Editor’s note: John and Noel (@noelportugal) need to chat about Google’s Physical Web gBeacons.
I have been a tad skeptical about the usefulness of smart watches, but my colleague Julia Blyumen has changed my thinking.
In her recent blog post, Julia noted that a smart watch could become both a detector and a universal remote control for all IoT “smart things”. She backed this up with a link to an excellent academic paper (pdf) “User Interfaces for Smart Things: A Generative Approach with Semantic Interaction Descriptions.”
I strongly encourage anyone interested in the Internet of Things to read this paper. In it the authors lay the foundations for a general purpose way of interacting with “smart things”, interactive components that can sense and report on current conditions (counters, thermometers), or respond to commands (light switches, volume knobs).
These smarties (as I like to call them) will have much to tell us and will be eager to accept our commands. But how will we interact with them? Will they adapt to us or must we adapt to them? How will we even find them?
The authors propose a brilliant solution: let each smartie emit a description of what it can show or do. Once we have that description, we can devise whatever graphical user interface (or voice command or gesture) we want. And we could display that interface anywhere: on a webpage or a smartphone – or a watch!
Another one of my AppsLab colleagues, Raymond Xie, immediately saw a logical division of labor: use a phone or tablet for complex interactions, use a watch for simple monitoring and short command bursts.
Another way a watch could work in concert with a phone would be as a “smartie detector.” It will be a long time (if ever) before every thing is smart. Until then it will often not be obvious whether the nearby refrigerator, copy machine, projector, or lamp is controllable.
Watches could fill this gap nicely. Every time your watch comes within a few feet of a smartie it could vibrate or display an icon or show the object’s state or whatever. You could then just glance at your wrist to see if the object is smart instead of pulling out your phone and using it as a dowsing rod.
One way of implementing this would be for objects or fixed locations (room doors, cubicles, etc.) to simply emit a short-range bluetooth ID beacon. The watch or its paired phone could constantly scan for such signals (as long as its battery holds out). If one was detected it would use local wifi to query for the ID and pull up an associated web page. Embedded code in the web page would provide enough information to display a simple readout or controller. The watch could either display it automatically or just show an indicator to let the user know she could tap or speak for quick interactions or pull out her phone to play with a complete web interface.
An example I would find useful would be meeting room scheduling. I often arrive at a meeting room to find someone else is already using it. It would be nice to wave my watch at the door and have it confirm who had reserved the room or when it would next be free. Ideally, I could reserve it myself just by tapping my watch. If I realized that I was in the wrong place or needed to find another room, I could then pull out my phone or tablet with a meeting room search-and-reserve interface already up and running.
But that’s just the beginning.
One of the possibilities that excites me the most about this idea is the ability to override all the confusing and aggravating UIs that currently assault me from every direction and replace them with my own UIs, customized to my tastes. So whenever I am confronted with a mysterious copy machine or the ridiculously complicated internet phone we use at work, or a pile of TV remote controls with 80 buttons apiece, or a BART ticket machine with poorly marked slots and multiple OK buttons, or a rental car with diabolically hidden wiper controls, I could pull out my phone (or maybe even just glance at my watch) to see a more sane and sensible UI.
Designers could perfect and sell these replacement UIs, thus freeing users from the tyranny of having to rely on whatever built-in UI is provided. This would democratize the user experience in a revolutionary way. It would also be a boon for accessibility. Blind users or old people or children or the wheelchair-bound could replace any UI they encounter in the wild with one specially adapted for them.
Virtual interfaces could also end the tedium of waiting in lines. Lines tend to form in parking garages and conference registration because only one person can use a kiosk at a time. But if you could tap into a kiosk from your smart watch, dozens of people could complete their transactions at the same time.
Things get even more interesting if people start wearing their own beacons. You could then use your watch to quickly capture contact information or create reminders; during a hallway conversation, a single tap could “set up meeting with Jake.” Even automatically displaying the full name of the person next to you would be helpful to those of us who sometimes have trouble remembering names.
If this capability was ubiquitous and the range was a bit wider you could see and interact with a whole roomful of people or even make friends during a plane ride. Even a watch could display avatars for nearby people and let you bring any one into focus. You could then take a quick action from the watch or pass the selected avatar to your phone/tablet/laptop to initiate something more complex like transferring a file.
Of course this could get creepy pretty fast. People should have some control over the information they are willing to share and the kind of interactions they wish to permit. It’s an interesting design question: “What interaction UIs should a person emit?”
We are still at the dawn of the Internet of Things, of course, so it will be a while before all of this comes to pass. But after reading this paper I now look at the things (and people) around me with new eyes. What kind of interfaces could they emit? Suddenly the idea of using a watch to dowse for smarties seems pretty cool.
