All problems in computer science can be solved by another level of indirection
Last week my London based C++ user group, C++ London, joined forces with SwedenCpp, based in Stockholm, for a distributed event where we shared video streams with each other. The whole thing was hosted by King, who took care of the audio-visual link up, so we just had to organise the speakers.
We did this as a series of lightning talks (5 or 10 minutes each) - 5 in Stockholm and 7 in London.
This was an experiment. The idea was that, if successful, we could do more like this. I know for a fact that other user groups have been waiting to hear how it went as they think about doing something similar too.
So how did it go? Well, you can see for yourself. The video is here (if you view it on YouTube you'll also find a Table Of Contents to jump to individual talks):
Overall I think it went very well. There were certainly some rough edges, and opportunities to do better next time - but the basic concept worked well, I think. Feedback from attendees in both locations also bore that out. Everyone is keen to do this again!
Paul Dreik, in Stockholm, sent out a survey to all attendees after the meeting and collected some stats and comments. Regarding the overall event, over 80% said "Great, let's do it again!". Of the rest, all but one said it was at least as good as a normal meet-up. Only one person thought it was a step down (you can't please everyone).Interestingly, while lightning talks, as a format, was popular (over 50% thought at least most of the time should be spent on them), a sizeable 40% thought more time should be spent on longer talks. So maybe some combination could work well?
Beyond that there were a couple of people who said that it was too long without a break, and some people wanted time for questions as we go (which is hard to make work for lightning talks).
So, some of the things that could be improved (and notes for other groups looking to try):
We could have planned more interaction between the locations. We were very much focused on our own schedules and, other than the handover in the middle, there was not much beyond two independent sites that happened to be watching each other. There was an "open questions" section at the end. Jean Guegant, in Sweden, had mentioned that we'd do this in the intro but I'd missed it - and I think others had too, so we weren't really prepared for it. Between that and everyone forgetting any questions they'd had during the talks meant we didn't get many questions. I think we can do better here - perhaps collecting questions during the talks via a web app, or maybe a Twitter hashtag?
Are there other points during the course of the event that we could interact more between the sites? It can be a tough balance, but I think there is scope to experiment here.
And finally, while we're very grateful to King, and their AV staff, for providing, and setting up, the live stream, as well as recording, I think we made too many assumptions about how that was going to work. In particular we were left with an audio recording that was not as good as we had hoped for. In the London audio, especially, the hand mic and ambient mics were mixed together before recording - so we couldn't separate them out - which would have been very useful in the editing.
So. Would we do it again? Yes, definitely! I think this is a great way to expand the reach of our speakers, and give our members more variety of speakers and topics to listen to. In that respect this was a great success, and I'm excited to see what happens next.