I just came back from New York where I hosted the second edition of Edge, a web conference organised jointly by the FT and Google. It was a great success, and we raised over $8000 for Girls who code, a wonderful charity and a great cause for the technology industry to get behind.
Edge is the product of a lot of thought about what I don’t like about other web conferences, so I thought I’d share those thoughts with you:
- Cost: A lot of conferences cost upwards of $1000 to attend. And they often don’t pass that on to the speakers (at least, not in my experience!). If you want your conference to be about smart people discussing the way forward, selecting for wealth doesn’t seem like a good way of doing that. So we decided Edge should be cheap enough that anyone can go.
- Expertise level: It used to be that the web was fairly simple, so an expert was not a million miles away from a novice. Now the complexity is phenomenal, and conference organisers too often don’t do a good job of grading talks and matching talks to audiences. It’s hugely frustrating if you’re in a talk that turns out to be entry level only to later discover you missed one that was deep and complex. We run a call for delegates in a similar way to a call for speakers, to try and get a fairly uniform, normalised level of expertise in the room.
- Verbosity: Everyone hates the person that stands up and spends 5 mins asking a question in the longest winded way possible that no-one wanted to know the answer to anyway. So we impose speaking time limits on delegates as well as panelists.
- Format: Having a 30-60 minute talk with a brief Q&A is fundamentally the wrong way around. We do need that long-talk format, in a training/teaching scenario, and many web conferences have essentially become that: a place for the less experienced to go and learn from experts. But that’s not the way standards are made – they’re made in small meetings of invited experts. As an outsider, they can be very difficult or impossible to interact with, and can operate in a very opaque way. So Edge tries to take the best of both worlds – the openness and acessibility of web conferences with the productivity and insight generated by working groups, by running panels with 10 minute opening talks.
- Online participation: A lot of conferences put their talks online, which is great, but very few of them allow you to watch the conference in real time online. We wanted to make the online streaming almost as good as being there, and as a result, for Edge 2 we had an online audience that at least doubled the audience in the room.
- Transparency: If an event is for the benefit of the community, and is paid for, there’s really no excuse not to be completely transparent about the way it’s funded, where it gets its money and what that money is spent on. For Edge, we present a summary of our accounts on a slide at the end of the event and announce the donation that we are making to our charity partner, which is whatever money we have left over.
Our first attempt was Edge 1 earlier this year in London, and this week at Edge 2 we iterated and changed several things – we added better timing control, developed a tool for promoting more audience interaction and instant feedback. We also pre-selected questions to ensure we covered a broad selection of topics, and that the questions were quick and succinct to ask.
The most striking thing we tried was OnSlyde: We worked with Wesley Hales who created a version of his OnSlyde audience interaction app adapted for panel debates, and that gave us the ability to offer the audience real time options for ‘Agree’, ‘Disagree’ and ‘I want to speak’. Compared to Edge 1, this gave us a massive increase in audience interventions, and some amazing data on the sentiment towards the opinions being expressed on the panel. Wesley crunched the stats and produced these spectacular reports:
That all worked pretty well, but we still have stuff to improve:
- People lost focus when we were debating a particular point, and forgot the context, instead just asking fresh questions. We should put the questions on the screen to maintain that context
- The interaction tool we used, OnSlyde (developed by Wesley Hales) we expected moderators to use to nominate speakers. We discovered very quickly that it was way better for a behind the scenes person to do that, so we should move that control device to the AV booth.
- We forgot to put power outlets out for the delegates, so people gradually ran out of juice during the day. Oops.
So, a lot to be proud of, and a lot of opportunity to do better. Sounds like we need an Edge 3. Watch this space.