Okay, it looks like I’ve decided that I need a series of “primer” posts in order to ensure some baseline level of knowledge among my readers. If this post or the big data post are uninteresting or too basic, hang tight: more complicated posts are coming. I just don’t want to lose anyone along the way. Today’s topic: “The Internet of Things.”
Like “Big Data,” “The Internet of Things” is neither radically new nor particularly earth-shaking, in and of itself. However, in the same way big data is already changing the way we answer questions, the Internet of Things will change the way we live.
Remember when the next big thing was going to be internet-connected toasters and smart refrigerators that knew when you were out of milk and would automatically order it for you? Or how cars would communicate with each other to pass traffic information between them and your house would know when you (as opposed to your spouse or your kids) came home and configure itself (set the lights the way you like them, turn on the TV or the stereo as appropriate, adjust the thermostat, etc.) just for you?
This phenomenon, when it had a name, was usually called a “Connected House” or “Smart Car” or some other catchy moniker. Advancing technology and nearly-ubiquitous WiFi or cellular connectivity have moved these out of the realm of Popular Science and into the realm of Amazon. As these things become more and more of a reality, they’ve all been collected under a single rubric, “The Internet of Things.” The basic idea is simple: there are lots of devices without user interfaces that might benefit from—or benefit us by—being attached to the internet. It’s that lack of user interface that makes this different from the plain-vanilla internet that we interact with on a daily basis.
For most of us, the internet is a collection of services that users interact with through various clients. There are web pages we access via a browser; email services and email clients; music services, like Spotify, which have their own clients and web-based access; and lots more. In all these cases, however, the primary consumer of information is a human and, generally, the information was created and published by a human, whether it was the author of the email, publisher of a web site or artist/producer of the songs.
But this doesn’t have to be the case. We already have machine-to-machine (M2M) communications allowing servers to talk to servers, but, what if servers talked to stuff—not just computers, but everyday objects with embedded technology—directly, and humans were only second (or higher) order consumers? If there is no need for a human interface, how many of these objects could share their useful data with other things?
For instance, what if your thermostat talked with the electrical grid to decide when it would cost least to turn the AC on during the day? It could even negotiate with the other air conditioners in the neighborhood to stagger run times so that the overall demand was flatter. Perhaps the electric company would negotiate with you for a reduced cost per kwh if you agreed to a usage schedule that your thermostat could implement and they could monitor. While the Nest thermostat doesn’t do all those things yet, it can be managed and set from your iPhone, today. Taking the next step is easy.
What if your car could contact the shop when it was broken, tell them what was wrong and have them contact you for an appointment time? Assuming you had already set up some preferences, they could even make sure you had an appropriate loaner car ready and a convenient drop-off time.
What if every traffic light had the ability to alert any passing maintenance vehicle that it was out? Better yet, what if the dispatch computer had already taken the outage into account and ensured that the right replacement lamp was on the vehicle that was scheduled to drive by that intersection anyway?
You might have even seen an ad by IBM describing machines that take themselves offline to receive maintenance when required, or the one CISCO is running about trees talking to networks and cars talking to roads. All of these are examples of the phenomenon known as the Internet of Things, or IOT (since we can’t turn down the opportunity to create another TLA*).
And don’t laugh. Toasters and coffeemakers are very important in technology development. The first webcam was installed to keep an eye on a remote coffee pot to avoid pointless trips to the coffee room (in case the pot was empty or not yet done brewing).
Admittedly, as neat as all of these things seem, none of them is particularly transformational by itself. The power of IOT comes as more and more things are online and producing and consuming data. This is one of those places where the emergent capabilities are going to far outstrip anything we can think of today or anything we can intentionally design.
Before we can get there, though, there are three primary problems that need to be addressed:
- Protocols, and
Addressing is the simplest of these to deal with, but is far from trivial, if a bit wonk-y. We need to find a way to provide these multiple billions of things with network addresses that are findable and usable by other arbitrary internet-connected things.
Protocols need to be standardized. Right now, nearly every manufacturer is using their own language to get devices to talk to one another. IOT doesn’t work, really, if your fridge doesn’t know how to speak toaster and if you have to worry about whether your Samsung phone can communicate with your BMW. Or if your car stops receiving traffic information when you cross city, state or county lines or international borders where they just happen to implement a different vendor’s equipment.
There are also going to be emergent problems.
All these problems, though, require significant discussion beyond this post and will be topics for future posts. In the meantime, hopefully, you feel like you have enough information to hold your own at the coffee shop when the Google Glass-wearing fellow next to you wants to talk about the internet-controlled mood lighting in his apartment.
* TLA is my all-time favorite Three Letter Acronym**
** My least favorite is “www.” It’s the only acronym I know of that has three times as many syllables as the thing it is supposedly shortening.