Doc Searls’ Clouds For Things: Yes, Please, But!


My friend Doc Searls describes the benefits of having an electronic place online for everyday things. He calls that place for a thing a “Cloud For Things”. For example:

The house where I’m a guest in London has clouds for all its appliances. All the clouds are physical. Here they are:

Each envelope contains installation and instruction manuals, warranty information and other useful stuff.

Now let’s imagine a simple digital container for each appliance’s information: its own cloud. In form and use, it would be as simple and standard as a file folder.

I’m totally in agreement that this would be really useful, and I wholeheartedly support the idea. I’d love to have it for all of my things in my house!

But there’s a big But. The But is that that “cloud” must not owned/controlled by the vendor of the thing after the thing has been sold to me. It must change ownership, i.e. full control, from the vendor to me, when I buy the thing, and it must move with the thing if I sell the thing to somebody else.

Why? It’s because if the thing’s electronic twin remains under control of the vendor, we end up with the all-so-pervasive Overlord architecture for the Internet of Things, where we never actually own anything, although we paid for it. My favorite example being the Nest thermostat, which we supposedly “buy” but over whose behavior we have no control, and which will stop working once we disagree with its terms of service. (Note: if there are terms of service for something, we don’t own it!)

Coincidentally I’ll be talking about the contrast of the Overlord architecture for the internet of things, and what I call the “Indie IoT” this Tuesday at the Silicon Valley IoT Meetup. The essence of Indie IoT is that the “clouds” for the things are independent from the vendor, and owned and controlled by the owner of the thing. They have many advantages, from privacy to being a much better platform for innovation.

Turns out that only the Indie IoT architecture can actually deliver many of the benefits that Doc sees in his Clouds For Things proposal: customers helping customers, uncensored, reduced customer support costs and so forth. If the vendor remains the Overlord (“Ha, these clouds allow me to dictate/upsell/advertise to the customer even after the sale”), customers won’t use them, and the whole effort is for naught.

Unfortunately, also, we do not have the entire technology in the market necessary for white goods vendors to actually deploy such Indie IoT clouds for things. (“How does the vendor initiate that cloud, and then transfer control, including root password, to the customer?”) But with things such as Personal Clouds, Linux distros such as UBOS, and personal servers such as Indie Box, we will get there and it won’t be very long!


3 responses to “Doc Searls’ Clouds For Things: Yes, Please, But!”

  1. Thanks, Johannes.
    I should have made clear that the thing’s cloud is clearly the customer’s. So I just updated the text in my post and pointed back to this post of yours as well.