A few months ago i started tinkering with the Raspberry Pi, it’s such an amazing and beautiful platform to start exploring your ideas. As i am also running a business which is a cloud-based targeting-platform it did not take long, until ideas came to my mind which were connected to my business. Because running a cloud-service for business clients has some issues. The cloud makes things disappear for the client – he used to have servers and noisy equipment in dedicated rooms and now everything is quiet because these services have been moved to the cloud. Sometimes you even wouldn’t now in which country or region of the world your machines are running. This is even more the case with a service like the one my company is providing – we are offering specific algorithms and a high-scalable computing platform to be plugged into our clients Ad-Servers and their Content-Management Systems – which are also in the cloud usually. So it’s a cloud-service which connects to other services (machines) in the cloud. Beautiful. But also a pain if your business still relies on a real person running these services and paying your bills etc.
Visibility and making things tangible is quite an issue when it comes to cloud-services. Therefore i thought about that and came to the conclusion that the Raspberry could add something to it. So i worked on a thing i called the “targetometer”, which is basically a Pi connected with a LCD-Display and some LEDs to retrieve statistics from our cloud service and display them to the client in a very physical and also entertaining way. The idea literally was to get back on the desk of the client with our service. This is how we hope to get there:
But we are not a hardware-company… As i developed my idea further, started discussions with colleagues, pitched it to our marketing folks and the data science department it began to live and grow in my hands, i mean really… And then – after presenting the first prototype to our clients and speaking at a conference about it there came this point of no return: we would be shipping these devices to our clients!
Luckily we were so wise to not make any promises when this would be. But a certain pressure once you started telling about it is hard to deny.
Although we are only thinking about a small number of devices in total – not more than 100 in the end – i had to remind myself what the guys from Highway1 were telling about the surprising difficulties for people who are getting into hardware for the first time (at things.con in Berlin). It’s physical. You have to buy tons of material, find people to assemble it, ship it to them, discuss optimizations of the different cases (we decided to go for three different cases, one out of paper, another from wood and one made out of metal…), talk back to the manufacturers, think about packaging, security and try to anticipate how these devices might be used, abused and probably also hacked…
It’s really an amazing experience (and we have just started, no device has been shipped yet…) because we software guys, after ages of web-development and app-crazyness tend to underestimate what it means to do something in the real-world i guess. But it’s also extremely satisfying to hold something in your hands that has python-code and json-objects inside and talks to the web but is blinking like a real thing…
Another physical experience from today… http://twitpic.com/e70pwj
Pingback: Raspberry with LCD Display | Making connected stuff
Pingback: Building a cellular tracker: Part0 – Why future IoT Devices might run on GSM and not Wifi | Making connected stuff