Against a background of decreasing funding in higher education for conferences and travel, this paper argues for an alternative approach to international collaboration, based on work by the author and colleagues on an online conference run in Adobe Connect: the ‘Giving Feedback to Writers Online’ conference which took place in June 2014. This paper draws on lessons learnt from that conference to demonstrate that running an online event is both easier and more effective than you might expect. The paper will be of interest to anyone who is looking for ways to widen the reach and accessibility of the seminars and conferences they run, and to learning technologists and others who might be called upon to support such events. As well as the discussion of our experience in running an online conference, the paper pulls together the outputs from that experience to offer a takeaway checklist which maps out the key stages in planning an online conference, and can help organisers of future online conferences to avoid some of the obstacles along the way.
This paper is drawn from work undertaken with colleagues Mary Deane and Marion Waite on their virtual conference ‘Giving Feedback to Writers Online’, which took place in June 2014. Mary and Marion wanted to get together with international colleagues to share ideas about online writing feedback, and their big idea was to run an online conference. My task became how to set this up, and how to prepare facilitators, presenters and the online room itself for this relatively novel way of meeting.
When setting up the conference, we were learning how to use Adobe Connect, our new real-time online teaching and meeting environment. We were still experimenting with the tools for audio, video, and text communication, and the opportunities offered by the interactive whiteboard, the polls, and the ways that ‘layouts’ allow us to create different spaces within the same online meeting room. Within this article, I hope to share some of the things we have learnt, and encourage colleagues to consider organising similar events. The article ends with a takeaway checklist of things to consider in running an online conference.
The literature on technology-enhanced learning is clear about the many advantages of using this kind of synchronous session to enhance student learning, but also about some of the pitfalls to which it is prone. Many of the same issues apply to the use of synchronous conferencing technology for collaboration amongst colleagues as to teaching and learning with students: in both cases, the technology is being used to facilitate conversations and the development of ideas between people who may be geographically distant from each other.
This possible distance is perhaps the most important reason to use synchronous online environments. Rhona Sharpe talks about this as ‘the killer app of technology’: ‘bringing together people who would otherwise not meet, giving learners access to tools, resources or materials which would otherwise not be available to them, giving individuals access to participate within communities’ (2012, p. 219): true as much for our work with colleagues as for our work with students.
As with students, successful online interaction requires scaffolding and preparation. A useful, commonly cited way to understand this is with Gilly Salmon’s ‘Five Stage Model’ of online learning, which suggests we need to pass through stages of checking access and motivation, and some online socialisation, before we can get to the ‘higher’ levels of information exchange, knowledge construction, and development, which we are aiming for.
(Image used as permitted under the terms of Creative Commons BY-NC-SA license.)
Ignoring these preparatory stages can make it very difficult for participants to engage in the online environment: as Stephen D. Brookfield and Stephen Preskill put it, ‘online environments are sometimes experienced […] as sterile, unfriendly, and alienating’ (2005, p. 231). Salmon’s stages one and two, and much of our preparation for running this conference was geared to avoiding these potential problems.
Preparing the presenters
The most important part of preparing the presenters for the virtual conference was to give them all a chance to play around with the software, with one or more of the facilitators present to guide them through the options and test out how things worked. Unfortunately we didn’t quite manage to achieve this for all of the presenters — some couldn’t make the same times we could, so simply had a play with the settings on their own, and others could only manage to meet in a fairly large group, which meant they didn’t all get the one-on-one attention from facilitators which might have given them the best chance to get to know the setting in advance.
But within these limitations, most presenters were able to spend time with an experienced guide exploring the menus and tools in Adobe Connect. Doing this well in advance of the conference date meant they had the chance to change their presentation plans, and their slides, to take account of the possibilities offered by the software. It might have been even more helpful if we had been able to offer an introductory session to all potential speakers before they had even begun to finalise their papers, to give them the maximum opportunity to take advantage of the available features.
We spent time in trial sessions discussing polls, whiteboards, and breakout rooms, as well as introducing the basic text, audio, video, and slides features. Polls were a popular feature, giving presenters the chance to quickly gather data and ideas from participants, and share those results back with the audience. A limitation of free-text polls, which we discovered too late to avoid, is the restriction to only one response per participant. We think that in future, whiteboards will be a better way of capturing this kind of brainstorming, with the advantages of being anonymous, and nicely visual (although they require someone to move answers around the screen to keep them all visible). See the image below, which is an example from a whiteboard used in a different context, one of the synchronous sessions in the Teaching Online Open Course (TOOC):
Breakout rooms weren’t a success. One presenting pair hoped to use them, but after test runs and discussion amongst the conference facilitators, we advised against it. In a one-off context break out rooms can be difficult to manage and confusing for participants. So with a relatively short time period, and a small group (we were expecting 32 participants), who are not necessarily all experienced with this type of environment, we decided against them. We’re all eager to give breakout rooms a try in future, though, when we have more time for participants and facilitators to get used to how they work.
With all that practice done, we asked presenters to send their slides, and detailed instructions for their other requirements, in plenty of time for us to set up the room for them.
Preparing the room
A big advantage of Adobe Connect is its ability to set multiple layouts within one room: our conference used eleven different layouts within the conference space, which facilitators could move between as needed. These were:
This is where people landed if they arrived early. It contained a slide welcoming participants, advising them how to check their audio settings, and a link to a video introducing the system [https://www.youtube.com/watch?v=716PsnddCZA]; a list of people who were currently present; a text chat so that people could start talking among themselves; and a video pod showing one of the conference hosts, welcoming people to the conference. In many ways this was the most important layout of all: it was the layout which we used to begin guiding people through Salmon’s first two stages, to check that they were able to access all the tools, and to start some informal chat, both audio and text, to help people relax as they arrived.
Welcome and introductions (pictured below)
We moved to this layout to officially begin the conference, and to run a short welcome and orientation presentation. This continued the scaffolding work begun in the lobby, formally introduced the tools available for participants, and the hosts who were supporting the session.
Eight speaker layouts
That is, one layout for each of the five papers, and three extra ones for one pair of presenters who were using some complicated poll arrangements. This was the ‘meat’ of the session, in which we moved back and forward through Salmon’s stages three, four and five.
Wrap up and goodbyes
This contained a single slide with a ‘thank you’, and a link to the conference evaluation survey.
All of our layouts contained the same attendance list, text chat and video pods, and the same basic arrangement of boxes, so that when we moved between layouts the transition was as seamless as possible for presenters and participants (only people with ‘host’ permissions in Adobe Connect can access the layouts; others are limited to the one which is currently active. In this case, that meant that only the conference facilitators could move between layouts, although we could have given host access to anyone else if we had needed to).
Just before the conference started, I made particular use of Adobe Connect’s ‘prepare mode’, which meant that while participants were arriving and being greeted by one of the conference hosts, I was running around behind the scenes checking the other layouts, without disturbing anyone else’s view.
Most of our speakers used polls, and setting them up was probably the part of the preparation which took the most time, but for the most reward. We asked speakers to give us poll wording in advance, and to signal on their slides when they wanted polls to appear, so when they were ready they only had to ask: ‘can I have poll number three now please?’ I named each poll with the presenter’s name and the the poll’s number in that paper, for example, ‘Ros poll 3’, to make it easy to bring up the right poll at the right time.
In preparing to use these ‘pop up polls’, I set up the polls with their questions and answers, and carefully positioned them to overlap the speakers’ slides, but not the chat or the video. This careful positioning didn’t work seamlessly: while the conference was running, I was present in the room on two different computers*, and did some minor tweaking to improve the display size while the polls were live. Next time I’ll spend more time checking how the layout looks on different size screens.
Designing the arrangement of polls within layouts was also difficult where presenters wanted polls and instructions on screen at the same time, which was why I used multiple layouts for one pair of presenters, recreating instructions from a slide in a notes pod, to make it easier to fit both the instructions and the polls on the screen at the same time. More time to tweak and practice in conjunction with speakers would make this easier in future.
Polls worked particularly well with an activity where participants were asked to download and read a short document, then tick boxes to answer two different questions about it, and the speakers used the poll results as a jumping off point for discussion.
(*Big screen on one; decent microphone on the other.)
Preparing the facilitators
Two days before the conference (when everything was theoretically set up, although unavoidable delays meant that wasn’t actually so), the conference facilitators met for a dress rehearsal. We went through the content and layouts in order, and made decisions about the responsibilities each of us would take. Going through all the content meant we were all familiar with what was happening when, where the polls were needed and how to find them, and when we were moving between layouts. It also gave us all a chance to look at the content in the ‘framing’ layouts — the lobby, welcome, and goodbyes — to check they included everything that we needed. And perhaps most importantly, it meant that if my connection had failed on the day, the other hosts would have known where to find the things I’d set up.
Online conferences operate in multiple channels at once, and paying attention to all of them is difficult, so it’s important to assign roles to facilitators to make sure everything is covered. The main tasks we assigned among ourselves were:
- Welcoming early arrivals
- Watching and interacting with the ‘back channels’: text chat, and other media like Twitter if you’re using them
- Making a note of any questions from the back channels to put to the speaker
- Listening to the speaker (and don’t forget the traditional role of the host to ask questions in case the audience don’t do so immediately)
- Tech wrangling (making polls appear, moving between layouts, muting participants’ microphones if they create feedback issues)
We split these tasks between four conference facilitators, which gave us enough overlap that we could have covered it if one of us had our own technical failure. However, we learnt that the sessions which had the best interaction between speaker and text chat were those with two presenters: at any time one of them was speaking, and the other watching the text chat. This worked as a great addition to the facilitator watching the text chat, because two presenters who knew each other well and had worked together to write their session were much more comfortable interrupting each other than the hosts would have been, and more able to do so at suitable moments. They were also able to answer questions and give extra information without having to interrupt the speaker.
We didn’t come up with a good online equivalent to the chair’s ‘five minutes’ notice that we use in face to face conferences. Adobe Connect’s ‘presenter area’ seemed like a good place to put that warning, but turning on the presenter area meant that all new pods (such as the pop up polls we were using) automatically appeared there, which made managing these slower and more work.
What we learnt
Audio quality was the most persistent problem. Some people’s microphones weren’t clear, some produced feedback, and not everyone was able to follow our advice to wear headphones as a bare minimum.
Free text polls only permit one answer per person, meaning we may have lost some data. They’re also impossible to extract results from: reusing the poll responses requires manually retyping the responses. Polls might also be problematic because they look anonymous to participants, but are not. A solution to these problems for free text entry is using the whiteboard: it is anonymous, it’s possible to copy-and-paste to get the text out, and people can type as many things as they like. It’s not as visually neat, but may be more fun.
We misjudged how much time to allow for questions: the timetable demanded we move on to the next slot when there were still questions to answer. Our solution was to ask the previous speaker to answer questions in the text chat, but this was probably distracting for the next speaker and for participants who were ready to move on. Collecting questions in text chat meant we could keep them for later, so perhaps a solution would be to ask the speakers to respond separately to the outstanding questions, either by email, or perhaps in a recorded Skype call with one of the conference hosts, to be distributed to participants.
What the participants said
We asked participants to complete a Google form evaluating the conference, and 16 of them did so. Of those, five explicitly mentioned some aspect of the online format of the conference as something they learned from their attendance, and in general the response was very positive, with high scores for the usefulness of sessions (average session rating 1.9 on a scale from 1, ‘very useful’, to 5, ‘not at all useful’), and on the overall conference organisation (1.1 on a scale from 1, ‘very organised’ to 5, ‘very disorganised’). Four respondents mentioned audio problems, and two asked for slightly longer breaks between speakers.
Some typical responses
Can you identify at least one thing that you have learned from your attendance at this conference?
- ‘The importance of having time to rehearse using the technology!!’
- ‘How well the polling activities worked – I particularly liked the [third] presentation in this respect when they used text input in the polls’
Can you identify any ways in which the conference could be improved?
- ‘Audio was inconsistent for me but likely had something to do with internet speed’.
- ‘More rehearsal time – but that was to do with technical problems my end’.
- ‘I thought it was pretty much perfect to be honest’.
- ‘It was very well run. If the few technical hitches (e.g. [Presenter’s] audio probs) could be addressed it would be even better.’
- ‘Sound was an issue at times, which is pretty unavoidable.’
- ‘The organisation and facilitation were excellent. Special thanks to the organisers for this opportunity to connect with people across the world.’
- ‘Perhaps a little more time between each presenter session to allow time to catch up on chat conversations’.
Do you have any additional feedback or comments that you would like to add about your experience of the conference?
- ‘A very positive experience! It was my first online conference and I was rather nervous / sceptical – I thought it would be hard to concentrate so long and that there would be all sorts of technical issues which would take time but in fact it was very clear / smooth. I just need time to get the hang of the multitasking – i.e. the slides / the chat – very easy to get distracted!’
- ‘I was impressed at the ease with which you used the environment – even the audio hitches didn’t create too much of a problem.’
The experience as a whole was an extremely positive one for the conference facilitators, and the feedback we collected suggests that participants and presenters felt likewise. The set up and preparation took time and planning, but probably less than a face-to-face equivalent would have involved, and the outcome was an event with much wider accessibility than a face-to-face event would have offered. Our online conference created the opportunity for a group who would not otherwise have been able to get together to spend time sharing and discussing their work, with no need to book and pay for travel, no need for extended periods of time away from home and work, and no special preparations required of participants.
Having done this once and learnt more about what works and what doesn’t, we are keen both to try it again (with improvements), and to share our experience with others. It is our hope that other people running similar events might be able to benefit from our experience, to improve on our approach, and let us know more about what worked in their contexts. To this end, we have put together a checklist of things to consider in running online conferences, and we welcome suggestions for updates and improvements to it, by email to email@example.com.
(Note: some of these are specific to the environment we were using, Adobe Connect, but most apply to all similar environments.)
Prepare the programme
- Include explicit opportunities for participants to try out the tools
- Leave longer breaks between sessions than the 5-10 minutes we allowed
- Encourage presenters to use interactive tools to engage participants
- Group together presentations on a theme
- Consider how long participants can concentrate online
- Give them a chance to test out the room, with an experienced facilitator, before finalising their content
- Encourage them to use interactive tools such as polls to gather data and run activities
- Encourage them to work in pairs: two presenters can take it in turns to speak and watch the text chat. Speakers who are presenting solo work can still bring a co-presenter who can be better placed than the conference hosts to answer questions in the text chat, and to know how and when to interrupt the speaker.
- Ask them to use blank slides with, for example, ‘Poll 3’ at the appropriate point in their presentations for the poll
- Allow plenty of time for questions
- Make sure they test their audio well in advance, to leave time to fix any issues
Prepare the room
- Plan what people will see when they arrive, and at the end of the conference
- Use layouts to set up content in advance for quicker transitions between sessions and activities
- Keep the room arrangement consistent except where there’s a reason not to
- Use clear consistent naming for items that need to appear on the fly
- Check the appearance of layouts with different size screens and adjust as necessary
- Run a dress rehearsal with all the technology and content set up
- Agree who is responsible for which tasks, but also…
- …Build in redundancy of roles in case of technical failure
- Designate someone to welcome early arrivals
Brookfield, S.D., Preskill, S. (2005) Discussion as a Way of Teaching. San Francisco: Jossey-Bass.
Salmon, G. (no date). ‘The five stage model.’ Available from: http://www.gillysalmon.com/five-stage-model.html
Sharpe, R. (2012). ‘Accessibility: the killer app of learning technology?’ Research in Learning Technology 20 (3): 219–222. Available from: http://www.researchinlearningtechnology.net/index.php/rlt/article/view/19584.