> I think you're not using Facebook appropriately. I respectfully disagree. The dissociated graphs and/or the wrong usage of social network platforms are minor

> I think you're not using Facebook appropriately. I respectfully disagree. The dissociated graphs and/or the wrong usage of social network platforms are minor issues. They are issues, but not the cause of the matter. The wrongfully usage of graph theory is the main issue that (most social networks) should address. The first usage of the sociogram (at least the first I am aware of) is in the psychosociological studies of Iacob Levy. From there the graph theory has seen a huge improvement as a very powerful analytical tool. Using this tool to automatise sociality is [clumsily] easy, but [refinedly] hard: we are basically using a sophisticated querying tool to filter, predict, rate future interest and to ease future communications. And this is not wrong, unless you add strong requirements to the problem: free flow of information but avoiding hype-driven sharing, freedom of connection but avoiding the filter bubble and self-radicalisation, etc. Apparently many are starting to think that what's probably causing disconnected graphs, inappropriate usage, fake news, revenge porn et all is "feeding users with feedbacked data extracted from big data analysis over graphs and metadata". I wonder if anyone has ever bothered reading "Psychodrama & Group Psychotherapy", that highlights how doing this is inserting psychopathic traits in a community. Solving this matter means seeking a solution to precautionary principles, and is probably something that requires strong AIs. Since we don't have strong AIs, in the meanwhile, unfriending everyone is a simple and elegant solution. chapeaux