The mythical 17:00 split

Here is the new talk I just made at #pubconf organised fantastically by amazing people. If you were not there you missed something rather amazing.

The mythical 17:00 split from Sebastien Lambla


Versions Are Evil

Thanks to all that came to my short session on versioning. The slides have now been published, see below.

Feel free to provide feedback and comments here, or wait next week for the new speaker section of the site to be ready.

Versions are evil – how to do without in your API from Sebastien Lambla


What's good feedback from a talk?

As happens often when hanging out with speakers, the discussions started gravitating towards the ever fundamental question of talk feedback.

For those of you not in the known, most conferences now have a system of red/green/yellow to represent the mood of the room (I think). The temperature is a good indicator of wherether your talk polarised the audience. I tend to prefer green and reds to yellows, as it means I triggered some reaction in the audience. A large amount of yellow is a meh talk, and when you get that, you need to go back to the drawing board.

All this however is not enough to decide on how to improve your content, and that’s where the feedback forms come into play. I think they’re quite a flawed way of getting feedback.

They’re used more often by people that are unhappy than by people that are happy enough. That skews the results towards negativity. For example, at build stuff, the majority of talk feedback i saw published by speaker friends seemed to show a trend of lower aproval ratings than the tricolor system, although that would require a better understanding of stats than I seem to have, as highlighted by the brilliant Barbara Fusinska.

Another speaker I know was mentioning that the system was very unfair. Popular speakers sometimes get rated fully green just for the effort of showing up. I unfortunately think it comes more from the popularity culture than from the feedback system itself.

More importantly, for me to use the feedback, I sometimes need to drill into it. Did the attendees not understand something? Is something missing? Can the content or the delivery be made better? Can my jokes be any worse? By receiving only comments, I tend to not be given the opportunity to understand better. My good friend Sam Elamin made a very good point, highlighting that attendees may want to stay anonymous while providing feedback, and that conversations while protected by that anonymity would allow everyone to express their opinion and allow me to drill deeper.

Taking all this into account, I’m in the process of rearchitecting the part of where talks are indexed. Each talk will have its own discussion, and I’ll enable anonymous comments on those, so we can start conversing about these. I’m also gonna go ahead and create hash tags for each talk.

Have any ideas? Share them in the comments below!