1. Every activity that can be made digital will be. I don’t have to physically adjust my thermostat, it’s controlled by a computer.

2. Every activity that can be made digital will be put on a multi-purpose (or “convergent”) device as an app. I don’t need an alarm clock, I’ve got an alarm on my mobile phone. This phenomenon is starting to be labeled “ephemeralization.”

Where then does this leave single-purpose, single-function devices? Are they to vanish into history? Some certainly will. For example, cameras, MP3 players, and video cameras are taking a beating from mobile phones. But I believe there is still a place for single-function devices, under specific circumstances.

A single-function device (an “appliance” as Mike Kuniavsky calls them, vs. a multi-function “terminal”) makes sense when:

• It can do more/is more powerful than similar functionality in a terminal. You won’t find professional photographers using Android phones to take wedding pictures, for instance. The power and control of an object designed to just be a camera outweighs any inconvenience of carrying around a specialty object to take pictures.
• The physical activity the appliance engenders cannot be replicated digitally. In other words, an iPad isn’t going to substitute for your dishwasher.
• The integration of hardware and software is key. A variation on not being able to be replicated digitally, some activities are simply easier when there are physical controls. It is still easier to type on a physical keyboard than it is on a touchscreen one.
• It is prohibitively expensive or dangerous to make the activity digital. In theory, I could make an iPhone app to drive my car, but I think government agencies might have something to say about that.
• There are a few billion people out there without a smartphone, tablet, or laptop. If your target base isn’t likely to have an expensive terminal device, an app on one isn’t going to work for them.
• The device has to be stationary. Some devices have to be hooked into the systems of a location (plumbing, electricity, gas, water, etc.) to work. Likewise, some sensor-driven devices only work (or only work well) if they are calibrated for a certain space. For instance, Microsoft Kinect.
• The form factor of any terminal device doesn’t match the activity. My $500 non-waterproof iPad isn’t the object I want to take on a scuba-diving trip.

Now I grant you some of these items are conditional—perhaps in the future the cameras in multi-use devices will be as good as professional ones. Even so, I feel there is always going to be a place for beautiful objects that one do one thing—but only if the integration of hardware and software is such that it makes accomplishing the task so much easier/more efficient/more fun than a multipurpose device would. That’s the challenge: how to differentiate your single-purpose device so that consumers will buy it instead of (or in addition to) an app that does the same thing. Designers have to not only explain why you’d want to own the chef’s knife instead of a Swiss Army knife, but also make the specialty item worth owning. That’s the challenge.

• Tabs: “inch-scale machines that approximate active Post-It notes” Mobile phones, and devices such as mobile cameras and MP3 players fall into this category, although most of these are larger than Weiser likely envisioned. Very few devices currently are as small and light as a pad of post-it notes. Digital watches and some medical devices, perhaps.
• Pads: “foot-scale [devices] that behave something like a sheet of paper (or a book or a magazine)” Laptops, tablet PCs, and e-readers are examples of this category.
• and Boards: “yard-scale displays that are the equivalent of a blackboard or bulletin board” We’re making progress on this front, with large touchscreen walls, not to mention our flat screen TVs.

Clearly, Weiser was thinking mostly of an office/workplace environment with his categorization scheme. If we expand our view out a little more to include homes and public spaces, I think there are some other general types that could be included in Weiser’s categories:

• Dots: tiny, nearly- or completely-invisible devices.
• Boxes: devices that are slightly too large, bulky or heavy to be portable. Kitchen appliances such as toasters and many consumer electronics like stereo equipment fall into this category.
• Chests: large, heavy devices that do not move, such as dishwashers, stoves, and refrigerators.
• Vehicles: large, heavy devices that do move. Yes, cars are devices too, writ large.

It helps to consider these categories, because each brings with it different expectations as to how to engage with the device and what kinds of functionality might be available, not to mention qualities such as durability and cost. Device designers need to be aware of this when considering what form a device should take.

Of course, traditionally, this is has likely been determined before the designer even starts the project: “We need a new line of washing machines. Go make us some.” But in the 21st century, it’s not quite as simple. Frequently now, the starting point is not the form, but instead the functionality, with the form enabling and enhancing that. As electronics and computer components get smaller, it’s easier for what might have been a Box at one point to become a Pad, Tab, or even a Dot and “disappear” all together. There’s more computing power in a digital watch these days than in the NASA capsules that went to the moon. Functionality, and thus form, is more fluid and cross-category.

Of course, a Vehicle is not likely to become a Dot, but the functionality such as a navigation system that might have previously been stored in a car (in the dashboard, say), might now be in the Tab that is your mobile phone.

There is also an emotional response around different forms. We have a different relationship to our mobile phones than we do to our stoves, or to our cars as to our laptops. Some of this is certainly about price point (our cars cost more than our mobile phones), but some is the pure physicality of the object itself. A large, heavy, immobile object usually has more emotional weight than a small, mobile one. If just for the fact that it has a larger spatial (and fixed) presence in a room. You would immediately notice if it was gone. This is a tricky problem as more objects dematerialize and become Dots. Will you feel differently about your TiVo/DVR if there was no Box and no remote (only gestures for control)?