Proposing more ecologically-valid experiment protocol using YouTube platform

Video streaming is becoming increasingly popular, and with platforms like YouTube, users do not watch the video passively but seek, pause, read comments. The popularity of services possible due to development compression quality prediction algorithms. However, those algorithms are developed based on classic experiments, which non-ecologically valid. Therefore, experiments mimic real user interaction. Further depends results coming from ecologically-valid experiments. we aim propose such Nevertheless, proposing a new experimental protocol difficult, especially when there no limitation content selection control video. freedom makes data analysis more challenging. In this paper, present an in subject assessed while freely using YouTube. To achieve goal, Chrome extension that collects objective allows network manipulation. Our deep shows correlation between MOS objectively measured as resolution, proves test works. Moreover, have shown significant differences subjects, allowing for detailed understanding, how influences interaction service.