The problem with the Internet of Things

Updated: April 2, 2021

It's been about 10-15 years since I first heard the term Internet of Things (IoT). Like a wild ninja, it appeared out of nowhere, harbingered by excited management, who then tasked different teams to find problems to solve using this new concept. And I thought, wait a mo! Shouldn't it be the other way around? As in, you have real problems, and then you find the most adequate solution for them? Well, it's been 10-15 years since. And like U2 sings, they still haven't found what they're looking for.

Which is why I wanted to write this article. Ponder a little about the whole concept. What does IoT really do, and how it actually, supposedly helps and improves lives, if at all? So turn your cynicism thrusters to 11, lean back and listen to a dinosaur explain why, sometimes, despite your best enthusiasm, the fact you can do something doesn't mean you should.

Teaser

Everything is connected, but why

The idea, on its own, sounds pretty nifty. Everything talks to everything, and this gives you an elevated level of situational awareness, which then lets you optimize the reality around you. Think smart cities, where you have an instant, up-to-date view of the traffic at any point along the infrastructure. Think global container shipping - billions of tons of cargo going around the world in ships, sometimes months out of port, and being able to fully and accurately correlate all this in real time and not having to rely on paper manifests. These are real problems, and they do require solutions. Embedding low-power radio sensors in the moving parts makes a lot of sense.

But the leap from the industrial and urban backbone to ... everyday crap around you is a big one. Perhaps too big to do without some levitation assistance from over-enthused sales buzzword armadas. And this is where a new problem is born out of a solution to a problem that doesn't really exist. By connecting everything, you simply complicate the reality without providing the simplicity the "solution" ought to bring.

I am talking of course about home automation et al, and this is what most people coincide with the notion and application of the IoT. After at all, it is much easier to demonstrate and put into practice than having to wait 40 years for the world to lurch into the next generation of digitally supported commerce. Thus, actually, tragically, the IoT sells itself short by talking about small and impractical ideas like smart refrigerators, smart lighting and whatnot, instead of focusing on genuine problems.

Home automation = lazy done the wrong way

Yes. You want your home to be automated. But what does this actually mean? As it turns out, it means putting devices with embedded operating systems around your house and making them do stuff - stuff that you don't actually need, but stuff that you CAN do because 1) it's cool 2) you're a nerd who likes techy stuff 3) you have the technological ability to do.

Let's have a look at the two examples I mentioned earlier. Smart lighting. You can control your light level and hue using an app. Okay. Why? How's this different from manual dimmers that were commercially available back in the 80s? How's this necessary? Why would anyone actually need to change the lighting levels in their home? And then, if they actually do need (not just want because they can), how often would they need this? Also, practically, the time it takes one to unlock their phone, launch their app and then adjust this or that thing isn't any faster or more efficient than getting up and doing it yourself.

Along the way, we ignored the actual setup - the time it takes to configure the smart lighting. Then, the biggest issue of them all, by adding this connectivity (possibly over the Internet too), you actually expose your home network to the wider Web, just so you could control lights from a distance, which most likely serves no purpose, as you're not around to enjoy said lights. Indeed, the security and privacy problems issued embedded in the smart devices are significantly greater, more complex and harder to solve than the original problem - which is controlling the lighting in your house.

Drone

Imagine a drone that cleans windows, cleans drains, scares pigeons away, all the while playing Wagner's Ride of the Valkyries, and you pilot it with your phone. Thinking about it, sounds like something I may actually buy ...

You must have read about various problems related to smart devices - root certificates that expire, making the devices unable to communicate with online servers, hard-coded admin users and passwords, open ports, out-of-date firmware, companies going out of business and shuttering their gateways, rendering devices into lovely plastic bricks, and then some. Of course, that doesn't mean the entire IoT experience is like this. Far from it. But even if we're talking about isolated cases, it's still more than ZERO problems one has with a wall switch fixture that dumbly turns electricity on/off.

Smart fridges - another example. Say your fridge can help you track your inventory of perishables and even assist you in repurchasing them, so you never run out of stock on your eggs, milk and lettuce. Not bad, but ... What if you want to buy something that the machine cannot guess or predict? What if you want to buy different quantities? How does one actually track the contents of the fridge? How does one avoid the almost inescapable trap of "personalized" ads associated with one's shopping habits? Does it make sense for a cooling box in your kitchen to run an entire operating system and a dozen applications just so it can ping: you need more carrots.

But even if the above solution works (smoothly), there's a bigger question. Do you need it? How many people really live such hectic, modern lives that they can't invest three minutes a day opening the fridge door, checking out their stock, and then adding a note for their next grocery trip, or for that matter, the online order in the evening? And if they really live like that, perhaps they should fix the root cause - their lives.

You can [insert favorite fad here]

When confronted, this is how IoT enthusiasts explain their new hobby. You ask them why they NEED it, and they say BECAUSE YOU CAN. But I never heard an actual explanation what fundamental human need is satisfied by the introduction of this technology. This reminds me of conversations I had with people when smartphones became popular. I was being bombarded with: you can edit files, you can crop photos, etc. My take on this has always been, well, you can, but should you? You can edit files and crop photos on the desktop, and more efficiently, too. Why would you do old, already solved things in a new, suboptimal way?

Smartphones did solve problems - but they are none of what the nerds want or need. They allowed people to have cheaper, opportunistic computing devices that they can carry everywhere. They allowed people to combine several functions that used to require multiple devices in one, like music and camera. They made it possible for people who barely use their computing devices not to have to have a huge desktop PC tower in a room somewhere. Real problem, real solutions.

Even so, the touch frenzy also ruined a whole bunch of things. Many vehicles have touch-only infotainment systems, including things like AC controls! So now you have to take your eyes off the road to adjust simple things that you could previously do without actually having to look. Silly beyond words. Just because you can, doesn't mean you should.

But let's go back to the debate about practical value of new technology. Today, the same is happening, except the terminology has changed. When I ask people who added a whole bunch of sensors to their homes why they've done it, the answers always get elaborate and defensive. One could always say: BECAUSE IT'S FUN, and I can't argue with that. If you enjoy doing techie stuff, then go for it! Lovely jubbly. There's no reason to pretend that there's some higher functional goal. And let's be frank; how many people you actually know that have introduced automation to their homes and who are NOT techie lovers? Exactemondo.

It's hard predicting the future, especially when one looks at the world through the lens of their own personal experiences. There was no guarantee the phones would take off so majestically. For instance, netbooks, the low-end quasi-laptop-cum-tablets, didn't. Thus, the future of the IoT remains ... veiled. That said, I still struggle with the mission and purpose, especially when I base my judgment on the current usage models. I just can't seem to get a single, nice, definitive answer - from myself or any of the IoT diehards - the one thing that this new field truly solves (for the ordinary person).

The great divergence

Another problem facing the IoT world is the billion different ways it has been realized. There are as many mutually incompatible software permutations as there are IoT device models out there. In fact, there are more, because there are tons of variations in the operating system + app combo for any which hardware model. And largely, these are all incapable of talking to one another, because every company has its own unique implementation.

But it gets worse. We must also account for greed and hubris. Everyone wants to be THE company whose solutions prevail and become the dominant technology (ergo the deciding factor) in the IoT space, not at all unlike the Internet. Everyone wants their API to be the standard, everyone wants people to use their product. And exclusively so. Because there's power in numbers and even more power in data, especially if it flows through your network.

The same is happening across the board. So-called "exclusive content" - whereby you get data and information only if you submit yourself to a very specific, custom-tailored ecosystem. Want this or that movie or song or whatnot? Well, you need this DRM-ed app (or something like that). The old Internet used to be wild and open and it actually did what it was designed to do. The new Internet is a bunch of silos, isolated little domains that don't talk to the rest of the world.

And the same is happening in the IoT space. Not just that, the entire hardware space. Your camera will complain if you use non-vendor battery, your printer will complain if you use non-vendor ink, this or that thing won't work unless you do things in a very specific way. Perhaps this makes for a more predictable and robust experience, MAYBE, but it also transform the technological landscape into narrow, pinpoint areas of fierce exclusivity.

Vegetable skin peeler

Imagine a peeler that only works with exclusive, certified vegetables. Golly.

Does the ordinary person care? No, not really. Think automotive industry. People happily buy their favorite brand of car, they don't mind (much) that they can only service their vehicles in approved garages, and by and large, they have a pretty decent journey. As a petrolhead, I can tell you that cars today are safer, more reliable and overall better than they were 20-30 years back. So there's a benefit to this approach. And exclusivity.

But that's it. Each vendor does their own thing. It's not an interconnected world. It's many small isolated worlds. The same is happening to the Internet of Things. It should perhaps be renamed to Many Separate Local Area Networks of Things Done Over the Internet. Or the Internet of Nerds (IoN).

Maybe we're too early

I am being harsh, of course. The IoT is still very young. Perhaps we are simply trying to do something two decades too early. Compare the airplane from 1900 to one from 2000. A car from 1910 to a car from 2010. The Internet from 1990 to what we have today. Things have changed, considerably. But.

The core purpose of the car or the plane has not changed in the last 100 years. Technology lets us do things faster, safer, cleaner, but we still lug ourselves at speeds that exceed foot or horse, distances that exceed the biological engines, we can have AC in the summer, and we can even listen to music while doing it. Great. But at the end of the day, 'tis just transportation.

The Internet is a different beast. It's changing (and not for better). The purpose of the whole thing was to interconnect people and machines around the world. It's an information medium. Now that this purpose has largely been fulfilled, new problems are being designed to justify the new incarnations of the Web (the whole 2.0 and 3.0 buzzwordology). But then, ask yourselves, is the Internet today any better than it was several years ago? Are you having more fun and purpose in life now that you have social media and apps? Is your life fundamentally different?

Future

Believe it or not, this was a real thing.

Regardless, with the IoT, we're definitely too early. Because the current state of the technology only allows for trivial, banal and useless use cases. The implementation is far from being robust enough to allow ordinary folks to enjoy the IoT without introducing seven new problems along with their one shiny solution. Twenty years down the line, we will definitely have better capabilities.

But my original question remains: what will the IoT really solve? Or more precisely, in the consumer space (not industry), what will the IoT solve? It's been 10-15 years, and I still don't have an answer. Perhaps I'm an idiot. Or perhaps there's no problem to solve to begin with.

Conclusion

You may think I'm an imagination-less oldster who cannot immerse themselves in the new and innovative. Wrong. There are lot of new technologies out there, which definitely have amazing potential and use cases. Just to give you my perspective, say AR/VR. Lots of buzzwords, but this is actually something cool. AR/VR will solve real problems in hazardous environment work, healthcare. AR/VR will provide significant benefits to disabled people or those with learning difficulties. AI/ML may have some actual value, if we actually ever get to understand how the engines self-optimize.

IoT? Don't know. It may have value in the industry space, but in the consumer space? Ah. This is where I don't see any practical application of smart devices. Not because they aren't smart or (potentially) practical - because humans are stupid, impractical and random. And because, after several thousand years of careful tweaking and optimization, we have pretty much nailed down the analog way of doing things. Anything else that requires the human to give control over to machines implies inefficiency. Replacing analog with digital isn't the solution. That's just refactoring. Furthermore, finding new, real needs in the consumer space is going to be hard. If people use it, the need has already been met, or if there's a need, something has or will have been made to satisfy it. Which is why IoT is a hammer looking for a nail to beat. Well, if someone has a good answer, I'd love to hear it. For the time being, I shall do my cynical, skeptical thing.

P.S. All of the images used above are in the public domain.

Cheers.

You may also like: