ask hard questions of those with whom we’re inclined to agree.

The best of our profession already does this. But we need to be honest about the gulf that lies between the best and the bulk.

It’s possible to build journalism self-aware enough to bridge that gap. But it will take moral clarity, which will require both editors and reporters to stop doing things like reflexively hiding behind euphemisms that obfuscate the truth, simply because we’ve always done it that way. Deference to precedent is a poor excuse for continuing to make decisions that potentially let powerful bad actors off the hook and harm the public we serve.

Entirely agreed. I would rather read honest reporting than coverage of an event that is purportedly neutral, especially when that apparent neutrality is contrived and built on a phoney belief that there are two equal sides to every story. Honest reporting does not pretend that every argument is valid or needs to be presented.

Alfred Ng, CNet:

On Tuesday, Republican lawmakers introduced the Lawful Access to Encrypted Data Act, which calls for an end to “warrant-proof” encryption that’s disrupted criminal investigations. The bill was proposed by Sen. Lindsey Graham, chairman of the Senate Judiciary committee, along with Sens. Tom Cotton and Marsha Blackburn. If passed, the act would require tech companies to help investigators access encrypted data if that assistance would help carry out a warrant.

Lawmakers and the US Justice Department have long battled with tech companies over encryption, which is used to encode data. The Justice Department argues that encryption prevents investigators from getting necessary evidence from suspects’ devices and has requested that tech giants provide “lawful access.”

[…]

The proposed legislation stops short of requiring tech companies to create a backdoor, noting that the attorney general is prohibited from giving specific steps on how tech companies need to comply with lawful access orders.

It may not require a specific implementation, but eradicating meaningful encryption by introducing vulnerabilities is exactly what this bill mandates:

The debate over encryption and lawful access has raged on, unresolved, for years. The Lawful Access to Encrypted Data Act would bring an end to warrant-proof encryption in devices, platforms, and systems.

Pay little attention to the deliberate use of “warrant-proof” to describe end-to-end encryption. All end-to-end encryption is unable to be accessed by anyone other than the users at each endpoint; that is, almost always, a very good thing.

There is simply no way to do what Senate Republicans are envisioning without some form of back door access. But, as writing that into the bill would likely trigger a First Amendment case should it be voted and signed into law, it instead includes some magical thinking:

Directs the Attorney General to create a prize competition to award participants who create a lawful access solution in an encrypted environment, while maximizing privacy and security.

And I would very much like to acquire a house without expending any money.

There are clearly concerns about what nefarious users of end-to-end encryption are hiding, but requiring everyone to bend to that level means that we all become vulnerable. Making it easier for law enforcement to look into the activities of terrible people makes it easier for terrible people to take advantage of everyone else.

Besides, U.S. intelligence took over a year to discover that their most sensitive and powerful hacking tools had been sent outside its ostensibly secure walls. I don’t trust them with having a key to my phone.

Caitlin McGarry, Gizmodo:

Apple is making a family of custom ARM-based processors for Macs, which is a huge shift that brings with it rippling effects. One change that some might not be too pleased about is that Boot Camp, the free utility that allows you to dual-boot Windows and macOS, is going away on the new Macs powered by Apple silicon.

Rosyna Keller:

You can still disable Secure Boot using the same method as on Intel-based Mac. But it’s entirely up to Microsoft to support booting from something other than the 3 supported Qualcomm SoCs Windows on ARM currently supports.

I don’t know what fraction of Macs today are currently using Boot Camp but, I imagine, anyone who is using it really does rely upon it. It sounds like Apple would like to support Windows on ARM but, if anyone was wondering why Linux was demonstrated during the keynote instead of ARM Windows, it sounds like a technical limitation on the Windows side and not with Apple’s own processors.

While I was writing about Apple’s big ARM transition from the perspective of a technically interested user, the ATP folks recorded over thirty minutes of discussion from the perspective of people with far more knowledge. This is a three hour long episode; buckle up.

The introduction of the iPad in January 2010 will be remembered for lots of reasons. It was the first tablet computer to sell in large numbers and remains a category-defining product. It was the last new category of product to be introduced by Steve Jobs. It was the first expansion of the iOS codebase, which now underpins the Apple TV, Apple Watch, HomePod, and MacBook Pro Touch Bar. And, of course, it debuted Apple’s A-series of ARM processors designed in-house.

Ever since, there has been growing speculation about the viability of a Mac featuring an Apple-designed processor; as those custom processors have far outperformed competing offerings for phones and tablets, and spread across the product line, it has felt inevitable.

Exactly fifteen years and two weeks after Apple’s last processor transition was announced, the Mac is about to make the leap. It will join the rest of the company’s product lineup in running on processors developed in-house, using the ARM instruction set instead of the x86 architecture used by Intel. Just like that — boom — the inevitable became reality.

The way Apple is describing its current motivations for switching, you’d think they could almost copy the press releases from last time but swapping “PowerPC” and “Intel” for “Intel” and “Apple Silicon”, respectively. Everything feels a bit deja vu — efficiency, for example, is a primary motivator. The constraints of performance are primarily determined by energy use as it strongly correlates with battery consumption — relevant in notebooks — and heat output – relevant everywhere. Apple says that its own processors perform more efficiently than the Intel processors it has been using for the past fifteen years just as, during the transition away from PowerPC, Intel processors were described as providing more performance-per-watt.

The timing also recalls the last transition. When Apple moved away from PowerPC, Jobs specifically mentioned the company’s frustration at being unable to deliver a G5 PowerBook as there wasn’t a suitable processor in IBM’s pipeline. Intel, on the other hand, had a full portfolio of processors for everything from notebooks to professional desktops. During this year’s keynote, Johny Srouji spoke about how the company was developing a family of processors for the entire Mac lineup, and Craig Federighi demonstrated great performance from professional apps compiled for Apple’s own processors. What was unspoken but has been plainly obvious for years was how Intel has struggled to deliver across the board. Knowledgeable people say that this is a primary reason why Apple has struggled to update its Mac lineup as regularly as it once did.

Apple gains the ability to update its Macs as frequently as Srouji’s team is able to deliver new processors, but it also assumes the responsibility of being a more vertically integrated single vendor platform. Bold.

I don’t think anyone should be surprised by how closely this transition appears to be modelled on the last one — so much so that many of the technologies that are intended to create a seamless experience for users are billed as sequels to the ones used for the Intel transition: Universal 2 and Rosetta 2. Federighi even made a reference to Jobs’ introduction of Intel processors by introducing the transition after showing new MacOS features and saying that all of the new stuff was demonstrated on a Mac powered by Apple’s own processors.

But the big difference this transition is in the inherent possibilities of custom, product-specific processors with architecture shared by the rest of the product line. By not relying upon a third-party supplier, Apple is free to build into its Mac processors the kind of unique capabilities that have long been afforded to every other product the company makes. I have no dirt on this, but I would not be surprised if the reason no Mac has Face ID yet is because it demands Apple’s own ARM architecture.

MacOS Big Sur running on ARM will also allow iOS and iPadOS apps to run unmodified on a Mac. I do not know what to make of this. Where once there were separate worlds of iOS apps and Mac apps, there are now several app environments that you can run on a Mac: iOS apps, iPad-like Catalyst apps, Catalyst apps modified to be more Mac-like, Cocoa Mac apps — which are, in a sense, the new Classic — and SwiftUI-based apps. Oh, and Electron “apps”. These all feel a little different and, I imagine, iOS apps will sit with Electron and typical Catalyst apps on the ass-end of a scale of Mac-likeness. Whether this is something to look forward to or just something different about ARM Macs is yet to be seen. I think any Mac app that you love today from a developer that you trust will continue to be a true Mac app, so long as that is the best way to deliver a superior product. But whether Apple will continue to deliver terrific first-party apps designed and built specifically for the Mac is a concerning question.

Speaking of ARM, I feel compelled to point out that “ARM” is not mentioned once in the press release announcing this switch. This is something else that is not shocking — it isn’t as though Apple has spent the last fifteen years marketing its products as “x86-based”. That said, I think the use of “Apple silicon” in marketing materials is temporary. I expect the first ARM-based Mac to arrive with a processor branded in true Apple fashion.

Recall the Cook doctrine, via Adam Lashinsky in 2009, with my own emphasis:

We believe that we are on the face of the earth to make great products and that’s not changing. We are constantly focusing on innovating. We believe in the simple not the complex. We believe that we need to own and control the primary technologies behind the products that we make, and participate only in markets where we can make a significant contribution. We believe in saying no to thousands of projects, so that we can really focus on the few that are truly important and meaningful to us. We believe in deep collaboration and cross-pollination of our groups, which allow us to innovate in a way that others cannot. And frankly, we don’t settle for anything less than excellence in every group in the company, and we have the self-honesty to admit when we’re wrong and the courage to change. And I think regardless of who is in what job those values are so embedded in this company that Apple will do extremely well.

You can judge for yourself whether Apple is living up to the simple-versus-complex qualities, or if it is not settling for less than excellence. But any time it has not controlled the technologies that underpin its products is a time when that has eventually bitten them in the ass. Ensuring that Apple’s products have wholly adequate processors tailored for its specific needs makes complete sense.

Does that mean that I am not a little bit anxious? No — of course I have some concern that, when I am next in the market for a Mac, it won’t be quite what I need and some piece of software I love won’t work properly on the new machines. But it isn’t as though that’s an entirely new concern. Also, Apple has set a high bar for both native applications and those running with the help of Rosetta 2. This is, truly, a new era of Mac. No wonder they called the software that will usher it in MacOS 11.

One of the slogans tying together the Tim Cook era is “only Apple” – as in, only Apple could have done this. Well, this truly feels like one of those moments, in more ways than one. Yes, only Apple can convince us that this platform transition will be near-seamless. But this also means that the next generation of Macs will be only Apple products through and through. From the processor to the operating system — and to many of the apps and services users will touch — these Macs are increasingly products defined solely by Apple.

I suppose that is my biggest concern. It is as though Apple now has no escape hatch — no obvious Plan B. When Intel began to struggle, it felt as though AMD could, perhaps, take the reins. That simply isn’t the case any longer. If Apple cannot deliver on its processors for its entire lineup — from the massively popular MacBook Air to the niche Mac Pro, and absolutely everything that lies between — it stops delivering period.

Thankfully, ten solid years of unmatched growth in CPU and GPU performance is a pretty solid C.V. to build the next generation of a business on top of. Apple has proved that it can build powerful and energy-efficient processors in devices as small as earbuds. I am excited to see what it can do with a chassis the size of my iMac, and the kind of vision and capabilities it has shown with its custom processors so far.

In an Apple press release summarizing the developer-specific announcements at WWDC, there are these two significant changes to App Review:

Additionally, two changes are coming to the app review process and will be implemented this summer. First, developers will not only be able to appeal decisions about whether an app violates a given guideline of the App Store Review Guidelines, but will also have a mechanism to challenge the guideline itself. Second, for apps that are already on the App Store, bug fixes will no longer be delayed over guideline violations except for those related to legal issues. Developers will instead be able to address the issue in their next submission.

It is anyone’s guess how likely a rules dispute will result in modifications, but at least the avenue is now there. As for no longer withholding bug fix updates — it’s about time. Unnecessarily punishing users for a developer dispute is unfair and I am glad to see this change.

Tim Hardwick, MacRumors:

Force Touch can be used in watchOS 6 to reveal hidden menus on Apple Watch , such as options to clear notifications and customize the current Watch Face. These options will no longer be accessed using the Force Touch gesture when watchOS 7 is released. Apple’s new Human Interface Guidelines for developers making apps for watchOS 7 confirms the change:

Firm press and long press. In versions of watchOS before watchOS 7, people could press firmly on the display to do things like change the watch face or reveal a hidden menu called a Force Touch menu. In watchOS 7 and later, system apps make previously hidden menu items accessible in a related screen or a settings screen. If you formerly supported a long-press gesture to open a hidden menu, consider relocating the menu items elsewhere.

Several native apps in the watchOS 7 beta already reflect the gesture’s removal. For example, the Force Touch gesture for the app layout Grid/List View has been replaced by a menu option in the Settings app. Similarly, changing the Calendar view must now be done in Settings, while the gesture to Change Move Goal in the Activity app has become just another menu item. The Customize Watch Face menu is now accessed via a long press.

Not a particularly surprising change; even I called it, and I am a dummy. Force Touch and 3D Touch were both interesting experiments in different interaction models, but they were invisible and implemented too inconsistently. As of the first beta of Big Sur, Force Touch trackpad options remain in MacOS, but I suspect not for long.