Generation of Reliable Randomness via Social Phenomena

Randomness is a hot topic in computer science due to its important implications such as cryptography, gambling, hashing algorithms and so on. Due to the implicit determinism of computer systems, randomness can only be simulated. In order to generate reliable random sequences, IT systems have to rely on hardware random number generators. Unfortunately, these devices are not always affordable and suitable in all the circumstances (e.g., personal use, data-intensive systems, mobile devices, etc.). Human-computer interaction (HCI) has recently become bidirectional: computers help human beings in carrying out their issues and human beings support computers in hard tasks. Following this trend, we introduce RANDOMDB, a database system that is able to generate reliable randomness from social phenomena. RANDOMDB extracts data from social networks to answer random queries in a flexible way. We prototyped RANDOMDB and we conducted some experiments in order to show the effectiveness and the advantages of the system.