Information system
An information system is used to retrieve information and process business transactions. Users in different roles work collaboratively on an overall result. The system takes care of the storage, input and output, validation and processing of the data for the users. The aggregation of data into reports also occurs.
Information systems are often used in a company’s intranet. The users in an intranet usually are expected to have a behaviour that is predictable (e.g. with respect to scalability) and regular within certain limits (not a “wilderness” like on the Internet).
Often rather complex business processes are important, which can make it challenging to properly authorise users for the business processes.
Information systems can also be web-based, but they are not a “web system” (see the next section for a more precise definition).
Web system
A web system basically serves the same purposes as an information system. The difference lies in the target group and the type of use. A web system is intended for use on the Internet (the “wilderness”). The Internet is an environment in which large numbers of users, heavily fluctuating loads and occasionally malicious attackers can be expected.
Due to the use of the Internet, higher network latencies are often to be expected than with in-house applications, which can have a negative effect on the system response time behaviour and thus (also) its usability.
Attention must be paid to these special circumstances when designing a web system (in addition to the business logic itself).
Embedded system
An embedded system is directly connected to its environment through sensors and can also influence the environment through actuators. The tasks of an embedded system are often measurement and control of processes in its environment. In the case of information or web systems, on the other hand, the environment consists exclusively of human users or other IT systems. In contrast to IT and web systems, resources (e.g., memory, energy, computing time) in embedded systems are limited and architectures for those systems have to consider this. Often, latencies are important, that is, the reaction time to external events and their handling. Special cases are embedded real-time systems in which timely task processing is essential, as well as safety-critical systems.
