One of the most tangible and noticeable trends of the past decade is the overwhelming amount of data of all sorts we need to work with and make sense of. From car telemetry to grocery chains’ sales statistics, enormous heaps of numbers and figures are gathered nowadays. Obviously, an important task here is to make sense of the numbers, monitor them and make decisions accordingly and in time.
The Internet of Things has contributed quite significantly to the global flow of data with its sensors, logic controllers, user inputs and so on. It’s no less important in this industry to swiftly crunch the numbers, see what’s behind them and assess the situation.
From the trend of “quantified self” to industrial and smart-home sensors, comprehensive and straight-to-the point visualization of data is a must for any IoT appliance. The task becomes ever harder with the flow of data increasing.
Let’s take a fresh look at the state of the art in visualization: what are the dos and don’ts, and which tools you can use to create clear and smart visuals for your valuable data.
Tips and tricks
The basics of good visualization are simple, and yet examples of awfully designed dashboards are aplenty, so it makes sense to repeat them once again.
Align your information horizontally and vertically so that patterns in graphics would be clearly visible. In addition to that, make sure your alignment doesn’t suggest irrelevant correlations and comparisons.
Consolidate data you’re visualizing whenever possible. If two different graphs can be represented as one (for instance, when they share one axis), you probably should go for it to make your visualization clearer.
Don’t over-adorn. In fact, don’t adorn your visualizations at all, it’s a bad practice. Try to refrain from using all kinds of 3D effects and skeuomorphism in your graphs and diagrams. These only make the visualization more difficult to understand, and also look out of place in most cases.
User comes first. The best practice in visualization is to think about the user and their needs before getting to design the dashboards and charts. Try to look at your product from the point of view of different types of users and write down possible scenarios and user stories. When you have the first version of how your product’s visualization is going to look like, test it against those stories and see if it’s clear enough for the customer.
We’ve taken a look at the wide variety of tools you can use to visualize your priceless IoT data and selected four of them for you to consider.
Launched by the creators of Matlab, this is a versatile open source platform and API that allows developers and app designers to gather data from sensors and other sources, analyze and visualize it. If you’re familiar with Matlab, this is going to be a breeze; if not, ThingSpeak doesn’t seem too hard to learn, especially with good tutorials and documentation. The whole thing works around concepts of channels that store data and apps, which can transform and visualize it. What else is good about it is that the platform is completely free to use.
This is a purpose-built visualization tool for the Internet of Things. It allows you to create a dashboard full of different widgets and immediately share it with anyone. To collect data, Freeboard offers integration with the Dweet.io IoT messaging system but also can talk to any web-based API out there. Widgets, which are the main thing on the platform, can be either created from scratch or chosen from a list of pre-defined ones. The platform is open source but isn’t entirely free: the pricing goes from $0 to $100 a month, with the main difference being the number of private dashboards you can create.
The cloud solution from the Big Blue has a quite nice and easily deployable visualization tool set of its own. The centerpiece of the solution is the IBM Internet of Things Foundation, which works as a hub, to which sensors and other data sources can be connected. The data is sent via the MQTT lightweight messaging protocol.
Here’s a visualization example of Sogeti with an explanation:
“Data and events are forwarded as they occur by each gateway to IBM’s IoT Cloud by sending JSON over the MQTT Protocol. A Bluemix application processes incoming messages based on a flow that can be configured separately per room. Some of the data is just captured and stored in the historian. Some of the data is consumed by the live user interface. Some of the data can trigger alerts.”
IBM offers a flexible pricing system, where users could be paying from nothing to $1,500 per month, depending on the number of instances and amount of memory used.
The IoT application platform by DGLogik offers a lot more than just visualization but can be used for that as well. It provides a number of assets like animated widgets, background themes, patterns, effects and so on. The programming of the dashboard is also visual, which makes the life of those less familiar with coding much easier.
Armed by the best practices of data visualization and your tool of choice, working on your IoT appliance is a breeze.