I agree in particular with 2. I’d have liked to see some modelling that could be criticised. I don’t think I want more processes here because those will probably clog things down in future, but I’d like to understand what caused decision making to be be hidden and hard to correct here. Were the CEA team understaffed/under-supported?
I’ll add that afterparties were unmentioned but clearly higher risk than the events in general and very predictable that they would take place. Some thought/modelling here would have been nice too.
Likewise it feels that if we can’t deal with risky(?) events well as a community, who can? And if we can’t do things like this well, I am concerned about our decision making around complex risk in general.
Thanks, Nathan! It was great to see you at EA Global.
You’re right that we didn’t share the specifics of our reasoning—I think that might have been the right call at the time given the tradeoffs of spending time on public communications vs. object-level preparation for the event, but I can understand how community members felt in the dark and we regret that.
To make explicit some of our reasoning: we think that the cost from COVID of the event was likely quite small relative to other costs of the conference. A back-of-the-envelope calculation from one of our COVID advisory board members suggests that the total counterfactually-adjusted COVID cost was ~1% of the other costs of the event (primarily attendee time, but also financial costs). We think that doubling the number of attendees may have grown the fraction of all costs attributed to COVID from 1% to something more like 2% — so while higher, it was still not one of the biggest costs of the event as a whole. So I want to acknowledge that, while we considered this aspect of the event and took precautions to limit the spread of COVID, we genuinely didn’t view COVID risk as a major cost to attendees: we probably spent more time worrying about whether the logistics would work. We stand behind this prioritization.
(The BOTEC was done after we had decided to expand the event, but when we consulted our advisory board, community leaders, and our own judgment of the benefits and risks before the expansion, it felt very clear that the benefits would exceed the costs by a large margin, to the extent that a detailed BOTEC didn’t seem like a good use of time. Our post-decision calculations seem to validate that view.)
Regarding staff capacity, we have been very stretched as a team this year. I fully came back from parental leave in mid-April. We ran EAG Reconnect in March, the EA Picnic in July, the EA Meta Coordination Forum in September, and EAG London in October. For much of that time, I was the only full-time staff member on the Events Team: we had one other part-time employee and two contractors. COVID-related precautions added (rough guess) 20-40% additional production time. We brought on two part-time contractors to help with admissions processing in the final weeks and brought on two additional contractors to assist with production for the week of the event (and, as usual, were helped by many wonderful volunteers during the event itself).
We knew we were stretched too thin and ran a hiring round, but the new hires had just started or not yet started at the time of EAG London (and running the round took capacity from the otherwise stretched team). I hope and expect this will pay off in 2022.
“Regarding staff capacity, we have been very stretched as a team this year. I fully came back from parental leave in mid-April. We ran EAG Reconnect in March, the EA Picnic in July, the EA Meta Coordination Forum in September, and EAG London in October. For much of that time, I was the only full-time staff member on the Events Team: we had one other part-time employee and two contractors. COVID-related precautions added (rough guess) 20-40% additional production time. We brought on two part-time contractors to help with admissions processing in the final weeks and brought on two additional contractors to assist with production for the week of the event (and, as usual, were helped by many wonderful volunteers during the event itself). “
I didn’t realise this. I’m impressed by how much you did with so little staff.
I agree in particular with 2. I’d have liked to see some modelling that could be criticised. I don’t think I want more processes here because those will probably clog things down in future, but I’d like to understand what caused decision making to be be hidden and hard to correct here. Were the CEA team understaffed/under-supported?
I’ll add that afterparties were unmentioned but clearly higher risk than the events in general and very predictable that they would take place. Some thought/modelling here would have been nice too.
Likewise it feels that if we can’t deal with risky(?) events well as a community, who can? And if we can’t do things like this well, I am concerned about our decision making around complex risk in general.
Thanks to the CEA team who ran a great event.
Thanks, Nathan! It was great to see you at EA Global.
You’re right that we didn’t share the specifics of our reasoning—I think that might have been the right call at the time given the tradeoffs of spending time on public communications vs. object-level preparation for the event, but I can understand how community members felt in the dark and we regret that.
To make explicit some of our reasoning: we think that the cost from COVID of the event was likely quite small relative to other costs of the conference. A back-of-the-envelope calculation from one of our COVID advisory board members suggests that the total counterfactually-adjusted COVID cost was ~1% of the other costs of the event (primarily attendee time, but also financial costs). We think that doubling the number of attendees may have grown the fraction of all costs attributed to COVID from 1% to something more like 2% — so while higher, it was still not one of the biggest costs of the event as a whole. So I want to acknowledge that, while we considered this aspect of the event and took precautions to limit the spread of COVID, we genuinely didn’t view COVID risk as a major cost to attendees: we probably spent more time worrying about whether the logistics would work. We stand behind this prioritization.
(The BOTEC was done after we had decided to expand the event, but when we consulted our advisory board, community leaders, and our own judgment of the benefits and risks before the expansion, it felt very clear that the benefits would exceed the costs by a large margin, to the extent that a detailed BOTEC didn’t seem like a good use of time. Our post-decision calculations seem to validate that view.)
Regarding staff capacity, we have been very stretched as a team this year. I fully came back from parental leave in mid-April. We ran EAG Reconnect in March, the EA Picnic in July, the EA Meta Coordination Forum in September, and EAG London in October. For much of that time, I was the only full-time staff member on the Events Team: we had one other part-time employee and two contractors. COVID-related precautions added (rough guess) 20-40% additional production time. We brought on two part-time contractors to help with admissions processing in the final weeks and brought on two additional contractors to assist with production for the week of the event (and, as usual, were helped by many wonderful volunteers during the event itself).
We knew we were stretched too thin and ran a hiring round, but the new hires had just started or not yet started at the time of EAG London (and running the round took capacity from the otherwise stretched team). I hope and expect this will pay off in 2022.
“Regarding staff capacity, we have been very stretched as a team this year. I fully came back from parental leave in mid-April. We ran EAG Reconnect in March, the EA Picnic in July, the EA Meta Coordination Forum in September, and EAG London in October. For much of that time, I was the only full-time staff member on the Events Team: we had one other part-time employee and two contractors. COVID-related precautions added (rough guess) 20-40% additional production time. We brought on two part-time contractors to help with admissions processing in the final weeks and brought on two additional contractors to assist with production for the week of the event (and, as usual, were helped by many wonderful volunteers during the event itself). “
I didn’t realise this. I’m impressed by how much you did with so little staff.
I agree, I didn’t realise (/appreciate) how understaffed the team was, and it’s amazing how well the event went considering that.
Thanks for the clear and thorough response Amy. It was great to see you also.