If you have been an Apple user long enough, then you probably lived through a few painful transitions. I remember the 68k to PPC to Intel hardware years. I was also around during the even more painful move from System 9 to OS X. Now, Apple plans to migrate its computers from Intel to ARM while upgrading macOS at the same time. If history repeats itself, we could be in for a few rough years as Apple makes the switch. Fortunately, Apple has an ace up their sleeve; their bread and butter devices have all been running on ARM for years.

如果您已经足够长时间成为Apple用户，那么您可能经历了一些痛苦的过渡。 我记得从PPC到Intel硬件的68k年。 从System 9迁移到OS X时，我也处于痛苦的境地。现在，Apple计划将其计算机从Intel迁移到ARM，同时升级macOS。 如果历史重演，随着苹果公司的改变，我们可能会经历几年的艰难时期。 幸运的是，苹果的袖手旁观。 他们的面包和黄油设备都已经在ARM上运行了多年。

When I joined Microsoft in 2012 to help with the launch of Windows Phone 8 and Windows 8, I received a non-descript prototype tablet. It ran on a custom OS called Windows RT. This was months before Microsoft would announce their intention to run Windows on ARM-based devices. The Surface RT was supposed to be the future of Windows computing. On paper, it sounded like a solid plan because even eight years ago, you could see the computer industry moving towards more power-efficient, custom ARM chips. In reality, however, when I used Windows RT for the first time, it was clear that this plan was not going to work.

当我于2012年加入Microsoft协助启动Windows Phone 8和Windows 8时，我收到了一个非描述性的原型平板电脑。 它在名为Windows RT的自定义操作系统上运行。 距离微软宣布打算在基于ARM的设备上运行Windows还要几个月。 Surface RT被认为是Windows计算的未来。 从表面上看，这听起来像是一个坚实的计划，因为甚至在八年前，您就可以看到计算机行业正在朝着更加节能，定制的ARM芯片发展。 但是，实际上，当我第一次使用Windows RT时，很明显该计划无法正常工作。

Microsoft hired me to convince developers to port their games to Windows 8’s various hardware architectures: Intel/AMD, ARM, and Mobile. Windows dominated the gaming space so this sounded like a slam dunk. How hard could it be? Well, if you’ve never been on the early side of a “clean-break” software transition, I’ll paint this picture for you.

微软聘请我说服开发人员将其游戏移植到Windows 8的各种硬件体系结构：Intel / AMD，ARM和Mobile。 Windows主导了游戏空间，因此听起来像是灌篮。 它能有多难？ 好吧，如果您从未涉足过“干净的”软件过渡的初期阶段，那么我将为您提供帮助。

Imagine you are sitting in front of a computer with a clean install of an OS that looks the same as what you’ve been using for a decade. The only things installed on it are a few bare-bone Apps for text editing and web browsing. You open the Web browser to start downloading the Apps you use daily. Then you go to install them only to find out that nothing works. The average customer doesn’t understand why their (insert any App they currently use) doesn’t work.

想象一下，您坐在一台全新安装的OS的计算机前，该OS看起来与您十年来使用的操作系统相同。 唯一安装的是一些用于文本编辑和网络浏览的准骨干应用程序。 您打开Web浏览器开始下载日常使用的应用程序。 然后，您去安装它们只是发现没有任何作用。 普通客户不理解为什么他们的(插入他们当前使用的任何应用程序)不起作用。

As you already know, Windows RT didn’t take off. Windows has legacy code going back to the early days of computers, and cutting all of that out invalidates the main selling feature of the OS, backward compatibility. I could write an entire book on how Microsoft’s plans for Windows on ARM was a disaster. Even years later, after multiple hardware and OS revisions, it continues to be a failure. Long story short, it’s all about software.

如您所知，Windows RT并没有起飞。 Windows具有可追溯到计算机早期的遗留代码，并且全部删除这些代码会使操作系统的主要销售功能，向后兼容性失效。 我可以写一本书，讲述微软在ARM上使用Windows的计划是多么的灾难。 甚至数年后，在多次修改硬件和操作系统之后，它仍然是失败的。 长话短说，这全都与软件有关。

Right now, modern computer operating systems are all but dependant on Intel/AMD chips. For a developer, any migration to ARM means at the bare minimum they’ll need to remove any legacy code, refactor what’s left to work on new APIs, do a lot of testing, recompile, and distributing it. Apple would like you to believe this will only take changing a few lines of code. I know, because that is what Microsoft had me tell the developers I worked with. The reality is that it takes more work than anyone could imagine, primarily when the tools developers use don’t support the new architecture well before launch.

[R现在洞察力;现代计算机操作系统是英特尔/ AMD芯片全部依赖却。 对于开发人员而言，任何向ARM的迁移都意味着他们至少需要删除所有旧代码，重构可用于新API的内容，进行大量测试，重新编译和分发。 苹果公司希望您相信，只需更改几行代码即可。 我知道，因为那是Microsoft告诉我与之合作的开发人员的。 现实情况是，这需要比任何人想象的都要多的工作，主要是因为开发人员使用的工具在发布之前就无法很好地支持新架构。

During the switch to OS X, I completely abandoned Apple computers.

在切换到OS X的过程中，我完全放弃了苹果计算机。

Depending on how long it takes these tools to adapt, it could take developers months before they can release a new build. For example, look at Adobe. It has years of legacy code and technical debt in most of their staple software like PhotoShop, Lightroom, After Effects, and Premiere. While Adobe always announces their support for Apple’s new hardware migrations, the professionals that depend on their software can tell you that it never goes smoothly. During the switch to OS X, I completely abandoned Apple computers. I had no choice but to use Windows because the first two generations of software Adobe release on OS X were almost unusable.

根据这些工具适应的时间长短，开发人员可能需要几个月的时间才能发布新版本。 例如，看一下Adobe。 它在大多数常用软件(如PhotoShop，Lightroom，After Effects和Premiere)中都有多年的遗留代码和技术债务。 尽管Adobe始终宣布支持Apple的新硬件迁移，但是依赖于其软件的专业人士可以告诉您，它永远无法顺利进行。 在切换到OS X的过程中，我完全放弃了苹果计算机。 我别无选择，只能使用Windows，因为在OS X上发布的前两代软件Adobe几乎无法使用。

Luckily for Apple, they have been developing iOS on ARM in parallel for over a decade now. The latest release of OS X is clearly a stopgap to prepare desktop developers for switching to ARM. Catalina no longer supports 32-bit Apps, has stricter security around installing 3rd party Apps outside of Apple’s App Store, and offers iOS developers a way to migrate iPad Apps over to macOS if they share a similar codebase. The writing for the switch to ARM was on the wall, but I’m not 100% convinced that even with millions of iOS mobile and tablet Apps, this transition will be easy for users. It will be a lot worse for developers, for sure.

对于Apple来说幸运的是，他们已经在ARM上并行开发iOS已有十多年了。 OS X的最新版本显然是为桌面开发人员准备转而使用ARM的一个权宜之计。 Catalina不再支持32位应用程序，在Apple App Store外部安装第3方应用程序方面具有更严格的安全性，并且如果iOS开发人员共享相似的代码库，则可为iOS开发人员提供将iPad Apps迁移到macOS的方法。 切换到ARM的写作几乎没有，但我不是100％相信即使有数百万个iOS移动和平板电脑应用程序，这种过渡对用户来说也很容易。 当然，对于开发人员来说，情况将会更糟。

I have done developer evangelism for 10+ years now. A lot of stars are going to have to align for Apple to pull this off. Here is a list of hurdles, off the top of my head, that Apple will face over the next two years until this transition is over:

我已经做过10多年的开发人员布道活动。 苹果要想做到这一点，将需要很多明星的帮助。 这是我的头等大事，在过渡结束之前，苹果将在未来两年内面临一系列障碍：

1. Software development tools will need to support the new ARM architecture.
   软件开发工具将需要支持新的ARM体系结构。
2. Access to test hardware for developers since emulators will not work here.
   开发人员可以使用测试硬件，因为仿真器无法在此处运行。
3. Customer adoption of new ARM hardware.
   客户采用新的ARM硬件。
4. A better software distribution model.
   更好的软件分发模型。
5. Income to cover the cost to migrate, test, and support new ARM software.
   收入用于支付迁移，测试和支持新的ARM软件的成本。
6. An emulation layer fast enough to run productivity Apps built for Intel chips on ARM.
   仿真层足够快，可以运行为ARM上的Intel芯片构建的生产力应用程序。

Apple has a few of these covered. Since most native Mac App developers have to use XCode, their primary tool can already help start the code transition. On the other hand, while Apple monopolizes software distribution on iOS, its desktop App Store has yet to take off. Apple’s mobile App Store is a race to the bottom model, so it’s unclear if premium-priced desktop Apps will ever generate enough money to help support both Intel and ARM versions. Finally, there may be some preliminary testing developers can do on an iPad Pro with a keyboard and mouse. It appears like Apple’s dev kit is running on almost identical hardware. But on the flip side, there is little shared UI between the tablet and desktop operating systems.

苹果公司涵盖了其中一些。 由于大多数本机Mac App开发人员必须使用XCode，因此他们的主要工具已经可以帮助开始代码转换。 另一方面，尽管苹果公司垄断了iOS上的软件发行，但其桌面App Store尚未起飞。 苹果的移动应用商店是向低端模型的竞赛，因此尚不清楚高端台式机应用是否会产生足够的收入来支持英特尔和ARM版本。 最后，开发人员可以在使用键盘和鼠标的iPad Pro上进行一些初步测试。 苹果的开发套件似乎在几乎相同的硬件上运行。 但另一方面，平板电脑和台式机操作系统之间几乎没有共享的UI。

While Microsoft had an idea to unify one OS across different form factors, Apple chose to unify multiple OS versions across a single chip architecture.

尽管微软有一个想法，可以在不同的外形尺寸下统一一个操作系统，但苹果公司却选择了在一个芯片架构上统一多个操作系统版本。

Even with Apple having many of the pieces it needs to pull this off thanks to iOS, this is an extraordinary time to transition over to ARM. Developers will have to reassess their desktop software development plans against the current reimagining of Apps on the iPad. While Microsoft had an idea to unify one OS across different form factors, Apple chose to unify multiple OS versions across a single chip architecture. The differences between the iPad Pro laptop and lower-end MacBooks, which will run on a similar architecture, are blurring the line between the two in unexpected ways.

即使苹果拥有许多功能，也需要借助iOS才能实现这一目标，但这仍然是过渡到ARM的绝佳时机。 开发人员将不得不根据当前在iPad上对Apps的重新构想来重新评估其桌面软件开发计划。 尽管微软有一个想法，可以在不同的外形尺寸下统一一个操作系统，但苹果公司却选择了在一个芯片架构上统一多个操作系统版本。 iPad Pro笔记本电脑和低端MacBooks(将在类似的架构上运行)之间的差异，以意想不到的方式模糊了两者之间的界限。

I bet that during this transition, macOS will move closer towards iOS as Apple intentionally forces developers to think iPadOS first when building desktop software. Developers, like always, are caught in the middle of this and will have to make some tough decisions on which OS to focus on, how to develop for it, and what they can support considering Intel Macs will continue to co-exist alongside these new ARM devices. The transition will be less painful for customers, assuming there is a proper Intel software emulation layer, and developers get on board early.

我敢打赌，在此过渡期间，macOS将向iOS靠拢，因为Apple在构建桌面软件时会强迫开发人员首先考虑iPadOS。 像往常一样，开发人员陷入其中，必须做出艰难的决定，要重点关注哪种操作系统，如何开发它，以及他们可以支持什么，因为英特尔Mac将继续与这些新产品共存。 ARM设备。 假设有适当的英特尔软件仿真层，并且开发人员可以早日投入使用，那么过渡对于客户来说将不会那么痛苦。

I guess we will find out in 24 months how this will work for Apple who, unlike Microsoft is going all in on its move to ARM chips.

我想我们将在24个月内发现这对苹果公司将如何工作，而与微软不同，苹果将全力以赴地转向ARM芯片。