Imagine a catastrophic social breakdown, Mad Max-style. An asteroid hits; pervasive coastal flooding causes sudden mass migrations; a genetically modified virus goes rogue; a crazed ideologue with an itchy trigger finger starts World War III. Whatever the case, the systems you’ve relied on for your survival are no longer functioning; you must fend for yourself and your loved ones. As you prepare to head out into the wrecked world in search of food, which watch will you wear?

Humor me with this. There are good reasons to want a timepiece in such a scenario. For example, you may wish to orchestrate maneuvers with fellow marauders as you embark on a raid. Or maybe you’re trying to calculate the speed of a raging river before you jump into it. The fact that all else has gone to hell doesn’t mean time has stopped. Sure, time is an abstraction — but a useful one, even under these dire conditions.

So, which watch will you wear? More specifically, consider these two options: An old Omega Speedmaster and a series four Apple Watch. The Speedmaster — a mechanical watch — only allows you to do a few things: you can tell what time it is, you can precisely measure how long something takes, you can measure how fast something is moving (including yourself, if you’re moving at high speed.) The Apple Watch allows you to do much more. Besides offering the same basic functionality of the Speedmaster, it also allows you to communicate with other people, see what the weather will be like, listen to Beethoven’s 9th Symphony (to remind you why persevering is worthwhile), etc.

In short, the Apple Watch allows you to do many more things that would be very useful in a post-apocalyptic world. The catch, of course, is that many of these features wouldn’t work in this case. When you strap an Apple Watch on your wrist, you’re not just wearing a device; you’re wearing an ecosystem. For example, the usefulness of the weather app depends on myriad things that exist well beyond the object on your person: sensors distributed around the planet, communication networks, data centers, the electrical grid, and so on.

The Speedmaster, on the other hand, depends on none of these things. As long as you remember to wind it every day, it’ll continue performing its functions for a long time. Of course, the day will eventually come when it’ll need repair. At that point, you’re out of luck. (I’m assuming Omega service centers will no longer be open for business.) But mechanical breakdown will likely be years away, even in this scenario. In the apocalypse, the smartwatch’s advantages would vanish within a few hours or days at most, and its essential functions would cease altogether when electricity reserves ran out.

The point of this mental exercise isn’t to get you into prepper mode. Instead, I’d like you to consider the nature of the things you interact with day-to-day — especially if you’re a designer.

We’re used to thinking of things as independent, self-contained objects. A cup is a cup. You can grab it, lift it, turn it around, dip it into liquid, bring that liquid to your mouth. Not much to it! A chair is just a chair. A watch is a watch. Except when it’s digitally enabled. Then it’s something more. Yes, it’s still an object you can pick up and manipulate. But that’s not the point. What’s essential about a smartwatch is that it gives you access to a range of useful features that are only available as long as the systems that enable them are in good working order.

For much of the time that people have been designing things, we’ve created things that are more like cups and chairs — and even Speedmasters — than Apple Watches. As a result, we tend to think of the things we design as individual artifacts with clearly defined boundaries; the kinds of things you can photograph and present in a beautiful coffee table book. Digital things aren’t like that. An app isn’t an individual artifact; it’s a part of (and a host to) very complex systems. Sure, you can show a comprehensive series of screen comps to illustrate what the app will “look and feel” like, but that’s not where its boundaries lie. The screens you interact with when you open the weather app on your smartwatch are a tiny shard of ice on the tip of an enormous iceberg.

Before digital, designers needed some degree of systems thinking. You can’t design something like a Speedmaster from scratch; there are hundreds of years of know-how that precede it, a specialized industrial ecosystem that will produce it, established business models that will get it onto people’s hands, and so on. A designer must understand these things to make the right tradeoffs. Many take these systems for granted as the context within which they’re working, but they must be aware of them nevertheless. Still, I’d venture most folks who design these things don’t think of themselves primarily as intervening in systems.

One possible exception is architects. More than other artifacts, buildings and towns depend on (and enable) rich interactions with their contexts; they depend on complex systems (e.g., transportation, energy, etc.) not just for their design but also to continue serving their functions. As Stewart Brand has pointed out, buildings also change over time as the needs of their occupants and stewards evolve. Smart designers create the means to accommodate change without making too much of a mess. This requires that they understand how the things they design function as systems.

Digital designers must think more like architects than like the designers of cups, chairs, or even mechanical watches. More than any other artifacts we’ve designed in the past, digital things participate in and enable systems. They’re also dynamic and interactive in ways that even complex mechanical devices like a Speedmaster aren’t. Bottom line: You can’t do a good job of designing a digital thing if you don’t understand systems. (This is one of the reasons why I think “product” is the wrong framing for digital things.)

I get tremendous value from my Apple Watch. However, I understand that that value is entirely dependent on complex systems that go well beyond the object on my wrist. When I design a digital thing, I frame it as a systemic design challenge: I look to understand the components and interdependencies that make the thing possible, and how they might change over time. I keep reminding myself that the boundaries for the thing I’m designing don’t lie with the organization that’s commissioned the work or even the operating system within which users will experience it, even if — especially if — stakeholders can’t easily see this. I must think of the thing I’m designing an intervention in one or more systems, and consider the second- and even third-order effects it implies.

And as much as I love my Apple Watch, if all goes to hell, I want a Speedmaster on my wrist.