As one who had the great fortune to be part of networking in the UK in the 70s and 80s, including a stint in UCL-CS working at the end of the SATNET/ARPANET years (UCL was the uk end of the ARPAnet of the day, with an IMP, and then a BBN butterfly, and a large tranche of addresses held I believe in conjunction with the UK signals directorate of the Ministry of Defence) -I can tell you its a bit from box A and a bit from box B.
It worked better than a lot of us (X.25!) thought.
It worked a lot better than the code a lot of us were working on (the OSI/GOSIP model)
USENET was both worse and better than online communities.
The 1984 RAND report on the social behaviours in EMail remain true to this day, discussing flaming and trolling behaviours.
This has to invoke Betteridges law of headlines.
As one who had the great fortune to be part of networking in the UK in the 70s and 80s, including a stint in UCL-CS working at the end of the SATNET/ARPANET years (UCL was the uk end of the ARPAnet of the day, with an IMP, and then a BBN butterfly, and a large tranche of addresses held I believe in conjunction with the UK signals directorate of the Ministry of Defence) -I can tell you its a bit from box A and a bit from box B.
It worked better than a lot of us (X.25!) thought.
It worked a lot better than the code a lot of us were working on (the OSI/GOSIP model)
USENET was both worse and better than online communities.
The 1984 RAND report on the social behaviours in EMail remain true to this day, discussing flaming and trolling behaviours.
https://datatracker.ietf.org/doc/html/draft-irtf-hrpc-associ... is an indication of how important the questions have become.
Also see https://reasonablypolymorphic.com/blog/monotonous-web/index.... which is at https://news.ycombinator.com/item?id=32348519