databox sdk primer

The SDK is a visual coding environment for building databox apps. All you need to get started is a github account. You can then login at to start building apps. This page provides a brief overview of the SDK, and will link to more detail where relevant. We have created a set of tasks which involve building databox apps. You don’t have to work through them sequentially, and we have grouped them into tasks that require NO programming, and tasks that assume (basic) programming proficiency. Feel free to ignore our suggestions and build your own apps.

A quick primer

The SDK is a visual coding environment that requires you to connect together ‘nodes’ into ‘flows’ to create an app. Nodes are small units of functionality. Aside from a couple of ‘special’ nodes, they will either be datastores (things that create data e.g. twitter, output from a smart plug), processors (things that operate on data, e.g. format it for display, filter out items) or outputs (things that do something with data e.g turn on a bulb). Typically a flow will consist of one or more datastores connected to one or more processors, connected to one or more outputs:

example flow
example flow

An app can be tested at any time during its creation. Once finished, it can be published, after which it can be run on a databox.

Your first flow

To get started, login at Once logged in, you will be presented with a palette of nodes down the left hand side of the screen and a grey canvas:

sdk main screen
sdk main screen

Drag the sensingkit datasource onto the main canvas. Now drag the debugger node onto the canvas and connect your datasource output to the debugger input. That’s it. You have created your first flow. All that the debugger node does is display the data that it receives.

To test the output, click on TEST on the top blue toolbar. This will build the app and run it in a test environment. After a short while, you should see the debugger node appear on a sidebar on the right hand side. Click on it and it will open a new tab with the debugger output. If you go back to the SDK screen and double click on the sensingkit node, you’ll see that there are a variety of sensors that you can choose. Select a different sensor, then click on TEST, and you should see data with a different format from before.

Modifying your first flow

Let’s modify your flow again. Delete the connection between sensingkit and the debuger. Now drag in the chartify node and the app node. Connect the sensingkit to the chartify node, then connect the chartify node to the app node. The app node simply displays information on the screen - in this case it will display whatever comes out of the charfity node. Double click the sensingkit and select the light source. Now double click the chartify node. You’ll see a bunch of options. For “type” click on gauge. In the “chart sources” section at the bottom, we want to plot the value of the light source. Select VALUE under the xvalues heading. Click on OK, and you are all done.

Click test again, wait a while for your app to build and you should now see that there is an app node in the right hand side sidebar. Note that the tab running the old test will no longer show data; each time you create a new test, a new tab will display the output from your new app. Click on the new tab and this time it should display a gauge, with a circle that moves in response to the change in data.

You can connect multiple sources of data to the app node. Try dragging in a listify node and connect it to sensingkit and the app node and re-run the test. Try changing the chart type to a bar graph, and now select the x and y values as “ts” and “value” respectively.

Next steps

This is all you need to get started with the SDK. Go here for a fuller overview. You’ll see that there are a bunch of examples of apps that do other things - by all means load them up and take a look. We have also provided a summary of the nodes available in the sdk. You should have most of the information that you’ll meed to work through a list of suggested apps.