As of the end of this week, I completed chairing the 30th annual international conference of the Society for Chaos Theory in Psychology and Life Sciences, with 54 participants from around the world. It basically went well, and it was kind of cool to make introductory remarks at 8 AM during EDT, with somebody on at 6 AM their time in Montana and someone else on at 10 PM their time in Sydney, Australia. It can be done, and even with parallel sessions happening.
Of course there were the usual snafus of people getting bad internet connections and disappearing or becoming mute while presenting, which does not happen in live sessions. There were also some people who failed to present due to not being able to properly load or manage their slides or videos, although I have seen problems with this sort of thing even in live conferences.
Something I throw out there for anybody managing one of these involves how we managed the parallel sessions. So we had both a co-host/moderator, who managed entry to a session, as well as a session chair who managed timing by speakers, with on this following the old incentive-compatible strategy of usually having that be the final speaker in the session, giving them incentive to keep the earlier presenter in line on timing. Indeed, in our wrapup session someone noted, accurately near as I could tell, that there may have been better adherence by speakers to time limits in this format than is often the case in live sessions.