21

Review of EA Global 2016

My name is Amy Labenz, and I was Executive Producer and Curator of EA Global 2016. On behalf of the team, I’m posting to provide the EA community with information about our:

  1. Goals for EA Global 2016
  2. Decisions and lessons
  3. Evaluation of EA Global 2016
  4. Plans for EA Global 2017

Summary

What was it?  EA Global 2016 was the largest gathering of the effective altruism community to date. Just over 1,000 attendees joined us for a two and a half day conference at the University of California, Berkeley where we had 3 stages, 3 workshop rooms, and 5 discussion rooms. You can see the complete schedule here.

What were the goals?  We set three goals for EA Global 2016: (1) to create connections and facilitate coordination among people in the effective altruism community, (2) to create an opportunity for people to immerse themselves in EA concepts and effectively communicate complex ideas, and (3) to encourage the development of the EA movement as an intellectual community.

How did it go?  Overall, we think EA Global 2016 was a success. In particular, we were pleased with our ability to encourage new connections between our 1,000 attendees as well as the positive feedback that we received about the program, theme, and production. We did, however, make mistakes, and received some negative feedback as a result, particularly concerning persistent marketing emails, late announcement of the program and dates, late admissions decisions, and last-minute program changes.  

What will we change? Going forward, I will transition to managing a dedicated Events Team full-time and will start preparing for EA Global 2017 much earlier. We intend to engage more with the community throughout the planning process to get input on programming decisions and to help potential attendees evaluate the value of attending. In addition, as part of the CEA merger, we have reallocated staff and have begun making improvements for next year - for example, with the new eaglobal.org.

Goals for EA Global 2016

We selected three main goals, based on an analysis of how EA Global, EAGx, and related events have created value in the past:

  1. To create connections and facilitate coordination among people in the effective altruism community,
  2. To create an opportunity for people to immerse themselves in EA concepts and effectively communicate complex ideas, and
  3. To encourage the development of the EA movement as an intellectual community.

The first two goals were easy to agree upon, but we were more cautious in deciding the third. Goal #3 reflects our view that the EA movement should cultivate an atmosphere of intellectual modesty and self-skepticism. Our reasons for this view are: (1) our expectation that new high-priority cause areas will continue to emerge and that a community-wide focus on cause prioritization remains important, and (2) feedback from community members that previous events had been too focused on specific causes.

Decisions and Lessons

Our goals influenced our decision-making process, particularly on strategic questions.

Programming

To help build connections between attendees, the schedule had many long breaks to allow people to mingle, as well as receptions and an afterparty to facilitate networking. To promote immersive information exchange, we had a large number of workshops and discussion groups on particular topic areas. We recorded all of the talks on the two main stages so that attendees were able to participate in workshops and discussions and did not feel rushed to leave valuable conversations. In addition, our talks and panels investigated research challenges and potential breakthroughs rather than particular organizations. We selected topics to attract experts who could discuss issues such as how EAs can influence policy and how to navigate intellectual disagreements. Finally, we introduced the Effective Altruism Global Research Meeting, which included lightning talks and academic poster presentations to promote intellectual exchange and a collaborative atmosphere.

Our main programming missteps were: (1) we did not post the program on the website early enough and failed to adequately communicate schedule and programming changes at the event, (2) the large number of concurrent activities limited the number of shared experiences at the conference, (3) the workshop rooms were not large enough, so some people who were excited to participate in the most popular workshops were disappointed, and (4) a number of the recorded talks and panels had poor attendance, which was uncomfortable for the speakers. We may want to consider having less complexity in the future, with fewer stages or fewer talks and more workshops with larger workshop spaces. We should also better communicate which talks will be recorded, and may consider recording additional or more varied content. 

Production and Logistics

Based on feedback from previous EA Global events, we tried to improve the production and logistics for EA Global 2016 so that people could better focus on the content of the event. This year, we spent more time working on room layout design, signage to direct people, catering estimates to ensure that there would be sufficient food, and volunteer recruitment, role descriptions, and training so that we could anticipate and resolve issues. In general, we made large improvements over previous years, but we still had issues with poor on-site communication of schedule changes that resulted in confusion and frustration for attendees.

Date and Venue

We decided to hold EA Global in August because: (1) it is a convenient time for students to travel, (2) university facilities and housing are more readily available, and (3) vendors lower prices to compete for business during their off-season. Given our goal of fostering an intellectual community, we thought that these benefits outweighed the fact that some high-value attendees, such as Bay Area VCs, might be less likely to attend at this time. People were generally happy with the venue and reported that its physical layout was one of the things they would like to see again. This layout, which was a large improvement over the sprawling Google venue from 2015, was one of the other main considerations that led us to select UC Berkeley. In addition, working with the EAs of Berkeley allowed us to get large discounts. One challenge of working with the student organization was that it introduced major bureaucratic hurdles, particularly with finalizing contracts with the university and reserving rooms. This caused us to delay our announcement of the event for months, which probably prevented some people from attending, and resulted in disruptive room changes, such as the last-minute changes to the workshop rooms on Friday.

Diversity

We’d like EA Global to help further diversity within the movement by attracting both attendees and speakers from underrepresented groups, particularly in terms of nationality, race, and gender. To recruit speakers, we consulted with the CEA team and leaders in the community to create a master list, which we evaluated on a number of dimensions. We optimized for fit with the program, selected experts in fields that we believe to be high value, and prioritized speakers based on their familiarity with effective altruism, their name recognition, and our personal connection. As part of that process, we made specific efforts to invite people from traditionally underrepresented groups. In addition, when building the program we made sure there were no all-male panels (we did have one panel entirely composed of women - this was not by design, but we noticed it with time to make adjustments and decided not to). We made significant improvements to our process over EA Global 2015; however, we’re not satisfied with our progress here and would like to do better in the future. We hope that starting the speaker selection process earlier will give us more opportunity to find speakers who better represent the diversity that we’d like to see in the movement.

Speakers

Overall, we did not optimize for attracting “big name” speakers, though we did invite some in that category. We found in the past that headline-making speakers can result in inaccurate perceptions outside of the community. For example, at EA Global 2015, Elon Musk participated in an AI side event, but because of his status and the team’s decision to feature him prominently in some event marketing, he was seen as the face of EA Global 2015. This contributed heavily to the impression that EA Global 2105 was focussed on technology and AI risk, even though this was not the intention of the organizing team. With more lead time and careful presentation of featured speakers and topics, we think it will be possible to select for “big name” speakers that also fit well with conference themes. This year, a number of speakers in this category were interested but had scheduling conflicts. Next year we will start the invitation process earlier. If you would like to nominate a speaker, please do so here.

This year we had ~80 speakers. Alison Woodman, our Speaker Liaison Team Leader, managed speaker communications before the event and helped train volunteers on the Speaker Liaison Team.  Each speaker had a Speaker Liaison to escort them to their stages and to make sure they were comfortable and understood the event layout. We also hosted a speaker reception where we connected speakers with one another and with EA organizations, and we expect to see friendships and collaborations as a result. Generally, our speaker interactions went well, and we have received positive feedback from many speakers. Unfortunately, we had communication delays, particularly about panel descriptions and lightning talk acceptances, which were likely stressful for the speakers. This was largely caused by a process mistake where we had only one employee drafting the descriptions and making decisions about lightning talks. Once we invited additional teammates to approve talk applications and write panel descriptions, things went more smoothly, but because of delays in making that call we probably appeared less professional than we would have liked.

Admissions and Marketing

We have an admissions process because we want to select for people that will get the most value out of the conference and who will contribute to the positive experiences of other attendees. We look for altruistic people who are curious and have an analytic mindset and try to screen out people who have histories of disruptive behavior. This year we received 2,152 applications for a target conference size of 1,000 people, which we successfully achieved. The application process, however, did have several problems: (1) conversion on our applications was lower than expected, and as a result, we did a push for additional applications that was seen as overly persistent or misleading (See Kerry Vaughan’s Marketing post), (2) spelling mistakes made some of the pre-event communications appear unprofessional, (3) some questions in the application may have been off-putting or discouraging, and (4) promotions, such as allowing some attendees to bring a friend, created inconsistent admissions standards. We plan to address these issues by opening applications earlier and making more use of waitlisting next year. This will allow for reconsideration of those who expressed early interest and will give us more time to build promotions that are consistent with our admissions standards. We also plan to have a focus group led by our Community Director, Julia Wise, in order to check the application for tone. Julia will also review the pre-event communications and marketing plan to make sure they do not clash with the intellectual and professional goals of the event.

Location

Our biggest considerations when determining where to hold EA Global 2016 were the convenience for and cost to the community. We considered (1) where most EAs that would like to attend are located (based on the map of Effective Altruists on the EA Hub, and the results of the most recent EA survey), (2) the strength of the existing communities in the various locations, (3) the cost of flights for the EA community, and (4) the location of EA organizations. Based on our analysis, we decided on the Bay Area. Our delayed release of the location likely prevented a number of Europe-based EAs from making the trip, and we plan to make a decision much earlier this year. Please let us know your location preferences here.

Food

Food was the largest single budget item for the conference; we spent approximately $100 / person for the weekend. We decided to have food on-site for the opening and closing receptions and for lunches because we wanted to create opportunities for interaction between conference goers. After discussions internally and on Facebook, we chose a vegetarian and vegan menu. We did this acknowledging that the community contains many who do not believe that personal diet changes are an effective way to reduce suffering, and also many who believe that a meatless menu is a basic gesture of respect to animal advocates. We included a restaurant guide in the program with the hope that nearby restaurants could provide further options if people found that our catering didn’t meet their needs. Food is always a touchy subject at conferences, and our survey results showed mixed responses on the meat question but trended positive in terms of overall food satisfaction.

Finances

CEA leadership set three high-level financial goals for the event: (1) use resources wisely, in keeping with our community values, while ensuring that attendees could focus on the content in a pleasant, stress-free environment, (2) keep tickets as affordable as possible while making scholarships available to applicants who could not have attended without assistance, and (3) roughly break even on the event. In the end, we were quite successful in achieving these goals. The total budget for the event was ~$250K (excluding the cost of staff time and travel). We estimate that CEA will spend $20K on costs not covered by ticket purchases and sponsorships (~$230K). We are still waiting on a full accounting from our bookkeeper and will provide an update when we have the final numbers. Here is a rough summary:

Note: financial aid and scholarships were our largest source of foregone revenue, however, we did not account for them as a “cost” in our spending overview. Students were given automatic discounts of $100 off, and everyone who applied for financial aid received some financial assistance. We heard substantial negative feedback about the posted ticket price, which suggests that we could have better communicated our financial aid policy. We instituted a pricing strategy where the ticket price was considerably higher than the price we expected the average attendee to pay, based on our estimates of the number of people who would need scholarships. It seems that the current pricing strategy might discourage value-aligned people who need financial assistance but are hesitant to ask for it. We plan to consider alternative pricing strategies and are open to suggestions. We will also consider adding more financial aid to our budget for next year.  

Evaluation of EA Global 2016

EA Global takes thousands of hours of EA time, so it is important that we earnestly assess how we did. We collected data from two surveys: (1) Closing Survey: 255 people responded to a 5-minute survey after the closing talk, (2) Follow-up Survey: 92 people responded to a survey that we sent a week after the event. The surveys were a mix of free-form responses and numerical ratings. (See the links for summaries and Oliver Habryka’s upcoming post for more detail).

On our Closing Survey, attendee satisfaction with the conference overall got the highest rating, with a score of 6.1 out of 7 (displayed as 2.1 below). Similarly, our Follow-up Survey results showed a 4.1 out of 5 overall satisfaction rating.

In our Follow-up Survey, we asked attendees to rate how much different parts of the event were issues for them. The mean was 1.5 on a 4 point scale, suggesting that none of these were terribly bothersome on average; however, of the issues listed, the clarity of the program, logistics communication, and the ticket price appear to have been the biggest concerns for attendees. This matches the feedback that we have received in-person and online.  

In our Closing Survey, we saw a significant impact as measured by the number of attendees who reported having changed their minds on particular topics. A large majority of people surveyed listed at least one area where they changed their minds (though this result may have been a consequence of the survey design).

In addition, in response to our Follow-up Survey, 14% of the ~90 respondents said that they changed their life plans significantly as a result of EA Global, 27% said that EA Global changed how much they intend to give to effective charities, and 5% of people reported that they took the Giving What We Can Pledge as a result of EA Global. We cannot simply extrapolate this proportion to all attendees of the event since there are likely selection effects at work (those responding to the Follow-up Survey may have been more moved by the event). It appears that 23 attendees signed the GWWC Pledge at EA Global or in the weeks after the event and 8 more signed up for Try Giving (only 3 cited the conference as how they first heard about GWWC, though this required a write-in answer, which may have resulted in underreporting).

We expect to see additional sources of value in the months to come, but based on attendee and EA organization feedback, as well as our own conversations with potential collaborators, potential hires, and new donors, we think that EA Global 2016 was a valuable use of CEA team time. On the margin, we might have saved time by outsourcing more of the work, or by limiting the complexity and/or size of the event. Given that we received feedback suggesting that people would like to see more facilitated interactions & networking and relatively fewer panels & interviews, CEA might be able to plan a version of EA Global with less curated content and more workshops, which would demand less staff time while better satisfying attendee preferences. 

The Communications and Outreach Team is currently working on improved methods for evaluating CEA projects against one another, but this is beyond the scope of this particular post.

Plans for EA Global 2017

For EA Global 2017, we will start the planning process earlier and invite more community participation throughout. We aim to start planning EA Global 2017 this month, with a goal of selecting a location and identifying keynote speakers in the next month, and working from there to select a date. If you are interested in contributing content or in having your organization represented at the event, please use our preliminary workshop submission form here.

To build on the success of EA Global 2016 and shore up our weak spots, I am transitioning to running the Events Team full-time, which will include EA Global, EAGx, and additional in-person outreach. The team will include Roxanne Heston, Julia Wise, and another teammate (description  here, apply here), and will receive support from the rest of the Communications and Outreach Team, led by Tara Mac Aulay. We will bring on a new teammate to manage CEA US Operations (description here, apply here). As a result of the new team structure, we have already started making improvements to our process for next year. For example, we have a new website, thanks to Sam Deere. With the new system, many staff members can edit the website and push changes quickly. This will allow us to share information about EA Global, including information about the program, speakers, and schedule, without having a single bottleneck.

Please visit the new website to see the videos and pictures from this year, and to get information about EAGx events and EA Global 2017. I appreciate all of the feedback that we have gotten so far and I look forward to EA Global 2017! If you have feedback but haven’t had a chance to share it yet, please do so in the comments or email us at hello@eaglobal.org.

Thank you!

Thank you to the attendees, speakers, staff, and volunteers who helped make EA Global 2016 possible. Special thanks to: Oliver Habryka (Director of Programming) and Julia Wise (Community Director) for help with this post. Thanks also to the rest of the EA Global team: Alison Woodman (Speaker Liaison Team Leader), Peter Buckley (Assistant Producer), Kerry Vaughan (Director of Marketing), Larissa Rowe (Marketing and Social Media), Tara Mac Aulay (COO of CEA), Will MacAskill (CEO of CEA), Anjali Gopal (Volunteer Team Leader), and our 50  volunteers. Also, thank you to our MCs: Nathan Labenz, Roxanne Heston, and Jonathan Courtney, and to Tessa Alexanian for helping me and others with our slides, Andrew Lapinski-Barker for creating the signs for the event, Katherine Xiang for designing the programs, Michelle Hutchinson for organizing the Research Meeting, and the many Pareto Fellows who joined in to help.

Comments (42)

Comment author: Ben_West  (EA Profile) 21 September 2016 06:22:17PM *  9 points [-]

Thanks for writing this up Amy! You and the EA Global team did an amazing job.

One comment/question on location and finance, sparked by Luke's comment.

My girlfriend (Gina) organized a one day vegan Fest with 4,000 people here in Madison Wisconsin, for which the venue (including AV) was around $5,000. The event was one third as long as EA Global, but there were three times as many people; eyeballing your graph above and assuming those factors cancel out would imply that events in the Bay Area cost eight times more. Let's just be conservative and say it's a 5X increase, which would change your $150,000 food and venue budget to $30,000 for a savings of $120,000.

If 10% of the conference is people who lived in the Bay Area and wouldn't otherwise come to EA Global, you could buy them round trip plane tickets to come to Madison (currently around $300 each) and still have a savings of $90,000. (You could actually buy them all first class tickets, and still save money.) And of course lodging would similarly be a lot cheaper etc.

I don't think Madison is particularly unusual – I would guess that most medium-sized cities have venues which can hold a couple thousand people for similar prices.

The EA Global team did an amazing job: if you hadn't put in a lot of work doing things like partnering with the student group the price difference would probably be way more. But there's only so much you can do while still being in the bay – have you considered a lower cost location?

Comment author: AmyLabenz 22 September 2016 02:04:00PM 4 points [-]

Thanks for the feedback! We'd absolutely be open to holding the event in a low-cost location if community members indicate that they are willing to make the trip and it makes sense in terms of expected savings (I have added this to the survey form). Based on my experience running events in various locations -- NYC, Bay Area, SLC, and Michigan -- the 5X estimate seems high, but there is definitely a meaningful difference and this could be a good idea. I live in Detroit, Michigan, and I expect that with my connections in the local community I would be able to get access to venues and catering at reduced rates. Aside from the fear that EAs would not make the trip, my biggest concern with having an event here, or in another medium-sized city with a relatively low number of EAs, is the cost to the EA community in terms of lost time spent traveling. Oliver Habryka has done Fermi estimates on the costs and it seems that even with conservative estimates of the value of each individual’s time, the costs likely outweigh the potential savings.

Comment author: rohinmshah  (EA Profile) 22 September 2016 08:23:44PM *  1 point [-]

10% of attendees = 100 people

Time spent traveling ~ 20 hours per person (Seems a bit high)

Value of time: $30 per hour (this seems high to me, though I know others claim $50-100. Either way, your conservative estimate shouldn't be higher than $30, I would think.)

Total: $60,000

It still seems like you'd be saving money according to Ben's calculation.

(It's definitely possible though that the time costs would discourage people from attending, resulting in less applications/attendees.)

Comment author: Habryka 22 September 2016 10:53:31PM 1 point [-]

I would expect that if we were to hold the event in Madison or a similar location, that a much larger percentage of the attendees than 10% would have to travel. My rough guess based on the distribution EAs around the globe, is that if we would want to get a good share of the community to attend, then at least 50% of attendees would have to travel to Madison.

My usual estimates in time costs usually ran into something around the $200,000 - $400,000 range for having the event in a more remote location, usually with a fairly low bound on how much people value their time (i.e. $20 or so), which would fit well with your estimate, when you increase the percentage of attendees who would have to travel.

I would also expect in practice that because of the heavy-tail distributed nature of income in the community (and the population at large) that the actual value of the average person's time would be a good bit higher than the value of the median person, which is where my $20 intuition would come from. So I am not that sure whether $30 is actually high. My guess is that it would still be an underestimate, though I wouldn't be confident. (E.g. if Dustin Moskovitz wants to attend, he is pretty justified in valuing his time at something on the order of $1,000 - $10,000 an hour, and weaker version of this are true for many other high-earning members of the community.)

I do think that ~20 hours in travel costs seems a bit high, but something on the order of ~10 hours is probably correct.

Comment author: Denkenberger 23 September 2016 01:05:47AM 2 points [-]

But remember nearly everyone who does not live in the Bay Area would have to travel a shorter distance. What about Denver, which is low-cost location and also low-cost flights?

Comment author: Julia_Wise 23 September 2016 02:20:08PM 4 points [-]

One thing I noticed while trying to find a city for a meeting of a group of people scattered across the US and UK: smaller cities are more expensive and more time-consuming to fly to from outside the country. For example, there are no nonstop flights from Berlin to Denver. Even places that are airport hubs within the US, like Atlanta, are not particularly good destinations if you're coming internationally.

It's much easier to find cheap, nonstop flights between major cities. Not that this is a dealbreaker, but it is something we consider.

Comment author: Jeff_Kaufman 28 September 2016 01:48:55PM *  2 points [-]

Boston does pretty well on the metrics of cost-to-fly and ease-of-flying from various places in the US and Europe.

(Not that I'm biased or anything...)

Comment author: rohinmshah  (EA Profile) 07 October 2016 12:15:29PM 0 points [-]

Sorry, totally missed the response here.

The question isn't "How many people have to travel", but "How many people have to travel who otherwise would not have to", which basically reduces to "How many people are coming from the Bay Area". I'll admit that I don't really know that number, but it seems implausible that it's 50%.

(I could believe that it's 50% for the last EA Global, but I expect that a lot of people showed up because it was so close to them, and so we'd get less of those people but more of other people who live closer to the chosen venue.)

In addition, as Denkenberger said, any participants from the US East Coast or Europe would travel less.

Good point about the heavy-tail distribution for time value, I hadn't thought about that. I agree that taking that into account $30 may not actually be too high. (Although travel time is not completely wasted, you can work along the way.)

Comment author: Ben_West  (EA Profile) 22 September 2016 02:56:21PM 1 point [-]

Interesting, I hadn't thought about the cost of time spent traveling. Thanks!

Comment author: lukeprog 20 September 2016 08:17:07PM 9 points [-]

In addition to making available scholarships clearer, I wonder whether there's anything that can be done to convince prospective attendees of how incredibly expensive it is to run this kind of event (thus necessitating "high" ticket prices). My sense is that people just do not understand how expensive it is to get acceptable venues and acceptable catered food.

Comment author: Robert_Wiblin 21 September 2016 02:33:50PM 3 points [-]

You'd think just pointing out that the goal is to break even would work, but it seems people don't find that fully persuasive! Maybe they assume the tickets look expensive because you're a spendthrift.

Comment author: Gregory_Lewis 21 September 2016 07:05:18PM 5 points [-]

I'm not sure many people were put off by, "This looks really expensive, I'm sure the conference team are wasting my money somehow", but perhaps more simply by, "This looks expensive, and I don't think the ROI is going to make it worth it for me".

Comment author: Ben_West  (EA Profile) 21 September 2016 05:12:12PM *  3 points [-]

My sense is that people just do not understand how expensive it is to get acceptable venues and acceptable catered food.

… in the Bay Area.

Maybe some people are just confused about what it takes to run an event, but I thought complaints meant something more like "an equivalent venue would cost a fraction as much back home".

Edit: some back of the napkin calculations make it seem like the price difference is around 5X. Let me know if I'm missing something!

Comment author: MichaelDickens  (EA Profile) 24 September 2016 03:18:36PM 0 points [-]

Why is it important to convince people that conferences are expensive? The real question for prospective attendees is not, are the organizers charging a reasonable price? but, is this worth the cost for me?

Comment author: BenMillwood  (EA Profile) 27 September 2016 04:40:54PM 2 points [-]

That would be true, except:

  • we may care about "fairness" of price-setting, or other non-economic motivations,
  • we may expect to be able to affect the price by objecting to it,
  • we may use reasonable pricing as a proxy for other forms of reasonableness.
Comment author: IanDavidMoss 20 September 2016 06:09:25PM *  9 points [-]

Great summary! Regarding location selection, one thing to keep in mind is that selecting based on existing concentrations of interest is likely to have self-reinforcing effects. With any conference, a significant percentage of the attendees are likely to come from the local area, and it stands to reason that those attendees are, on average, likely to have a looser or more casual connection to the community than those who have traveled from far away. That's a big reason why many industry and association conferences choose to alternate time zones, or continents, every year. Otherwise, you might risk unintentionally fostering a strong regional bias to community growth over time.

Comment author: AmyLabenz 20 September 2016 07:05:21PM 2 points [-]

Thanks! This is an important point that we are also considering.

When I produced the Singularity Summits we alternated between NY and SF. For EA Global, it is hard to say what the right decision is. We hope that having a number EAGx events in different cities will help build communities outside of our most heavily concentrated areas. At the same time, we want to draw from different pools for the large gathering as well. We plan to have a number of team discussions about this question over the next month, and also plan to consider feedback from the survey form.

Comment author: Robert_Wiblin 21 September 2016 02:32:52PM 8 points [-]

My guess is that after 4 years of the main event being in the Bay, it's worth mixing it up and moving it to either Boston, NY or the UK next year. Then someone can organise an unusually large EAGx in the Bay.

Comment author: Evan_Gaensbauer 22 September 2016 04:29:34AM 1 point [-]

Boston and the UK have the biggest EA communities outside the Bay Area, so this seems eminently sensible.

Comment author: AviN 22 September 2016 04:00:15PM 0 points [-]

+1 for NY! :)

Comment author: AGB 22 September 2016 07:44:52PM 4 points [-]

Some scattered thoughts:

  1. In a different comment Rob suggests mixing up the location after many years of being in the Bay Area, which is an argument I'm pretty sympathetic to. Similarly I wonder if it's worth mixing up the time. EA is so student-heavy that the argument for doing it in the summer holidays is pretty strong, but I wonder if we could at least do early July as opposed to early August.

  2. "Given that we received feedback suggesting that people would like to see more facilitated interactions & networking and relatively fewer panels & interviews." I mentioned this before the event and it's been a near-universal observation in what I've seen written about the event, so I'm glad this has been taken on board.

  3. "where most EAs that would like to attend are located (based on the map of Effective Altruists on the EA Hub, and the results of the most recent EA survey)." I was surprised to read this. I don't think these two are very representative at all, and I thought that w.r.t. the survey in particular this was common knowledge. For alternatives though, did you/do you have information for where the 2015/16 applicants are actually from? Especially for 2015 where the events were scattered around the globe, that would seem like a decent improvement to get a geographic map of where potential attendees live. Something like the location breakdown of who visits this forum (which I've seen posted somewhere before and could find again if necessary I expect) also seems preferable.

Full disclosure: I live in London, didn't attend EA Global this year because of the distance, and weighted my votes for next year towards European locations followed by the US East Coast.

Comment author: Habryka 22 September 2016 11:08:18PM *  2 points [-]

Regarding 3:

The EA Hub, the EA survey, the traffic numbers for the forum and the location of EAG attendees,as well as most other survey data we have all tend to agree quite well on the distribution of the EA community. They all look roughly like the EA Hub map:

https://eahub.org/map

For reference, here is the distribution of people who answered the EA survey (conditional on people who filled out the whole thing and gave additional information):

https://awesome-table.com/-K7ENuzdTLgyTnE0kf2r/view

For reference, here is the distribution of traffic to the Forum:

https://drive.google.com/file/d/0B09ZIi0x-2JtM2h3Tlo1MmhhNG8/view?usp=sharing

And here is the distribution by country:

https://drive.google.com/file/d/0B09ZIi0x-2JtanMybDBpNjZfREE/view?usp=sharing

It would take me a while to make the origins of the participants for EAG 2015 into a nice map, but it generally follows a similar distribution, with the East Coast being naturally somewhat underrepresented (since we didn't have an event there).

In general, San Francisco is the biggest hub, the East Coast has a good amount of people but is quite spread out, and London+Oxford is about half the size of the Bay Area, with a good amount of people spread around the UK. Usually London + Berlin still is only about 50% - 60% of the size of the Bay Area. (For the Google Analytics data above, make sure to add up Oakland, Berkeley and San Francisco to get an accurate number for the Bay Area, and probably add up Cambridge, Oxford and London to get a somewhat similar comparison for the London area).

Comment author: Richard_Batty 23 September 2016 10:13:14PM 3 points [-]

In addition to AGB's point about the forum data, the EA Hub map in its default zoom state shows 746 in Europe, 669 in Eastern US, and 460 in Western US.

For the EA survey in its default zoom state, you get 298 in Europe, 377 in Eastern US, and 289 in Western US.

Comment author: AGB 23 September 2016 08:09:40PM *  2 points [-]

"In general, San Francisco is the biggest hub, the East Coast has a good amount of people but is quite spread out, and London+Oxford is about half the size of the Bay Area, with a good amount of people spread around the UK. Usually London + Berlin still is only about 50% - 60% of the size of the Bay Area. (For the Google Analytics data above, make sure to add up Oakland, Berkeley and San Francisco to get an accurate number for the Bay Area, and probably add up Cambridge, Oxford and London to get a somewhat similar comparison for the London area)."

Um, Am I missing something? If I add London + Oxford + Cambridge in your EA forum data, that's actually >100% the three Bay Area locations put together, not 50%.

It's much less obvious there, but I think your distribution by country suggests something similar, given that a decent proportion of the overall US visitors are presumably areas other than California, whereas I would expect the majority of the UK visitors to be clustered close to London/Cambridge/Oxford.

Comment author: Habryka 24 September 2016 01:22:40AM *  0 points [-]

Ah, sorry for that. I was a bit unclear in what I wanted to express with the above, sorry for being confusing:

Here are the two separate things I wanted to say:

London itself is about as closely connected via public transport, group houses, people visiting each other and just physical density of people as the East Bay and SF are. Let's call this group the "core Bay Area", and the other part the "core London Area". It takes about 30-45 minutes to get from any point in the East Bay to any location in SF, and similarly it takes about 30-45 minutes to get from any point in London to any other point in London and both cost about $10.

Then the London core area, which in the EA Forum statistics is just London, has 2118 sessions in the period from July 1st to now. The core Bay Area, which consists in the statistic above of San Francisco, Berkeley and Oakland, has 3776 sessions in the same time, making the London core Area about 56% the size of the core Bay Area.

The wider Bay Area, including the South Bay, San Jose, etc. are about as closely connected as Cambridge, London and Oxford are. I.e. it takes about 2 hours to get from any point in the space to another, and it costs something around $30-$50 to do so. Then the total amount of sessions from the wider Bay Area in the same period is 4771, and 4144 in the wider London area, making the wider London Area about 86% the size of the wider Bay Area.

I did an analysis on the traffic data on that a while ago, and forgot to make this distinction clear.

(To quickly check the thing above I went through the top 50 entries for the Google Analytics account sorted via city, and added all of the ones that are close to London and all of the ones that are close to SF to this spreadsheet, together with the number of sessions. I did not find any entries of cities on that list that were not Oxford, London or Cambridge that had any significant amount of people and were comparatively close to London, though I might have missed one or two with <50 sessions, since I am not as familiar with the British city names. Here is the spreadsheet with my numbers:

https://docs.google.com/spreadsheets/d/1oAMqV67sJ3UEmIg4kw1gNgu95wtxODGg8FIyr_jglCo/edit?usp=sharing)

Comment author: AGB 24 September 2016 12:05:04PM *  3 points [-]

You included Los Angeles and San Diego in your 'wider Bay Area' sum? They're hundreds of miles away, so if you include those I want to start including West England/North England/Ireland/Belgium/Netherlands/France/West Germany/Switzerland in the 'wider London area' sum.

I don't know if there's any way for me to look through the Analytics data myself;, but even if I use the EA hub (which I'd expect to be Anglophone-skewed and so under-represent much of Western Europe), that would double the number of people in the 'wider London area' (see picture).

Maybe there are factors other than distance I'm overlooking though; does it really take 2 hours and $30-$50 to get from LA/SD to SF? Because I wouldn't have expected that and yeah, even Belgium -> London is going to be significantly longer or more expensive.

Edit: Someone else pointed out to me that using analytics data over a period including EAG itself should skew towards SF just because EAG was in SF during that period, i.e. there was a huge concentration of EAs there that there wouldn't have been without the conference. I don't have much opinion on how strong this effect is, but it seems so easy to eliminate by looking at a different date range that I thought I may as well mention it.

https://drive.google.com/file/d/0B35Cz9EPOFE_bkswb255aGkyWFE/view?usp=sharing

Comment author: MichaelDickens  (EA Profile) 24 September 2016 03:28:28PM 1 point [-]

does it really take 2 hours and $30-$50 to get from LA/SD to SF?

By plane, it takes about 3 hours (1 hour flight, 1 hour at the airport, ~1 hour driving to/from airports) and costs $50-100. By car, it takes 6-8 hours and costs about $100.*

*$50 for gas at around $4 per gallon; I read a while ago that car maintenance costs about as much as gas, so double this to get the full cost.

Note for non-Americans: California is almost twice the size of the UK. The distance from LA to SF roughly equals the distance from Oxford to Paris.

Comment author: Habryka 24 September 2016 09:35:57PM 0 points [-]

Yeah, LA and San Diego are probably a bit farther away than the other cities. I would be happy with a comparison that removes them.

Though removing them from the comparison doesn't change too much. The general point I was trying to make was more that the highest density areas of the two locations, in which frequent travel is actually feasible, aren't of equal size, though their wider areas are indeed quite comparable. And that there is a really big difference in a 2 hour drive and a 30 minute drive (i.e. over the past two years I've been to Oxford more often than to Stanford, simply because it's so far away).

For the sake of EA Global travel times, I think treating them both as about similar size seems reasonable to me. Which is what we did in the analysis for this year's EAG. Though for everyday community building considerations, the difference in density is actually pretty important (and is reflected in the number of meetups, social events, EA orgs, etc.).

Comment author: John_Maxwell_IV 25 September 2016 04:05:11AM *  3 points [-]

I see an implicit premise that we're best off creating a single EA super-hub, but is that true? Here are some reasons it might not be:

  • In academia, we see prestigious universities located in far-flung cities, and that seems to work pretty well. When a close friend of mine graduated from a prestigious university, his professors advised him to attend graduate school elsewhere to get a different way of looking at things. I assume this is a combination of different research groups developing their own views and also the influence of local culture on the university. Both could be factors for increasing intellectual diversity in EA.

  • You talk about the number of local meetups, but in practice the Bay Area has relatively few. Meetups focused on evangelizing EA to newcomers are especially rare (despite Bay Area residents being an ideal target audience: open-minded, altruistically inclined, high income, educated, etc.), and people have talked about the difficulty of breaking in to the community without already having connections. I suspect there's some sense in which the Bay Area community just doesn't want to grow more. Contrast with stories of the NY rationality community (in the days before the founders left for the Bay Area). I wonder about the wisdom of showering the Bay Area community with visitors year after year, instead of choosing a smaller hub such that EAG might actually make a difference in terms of having it be a Schelling point.

  • A single super-hub means a single point of failure in the event of disasters.

Comment author: John_Maxwell_IV 27 September 2016 12:39:31PM 0 points [-]

An interesting fact re: the Bay Area community is the number of people who were organizers in some other place and mostly stopped organizing some time after moving here. I think there are at least 4 different people like this and perhaps as many as 8. I don't know if it'd be valuable to interview them.

Comment author: MichaelDickens  (EA Profile) 24 September 2016 05:20:19AM 1 point [-]

EA Hub, EA survey, and traffic for EA forum are all samples of the sorts of people who actively participate in EA online. They're all going to be biased in roughly the same way, so the fact that they say similar things is not strong evidence that they provide a representative sample.

Comment author: Habryka 25 September 2016 01:39:57AM 0 points [-]

Yep, agree with this. Sadly since online surveys tend to be the easiest way to conduct these, we don't really have much different data. There are a few things we will hopefully be able to estimate soon, which might help us spot inconsistencies between these:

  • # of members in EA student chapters in different locations
  • # of people who attend different EAGx events
  • # Origin of people who attend EAG (sadly we only have country-wide data for this year's EAG, since our registration completion rates dropped quite a bit when we increased the length, so we had to cut some questions)
  • Distribution of people engaging with the EA Facebook groups
  • Distribution of people having taken the GWWC pledge
  • Distribution of people who donate to meta-EA charities
  • Geographic distribution of newsletter subscribers for 80K and the EA newsletter

I would guess that at some point CEA will look into all of these, though I would be somewhat surprised if any of these massively disagree with the EA-hub/survey data. Still seems valuable to check though.

Comment author: AmyLabenz 23 September 2016 03:25:40PM 1 point [-]
  1. We are certainly not set on August. This year we were open to different summer months and ended up selecting August based on a number of factors, particularly the venue availability. Next year we expect to weigh the venue availability, top choice speaker schedules, conflicting events, academic calendars, staff availability, and other factors. Is there a reason that you prefer July?

  2. Thanks!

  3. See Oliver Habryka's comment.

Comment author: AGB 24 September 2016 12:19:34PM 0 points [-]
  1. I have boring work-related reasons to prefer July just because August is my busiest month, and I also remember hearing something about a clash with an animal rights conference in early August (weak memory, take with a pinch of salt) but I'm more trying to add support for varying the logistical hurdles caused by the location and time as much as possible to catch different people in different years, rather than making any personal special plea :).

Venue availability seems like a very reasonable deciding factor, that's good to know.

Comment author: MichaelDickens  (EA Profile) 24 September 2016 03:29:50PM 1 point [-]

In 2015, EAG was scheduled at the same date as the animal rights conference, and I know several people who could not attend EAG for this reason. This year they were both in August but at different times.

Comment author: Evan_Gaensbauer 22 September 2016 04:58:38AM 1 point [-]

Last year, there were three (or four?) official events in Mountain View, Basel, Oxford and Switzerland. This year, the only official event was in Berekeley. I understand coordinating the 20-ish(?) EAGx events this year would've been too difficult if it had all been planned by just one team like last year. Hoever, I never got why there wasn't at least one other event that was centrally coordinated, like one in Oxford. Like, I too am sure Ben is a great EAGx coordinator as well, but my point is having a team, likeone or more people dedicated full-time to organizing an event, is a surer way it goes off without a hitch. Like, generally, if someone is dedicated to organizing an event as there only top priority, you can expect more care will go into planning it. So, you can scale it more. It seems to me there's so many EAs in Europe, an event could be hosted there just as big as EAG in Berkeley this year.

There could be a separate team paid to work full-time to organize just as big an event in Europe. They'd need to be paid salaries or as contractors, which could be a big added expense. However, it seems holding an additional big event in Europe (or at least in two different locations) creates lots of added value for people who can now attend with mostly mitigated travel expenses. Also, you could have multiple >500-attendee events without the financial costs and logisitical headaches unique To organizing 1000+ -attendee events. So, you could end up saving money, and charging lower ticket prices, this allowing even more local people to attend.

Another possibility is volunteer organizers who did an astounding job with bigger EAGx events in 2016 could receive (more) funding to throw events with more people, higher-quality programming and/or more marketing in 2017. This way, more people would be able to attend a top-notch event at lower cost to themselves, without anyone on any events team having the coordination challenges that come with the biggest of conferences. I'm assuming as events scale in attendance numbers, finding a venue to fit everyone becomes harder. I expect something like this needs to happen at some point anyway because EA will continually be spread out across the whole world. This would require raising more funds for more one or more professional events teams, but I think the community would find it given the case for extraordinary added value to increase the total number of EAG-event attendees around the world by, say, 50-100% over 2016.

Have either of these options been considered?

Comment author: AmyLabenz 22 September 2016 02:16:31PM 3 points [-]

Now that we have a dedicated Events Team, we are considering additional in-person events. Things currently under discussion are EA Salons, EA Global Reunions, additional large scale events in various locations with different focuses (for example, one that is more curated and one that is like a convention or unconference), and provision of greater support for EAGx teams so that they can improve their own events.

One concern we have with the idea of hosting multiple large and substantially similar EA Global events is that we expect that fewer people will come to the main event or to the nearby EAGx events in different locations (for example, we have heard concerns that the plans for a large EAGx Oxford might cause fewer attendees in some of the other EAGx conferences in Europe). Also, the financial and logistical costs of hosting an event do not scale linearly. So, hosting two 500+ person events would add a large amount of additional work for the team, and many of the large costs, such as AV, venue, and speaker costs, would remain mostly the same (unless we reduced the complexity and content in addition to the number of attendees).

Comment author: Evan_Gaensbauer 23 September 2016 05:06:40AM *  0 points [-]

Sounds like you should keep doing what you're planning anyway. Thanks for responding.

Comment author: AmyLabenz 20 October 2016 03:38:43PM 2 points [-]

Update: after comparing the location data alongside the attendee and registration data and also considering the location survey results and other factors, we are leaning towards having three distinct events. It remains true that this will likely be more expensive for CEA, but, it seems that it is likely worth the cost on balance.

I will post more extensively on this soon and hope to firm up the locations in the next few weeks. We are tentatively thinking of Boston / Cambridge, London / Oxford, and the Bay Area spread out over many months, possibly with one in the spring, one in the summer, and one in the fall.

Comment author: AmyLabenz 23 September 2016 03:15:38PM 1 point [-]

To be clear, we haven't made a final plan. As a team, we're having discussions about additional in-person events because we're trying to be sure we're not missing something that we should consider. I would be happy to hear thoughts about which of the options sound most appealing, and will also consider other suggestions.

Comment author: Jeff_Kaufman 28 September 2016 02:29:02PM 0 points [-]

Really glad you wrote this up; thanks for sharing this information!

Comment author: AmyLabenz 19 October 2016 09:15:59AM 0 points [-]

Thanks! I'm glad you found it useful.