“What does my chair do
when I am not in the room” –
This questions was asked
by the philosopher Ernst Bloch
more than 40 years ago.
Today we have the answer: It connects to the internet!
Digitization – once again – has brought on a fundamental change: Things are starting to communicate, even without asking us! It is a development that is already underway – expectations for 2020 are 20 – 50 billion “online” devices (according to the European research cluster CERP-IoT). By then, there will be 30 times more machine-to-machine sessions than person-to-person contacts. The refrigerator will order independently, the house will control the heat autonomously before the inhabitants come home, and the car will automatically transmit data to the service center. Even forests will post their condition independently, and will autonomously take action when they are plagued by the bark beetle.
Sensor and communication technologies (e.g., NFC) are often at the center of discussion, but the stakes are a lot higher – including changes to the Internet. The following is a selection of current research priorities (based on the Strategic Research Roadmap of the European Union):
- Identification technologies – a unique identity (UID) is necessary, e.g. IPv6, for which new technologies need to be developed or existing ones adapted.
- Architecture of the IoT – semantic interoperability is required, which calls for “semantic web based frameworks and semantic mediators.” Services will play a key role, providing functionality independent of heterogeneous hardware or software.
- Communication technologies and protocols (CoAP)
- Software and algorithms (procedures for self-organization, self-adapting software, etc.)
- Security technologies
- Standardization to cope with the variety of devices
Fear of the Things
At least two aspects of this scenario generate fear. For one, “the things” will have more information at their disposal than we do, they will “know more.” Cars will communicate with the vehicles ahead, thereby “seeing around the corner.” And thanks to “big data machines,” devices will be able to accurately predict our behavior.
Second, the things will start deciding autonomously, in the process developing a life of their own and gaining a decision-making capacity, and with it “power.” But where does this power end? Should a machine be able to “impose its will” over us, even if this seems to be in our best interest?
Software Testing: Ramp it up or re-think it?
We believe the answer is both. Traditional testing methods that assure functionality, reliability and performance will need to be intensified along with security testing, while unit testing will require more attention to detail. Given the large variety of components, layers and interfaces, integration testing will grow significantly more important.
However, verifiability, responsibility and autonomy open up completely new areas and pose new questions. These social and ethical issues cannot be discussed uncoupled from specific applications, for they need the collective thinking of engineers and technicians, experts from other fields and social scientists. Software testing can provide the platform – based on technical and professional expertise (the core business of testing), but also keeping in mind the multiple dimensions.