Event formats, including hybrid

I recently attended the hybrid SC21 conference in person, as one of the about 3000 in-person attendees, along with several thousand virtual attendees. (I don’t know the exact numbers, but in the past, an SC conference would have about 12000 attendees in-person.) In December, I’ll be attending the hybrid AGU Fall meeting remotely. And I’m also involved in planning a number of future workshops and conferences, so I’m thinking about how to run such events.

It’s clear how in-person events work. In my experience, the advantages to such events are the sense of community that develops, the opportunity to talk to people in hallways, and the fact that when I dedicate time to travel and to attending an in-person event, I end up learning about new things based on attending sessions that seem interesting at the time. The disadvantages are that such events involve travel and associated environmental impacts, and that they are often limited to those who can afford the cost of attending them, typically mostly those from the Global North and those in certain types of jobs.

It’s also reasonably clear how virtual events work at this point, after 20+ months of many of them. The advantages of these meetings are somewhat the opposite of the disadvantages of in-person events: they are generally cheaper than in-person events, have no travel and associate climate impacts, and are really open to anyone in the world with sufficient network connectivity. The disadvantages, at least for me, are mostly around timing: it’s harder for me to dedicate time to virtual events, so I often only attend sessions where I am talking or I am sure I will learn something, but I don’t have many serendipitous encounters; I often don’t feel very connected to others attending, or make social connections to people I didn’t already know; and for events that happen out of my time zone, I often cannot attend large parts of the event.

Some virtual events have worked on these issues, such as by running sessions multiple times in different time zones, either repeated or with different hosts and different speakers on the same topic or via recordings, and by creating social sessions (virtual icebreakers, use of gather.town and spatial chat, etc.,) and they do improve the virtual experience and make it more inclusive, but don’t really succeed in fully addressing the disadvantages and providing an experience that matches an in-person event.

I’m not really sure where this leaves us, as I don’t think the disadvantages of either in-person or virtual events can ever be fully addressed (e.g., it’s hard to see how to solve people being in different time zones without travel), I can imagine a world where we have both types of events (and gradually get better at running virtual events), and organizers develop a set of rules for deciding which kind works best for a specific event.

I think that having hybrid events instead of either purely in-person or purely virtual events is not the general answer, as they seem to be the worst of both in many regards. A hybrid event has to

  1. Provide something for in-person attendees that makes the extra cost of their attendance (in time, travel, and funds) worthwhile, and
  2. Make sure that remote participants don’t miss out on anything by not attending in-person.

It seems to me that it’s almost impossible to do both of these at the same time.

On the other hand, even given these disadvantages, maybe we should still have hybrid events, as neither in-person nor virtual events don’t also have disadvantages. If so, it seems we would want to provide guidance for participants about the advantages and disadvantages of in-person and virtual attendance, perhaps on a session-by-session basis as well as for the overall event, and let them decide.

I’m also interested in thinking about an event format where people meet in-person locally (or nationally), and these local meetings are then brought together in the overall event, similar to Research Bazaar (RezBaz) events.

Overall, I don’t think there’s ever going to be a solution. All three event formats have disadvantages, and none of them seem to be outweighed by their advantages. If this is correct, our goals should be to:

  • Understand the disadvantages of each type of event and reduce them as much as possible;
  • Understand how different types of events meet different objectives, and try to best match the format for an event to the objectives of that event; and
  • Develop and present clear guidance to potential attendees of their options.

Published by:


Chief Scientist at NCSA, Research Associate Professor in CS, ECE, and the iSchool at the University of Illinois Urbana-Champaign; works on systems and tools (aka cyberinfrastructure) and policy related to computational and data-enabled research, primarily in science and engineering

Categories Uncategorized6 Comments

6 thoughts on “Event formats, including hybrid”

  1. I don’t think SC21 was a good example of the potential for hybrid. It seemed to be more of a skeletal in person with mostly remote for the tech program. Once we are truly back to being able to have full in person events, then we can see how to make hybrid work. For example, force all questions to go through an online portal to avoid grandstanding at the audience microphone and to ensure that remote participants get equal chance to ask questions. Allowing remote participants to ask questions on video, assuming they have bandwidth, or via text entry and get live responses with discussion from the room would extend the room online making for a much better event. These are the things we can start experimenting with to make our events more inclusive for those that cannot attend in person for whatever reason.


    1. Thanks Jay. I will note that SC21 did do part of what you said, specifically, it required “all questions to go through an online portal to avoid grandstanding at the audience microphone and to ensure that remote participants get equal chance to ask questions”. I agree that allowing “remote participants to ask questions on video” would have provided more opportunity for back and forth discussion. Also overall, I completely agree with you that we should be doing these experiments, and not just doing them, but recording what we learn for others, so that each event series doesn’t have to do this alone


      1. SC actually had a bit more nuance than requring all questions to go through the online. They had the following session types: in person only, live stream, interactive zoom, and online only. Live stream rooms (panels, some workshops, and some paper sessions) had a microphone for in person people to ask questions and they didn’t have a good procedure for integrating the remote participants. They expected that the session chairs would just do it. Much of it was interactive zoom, which is as you describe. Part of my goal for SC22, since they are committed to offering hybrid again, is to develop some training for session chairs on how to manage a hybrid session equitably. This will certainly be open for community input and development to make this a better experience at all events. Anyone interested in helping this can contact me (I am easy to find).


      2. I was only involved in hybrid and virtual sessions, and for both of these, I was given clear instructions to force all questions to be submitted via sli.do, and that I would then read these out. One of these hybrid sessions was livestreamed, but both remote and in-person attendees used sli.do to suggest (and vote on) questions. I don’t think the session chairing was particularly difficult (and I skipped the training, assuming I would be able to figure it out), though more clear written instructions to both chairs and speakers would have helped.


      3. The session chair training covered the live streaming option as well and was far more complex than anyone expected. The training content took about 40 minutes with 20+ of good questions about scenarios that may not have been considered. Hubb used embedded zoom + sli.do and managing that for the live stream sessions was not obvious. How to and when to use pre-recorded videos and how to have remote live presenters vs. using the in room needed more practice to work out a simpler process. The biggest problem was that zoom does not handle floating licenses “properly”. Instead, we had to manually assign a license, and therefore a corresponding login/pw, to each session. Since those logins were not handed out as the licenses were reused for other sessions and they didn’t want inadvertent collisions, they had to manage all of the session with trusted people starting it and then assigning the host role. For a 16+ track wide conference, it required a large team.


  2. At SC21 in the audience (in-person) I was in a number of different types of sessions (workshops and seminars) and was surprised to find out that grandstanding could take place remotely; I honestly didn’t expect that to happen, I expected what you mentioned: local grandstanding. Depending on how well the session was organized, and how local host handled things with the remote hosts, determined the flow and participation of the local and remote audiences. I found some workshops well prepared and handled (Dan’s being one of them) vs others that weren’t, and it showed. Perhaps more so due to the technical nature of the sessions requiring audio and video in a mixed local and remote environment. I actually thought running ALL the questions through a remote hub made it easier for questions to be asked, managed, and responded to (when done well) from the local audience in particular (i didn’t have to get up and in line at a microphone). Everyone got to see the questions as they were proposed and could up-vote those one would have asked themselves. I often don’t ask questions at events, but the format allowed me to comment or ask questions I would not have done in front of a microphone.

    One thing the remote format running simultaneously allowed me to do: attend two sessions as once- one in person and the another virtually (hurrah for earbuds). Not ideal, but sometimes you want to listen in on another session while it’s happening rather than try to go back to a recording that may or may not have been able to take place. Granted, your focus is split, but if you find yourself in a session that turns out not to be what you expected, you can then focus on the other one… without having to rudely leave…

    I also found out I couldn’t count on recordings for all sessions as part of one session could not be made due to one of the speakers not giving permission. Just had to toss that in there.

    I agree with a lot of the comments above: I’m not sure SC21 is a good example of what a hybrid would normally be (i.e. near full participation WITH remote audience), but at the same time, while missing a large segment of folks who normally would have been there, I did manage to have some of the spontaneity and discovery one gets while in person with the encounters I did have that a remote conference mostly lacks. I have yet to see a successful remote conference duplicate the benefits of being at an in-person conference. I’ve seen many attempts to try (of the ones i’ve attended) – just in bits (pun intended) – but not byte sized enough… to be satisfying 😉

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s