Apry

Apry at

mlinksva, putting the involuntary aspect of requiring (or defaulting to) symmetric relationships in a FDSW aside for a bit. Assuming that would skew the distribution to be less top-heavy, there's the issue of how that would affect behavior. Were it (practically, b/c of client conspiracy) compulsory to be getting (probably a sampling of) audience feedback, would that significantly change the performer/audience relationship between hub people and "followers"? No first hand experience, but I can't imagine the best case would be substantially different than stand-up comedian and live crowd. But perhaps I'm missing your point altogether.

FWIW I know of an early ISP in the US which was started by highly technical people who wrote a lot of new software to offer services, but is probably an outlier even in my non-representative anecdotal dataset. However, I think my argument for portals in the mid to late 90s applies a forteriori to ISPs. So, to the degree that it's convincing (and it probably isn't all that convincing), it should supercede any business-centric interpretations.