The Knowledge Cloud: Evolving the data management paradigm

In the state of the art, software programs decide how data is (or may be) stored. The program's code determines in which file formats, which database management systems or which cloud services are supported. Sharing data between programs or devices is difficult: files are inadequate (too low-level); cloud services tend to be program-specific or interoperate badly with competing services. Consequentially, the user's data is splintered across various storage mechanisms. Programs are unnecessarily complex; data is badly interoperable; users are not in control of their data; developers must write inessential code. Our current ‘data management paradigm’ causes these problems by promoting a strong coupling between programs and storage. We propose a new data management paradigm to solve these problems. Our new data management paradigm separates programs and storage by a central system for information provision and exchange: the knowledge cloud. Programs are no longer concerned with storage; they exchange information with the knowledge cloud in the same information model the program uses internally—we call this quality storage-agnosticism. The user can configure the knowledge cloud, oblivious to the programs, in order to control storage and processing of the user's data—we call this quality context-adaptivity.