Microsoft Research was able to set up its New Future of Work conference (August 3-5) as a virtual conference right from the start. This meant that it was probably one of the first to be designed at the outset to be virtual rather than adopting and adapting an existing one-site conference agenda. The conference was excellent in terms of content, and all the papers (67 of them) and themes can be found on the conference web site. Given that Microsoft started with a clean slate I thought you might be interested in the approach that it took with the delivery platform. I’ll summarise the outcomes of the conference in Part 2 in the Autumn issue as it will take me some time to work through many pages of notes.
As you might have expected the technology platform was Teams, and even the Microsoft team were getting annoyed with its intransigence and poor usability. I’m beginning to see a marked difference between video meeting software and video conference software! There were two keynotes and some well managed plenary panels, but most of the presentations were presented in nine themed discussion sessions with three or four papers in each one-hour session.
All the presentations were delivered as voice-to-text streamed below the screen. The text stream was around 60 characters wide and in upper case only. The latency between the spoken word and the text was very low, and generally very accurate. I found this slightly annoying as I found myself checking out the quality of the text delivery when I should have been listening to the speaker. I’m not sure at all that I understand why this feature was included, given that it is very difficult to parse text which does not have the conventional mix of upper and lower case letters.
The presenters could either speak for 5 minutes (a hard limit!) or present a video (same length). These were presented one after another, after which there was a 40 minute Q&A using video and chat. Each session had a moderator and a technical support person. The moderators had clearly been chosen with care and were contributing, not just facilitating.
There was a standard set of topics for the discussions in each of the sessions. This enabled the moderators to create a single PPT at the end of the conference to summarise the discussions in their themes. I felt that the extended discussion sessions helped take the place of a lack of social interaction, as the Q&A helped in many cases for questions to make new connections in the discussion audience. At least you knew that everyone in the theme room had a common interest!
Microsoft did its best to encourage networking, for example making use of Minglr which was available from MIT in a beta release. However even the Microsoft team admitted at the end that this social aspect was a) very important and b) didn’t work at all well. For me there were two particular issues. The first was that there was no list of attendees and the second was that in the discussions most of the attendees were represented by one or two capital letters in a row at the bottom of the screen. The end result was that it was like attending on on-site event where everyone had a badge with just one or two capital letters. I don’t want this to be seen as a criticism of Microsoft. It was trying to create a social environment and the outcomes of they actions will be of benefit to others trying to break through the difficulties of providing both structured and serendipitous social networking.
The conference web site has all the papers and some useful ancillary information. I would expect more content to be added over the next few weeks, and it would not surprise me if Microsoft Research set up more conferences in the future, perhaps with more of an emphasis on information management than on employee behaviour.