Jonathan Cutrell

The Internet of Places: The Context of Things

2/25/2014

The Internet of Things is absolutely upon us.

Whether you’re fully aware and plugged in to tech rumor blogs and things like iotlist.co, or if you’ve never heard the term before, the Internet of Things is a huge wave in the way we think about technology.

A brief summary: everything is connected to the internet (is connected to everything), and everything is everywhere.

That’s the groundbreaking idea. To have an internet of things, you must have not only complex, multi-tasking machines like laptops and smartphones that connect to the internet, but also common elements of your every day life that connect to the internet. Lights, washing machines, thermostats, faucets, pet collars… You get the idea. If you don’t, try Googling an appliance, or really almost any concrete noun, followed by “wifi”.

This isn’t about a big red sticker saying “new and improved, now with WIFI!” - what it really comes down to is a different perception of control interfaces; I already do so much on my phone, why not centralize everything? Let me wash my dishes, pay my bills, set my DVR, and entertain my pet directly from my phone. And give me visibility to anything, anywhere. Oh, and save me some money in the process.

And it’s happening! It’s happening in my home, and it’s probably happening in yours. My nest has saved me… I don’t know how much money, but I really enjoy using it. And my Hue lights flash red when Alabama scores.

The ubiquitous dream. One day, the walls will be screens. Our homes will have ears, and our eyes will be everywhere. (This isn’t a discussion about privacy - imagine it’s magically solved.)

But let’s think further, beyond nouns. Let’s think in sentences, as humans do. Think about the context of our wifi-empowered nouns.

Context is everything. And the Internet of Place is the evolution of the Internet of Things. Things are nothing without their context, and the connectivity of things will only be followed by the connectivity of place.

Distinguishing Place From Space

At Georgia Tech, I studied mobile applications under Christopher Le Dantec. One of the primary themes of Chris’s work is understanding the relationship of mobilized technology as it relates to place, and the social science that turn a space into a place.

To unpack this a bit, let’s think about the lot that your home is built on. Scientifically speaking, that space has dimension and material makeup, and resulting properties. The space is located at a particular latitude and longitude.

Without any social understanding, the meaning of this space is lost.

Let’s start with sociocultural understanding: what country does that latitude and longitude land in? What state, city, and perhaps most importantly, what neighborhood?

We can get a sense for the mass sociocultural meaning, but what about the micro-cultural meaning in the space? A more elementary way of asking: who is your neighbor?

Now, the richest meaningful aspects of the space: who inhabits the space? Who has built the human history of participation in the space?

We attribute value to these things, both conceptually and concretely. Book two hotel rooms, exactly alike on their face; if one of them once held the President, I’d bet you’d be charged more for that room. Why? The space is the same; the place is different.

And truly, places aren’t the same to each person who comes along. The run-down garage where you and your father worked on your first car when you were a kid has a market worth of nearly nothing. But to you, it’s priceless.

The Internet of Places

Arguably, the most powerful motivator for technological innovation is the opportunity to empower us to make meaning in our lives more richly. The Internet of Things gives us the material to do this, but the Internet of Place gives us a full vocabulary.

What does it look like?

The Internet of Places will likely be made up of context aware applications combined with higher level machine learning to develop a sense of place-oriented interaction. Let’s imagine a few scenarios.

Football Game

I enjoy football. Going to a football game to me is a memorable experience. How awesome would it be for the stadium to know I was in it, and display my presence on a digital wall of inhabitants? Or maybe it’s as simple as having an application that allows me to press a button to order my hot dog and Coke from my seat so I don’t miss a play. This is absolutely possible right now with almost zero technological innovation. A more lofty scenario: my Google Glass listens to the game, and based on analysis of voice tension, language analysis, and game data, automatically records my perspective of the big plays. Later, I can go back and identify the ones it got right (hey machine learning folks - there’s your correction feedback loop), and share my perspective of the incredible touchdown pass. My memory of that play isn’t what they show on ESPN - it’s what I saw. This enriches my ability to revisit my important memories.

Shopping

Dear stores: Buy iBeacons and make awesome experiences for your buyers. Trade them their personal info for coupons. (Be responsible with their info.)

Imagine for a moment that you are in Wal Mart, and you can’t remember what you need. Your digital world remembers though, and your phone knows you’re at Wal Mart. You no longer keep a list of things you need to buy - your phone has already pre-ordered them and they are ready for you to pick up. (This seems like a very complicated scenario, but this is essentially how inventory management happens at the Wal Mart level; why can’t it happen at the My Home level, too?)

Another shopping example: you’re shopping for clothes. Your device of choice is your companion, letting you know what stores have your sizes in stock, and checking whether those sizes run smaller or larger than average. When you walk into the store, your device identifies, based on your past purchasing behavior, current fashion trends, local fashion trends, and friend comparisons via Facebook (or whatever network you choose), what you are most likely to genuinely be interested in. Perhaps a score is assigned to outfits, identifying things that are “out of the box” vs things that are “popular mainstream”. (Don’t crucify me - I’m not a fashion guy.) These value decisions you usually make on your own are now more informed. You have more confidence in your decisions, and the store has better information about you to ensure that your size is in stock next time you come in.

The implications are absolutely ginormous for retail; employing higher level learning algorithms for massive communities would change the concept of “lead time”, and could potentially virtually eliminate a massive margin of error in most inventory prediction methods.

At Home

When it comes down to it, what people want out of a smart home doesn’t put the emphasis on smart - it puts the emphasis on home. And that happens with context. For now, we have hyper-configuration, a tinkerer’s dream. Philips Hue lets me set a mood, for instance. We’re moving towards contextualization with automation and pipelines like IFTTT, and the next wave of home automation will move even further this direction. My home should know when it’s my anniversary, when I have friends coming over, and when I’m on vacation. I should be able to see meaningful analytics about my energy usage as it relates to my activities, not just simply the time of day. I don’t want my home to be a house-shaped human - I want it to be a representation of the people that live in the home.

My masters project was a good example of creating placeness in a home (or anywhere that music is played in an open space). I created an application that allowed people to check in to a given location, and would dynamically select music using everyone’s collective preferences. Once the person leaves the space, their effect on the music preferences decay over time; if they are a “regular”, their preferences remain strong for longer. This concept perpetuates a sense of collective identity of a place and its history.

The future of smart home technology doesn’t just let you set your alarm from anywhere in the world, or give out digital keys to let your friends in the front door. It knows you - it learns your moods and your behaviors, and connects with your personality. And not in a creepy invasive way, because your home isn’t personified. Instead, it picks relevant pictures and lighting to show on screens. It gives you suggestions for outfits on a given day, because it knows your wardrobe. It makes your coffee when you begin to wake up in the morning (instead of on a static timer). It keeps track of your belongings for you so losing your keys never happens again. Your presence isn’t simply presence; it’s contextualized presence.

Conclusion

The future of the Internet makes placeness a primary consideration, and elevates the Internet of Things. We begin to see machine learning conforming not only directly to measurable data, but also to subjective data - artificial intelligence in it’s smallest, most domain-specific application.

Discussion on Hacker News


Written by Jonathan Cutrell, Engineering Manager at Calendly and podcast host at Developer Tea.