I'd like to share some thoughts from @EAMSConf, which ran a couple of weeks ago, to help anyone else running an online conference.

We were lucky that we'd decided to do everything online back in January, so all the hard thinking was done before lockdown started.

4 favourites 3 retweets

In reply to @christianp

We decided to go online because of the subject of the conference (e-assessment) but also in order to include more people around the world who would find it hard or impossible to get to Newcastle for a physical conference.

0 favourites 0 retweets

In reply to @christianp

Travelling for a conference is a huge burden in time, energy and money, even when you don't need a visa. Personally, I'm restricted in what I can attend because Ehlers-Danlos syndrome leaves me shattered after a day of travel and standing talking.

0 favourites 0 retweets

In reply to @christianp

Having decided to move online, we realised that the normal format of a couple of days packed with activities wasn't necessary or even desirable. Unlike travelling to another city, there's nothing stopping you just abandoning an online conference halfway through.

0 favourites 0 retweets

In reply to @christianp

And, as most of us have discovered since, staring at a video call all day is knackering. There's also a timezone problem: there are only short times in the day when the UK working day overlaps with the Americas or the far east.

0 favourites 0 retweets

In reply to @christianp

So, we set a hard rule of two one-hour live sessions each day, at 9am and 4pm, spread across eight days instead of the normal two.
I think that worked really well, and the feedback from attendees agrees.

3 favourites 0 retweets

In reply to @christianp

A lot of the talks were pre-recorded. I think maybe they should all have been. A couple of the non-recorded talks had technology problems, in spite of our efforts to rehearse beforehand.

0 favourites 0 retweets

In reply to @christianp

In the live sessions, we had a chair who introduced speakers and read out questions at the end, and someone else (always me) who played out pre-recorded videos via screen share. This worked better than expected! I forgot the sound a couple of times, but nothing terrible happened.

0 favourites 0 retweets

In reply to @christianp

All questions were taken through the course website using a stackoverflow-like interface, rather than Zoom. This worked brilliantly. It let moderators carefully pick and reword questions, rather than the mess of unmuting people, and conversation could continue afterwards.

0 favourites 0 retweets

In reply to @christianp

(We used the MoodleOverflow plugin: moodle.org/plugins/mod_mo…)

0 favourites 0 retweets

In reply to @christianp

We initially planned for more interactive sessions, such as workshops for the popular systems, but they made way for more talks. I think we should have made sure to keep them. There was very little of the sort of casual discussion you get at break times in a physical conference.

0 favourites 0 retweets

In reply to @christianp

For non-Zoom activity, we set up a Moodle environment. It worked brilliantly, but some bits were used less than I hoped. We really wanted to give attendees an opportunity to see in action the systems we were talking about.
I set up four areas:

0 favourites 0 retweets

In reply to @christianp

"Main", for conference info, general discussion, and reference. The general questions MoodleOverflow was used a bit, but the standard forum and wiki-style reference not at all.
There was an "introduce yourself" glossary which a few dozen people used, but nothing happened with it.

0 favourites 0 retweets

In reply to @christianp

"Talks", with a link to the zoom meeting, recordings of all the sessions and individual talks, the question areas, and associated material such as slides. This worked very well.

0 favourites 0 retweets

In reply to @christianp

"Play area", where everyone was enrolled with teacher privileges. I hoped that people would use this to try out the systems we had access to or show off things arising in discussion, but nobody touched it. We'll have to rethink how this should work next time.

1 favourite 0 retweets

In reply to @christianp

"Software demos", with a course for each system being presented. There was a MoodleOverflow questions section for each system, which was well-used.
The stats show that only a couple dozen attendees clicked on anything in these sections.

0 favourites 0 retweets

In reply to @christianp

We kept registration open throughout the conference. We had just over 300 registrations on the morning it started, and 420 by the end. Attendance in the live sessions was in the 80s early on, down to around 60 by the end.

0 favourites 0 retweets

In reply to @christianp

I don't know how much people who didn't attend live sessions engaged with the non-live material. None of the recordings had more than 60 views at the end of the conference.
We were advised to expect about a quarter of people who signed up to actually attend live sessions.

0 favourites 0 retweets

In reply to @christianp

All the videos went on a YouTube channel (youtube.com/channel/UCMN_1…). Every video was initially unlisted, available only through the conference website. Now they're published and linked from the conference website: eams.ncl.ac.uk/programme/

0 favourites 1 retweet

In reply to @christianp

Feedback from attendees: they really liked the stretched-out format, and the short sessions. Most made use of the recordings. Many missed the ad hoc discussion you get in person.

1 favourite 0 retweets

In reply to @christianp

In conclusion: A++ would do it this way again.

1 favourite 0 retweets

View this tweet on twitter.com

This tweet as JSON