IoT and integration–beyond simply a mesh network…

As I write and talk about IoT with various people and groups I have shared my three buckets a number of times. They seems to resonate with people. The concept is simple, there are devices you leave (home, office etc.) that you want to connect to. They are sensors or produce useful information you need to have. Those are stayables. There are devices you wear and use that screen as a service to have quick information available. Those are wearables. finally the last one is portable and it in most cases represents the device you carry with you that connects you to the other devices.

The next part of IoT that I’ve talked about a lot is the overall reality of bandwidth, personal privacy (and beyond that security) and integration. I realized that I haven’t really talked a lot about integration and it is one of the big three concerns.

In a mesh network the transitional reality of receivers becomes the issue. Football is a great example to consider for this. One quarterback, one ball and multiple choices for receivers. Receivers are engaged one at a time. The reality of sensors is similar today. Sensors are engaged normally one receiver at a time, although many more sensors now broadcast so you could in effect have multiple receivers. So a formalized connection is required. Be it a Bluetooth pairing or a connection to a secure wi-fi and then hardware coded to go out to the Internet.

Interesting problem of tomorrow – what happens when there are 300 sensors in your house call clamoring for your mobile devices attention as you walk through the house? That will drain your battery faster than talking to Aunt Jean for 12 hours will.

Integration becomes an interesting question. First off there would have to be two types of security applied. Information that you don’t care if everyone knows (your current weather from your personal station, anything you’ve shared publicly such as photos and videos and anything you publish (blogs etc). Information that is private (PII). Once we establish the rules around these two types of data we can start to create the integration points.

The second component of integration is the data. Data analytics is the meshing of what people call Big Data with the concept of delivery. The analytics are how you present the information for the user to act on. In the broad swath that is IoT integration you have to also consider the attributes of the receiver. I’ve talked about the concept of the screen as a service many times. Where the intelligence is in what is presented on the screen. Weather information on your watch face – useful. An entire spreadsheet with graphs showing the economic impact of a new item your company is selling on your watch – useless. Unless of course you are selling smart watches than showing your sales figures on your watch is – priceless.

The third (and while not final overall final for today) concept then is how do you get the information to the right device at the right time. Where data is the JIT attributed (just-in-time). Information has a life cycle. The value of information considers time, delivery and once delivered usability or ease of consumption. The three rules for this that I follow are below:

  1. Timely information (not out of date)
  2. Delivery doesn’t derail other activities (don’t use all my bandwidth to deliver the information unless it is life threatening information)
  3. Mind the gap

I did in fact steal the last one from the London Metro. What Mind the gap means is make sure the device you are sending to is capable of presenting the information you are sending. That excel sheet you just sent that ended up on the screen of my smart watch – overloaded the smart watch and made it reboot.

more to come…


Scott Andersen

IASA Fellow