> or at least say how confident they are in what they know
This I can very much underwrite. Error bars or rough confidence indicators are missing far too often, also from sites reporting on e.g. benchmark values of hardware they've been testing... such professional organisations yet such basic omissions
