Talk format.

[nrn]

One sacred thing I'd like to see challenged is the standard talk format, where one person has a polished presentation trying to convince a passive audience to see things the way they do. This is great in that it is the most predictable, presenters work hard to make this format good for the audience, and ideas get to spread from one to many. However it doesn't create anything new.

I think there are a lot of possible options here, including just sticking with the standard :) One that has proven to work well other places is the bar camp style unconf. Whatever we do I really like the idea of building something or moving understanding forward instead of just disseminating knowledge.

[junosuarez]

i would love to see some form of debate or critical inquiry. needed: a controversial proposition, 1+ persons willing to advance arguments for each side, and a good moderator

[ceejbot]

I would love to see some moderator & point/counterpoint. Panel discussions for some topics. Chairs in a circle with everybody getting a turn to talk for others.

[junosuarez]

a pair talk with a system author and a system user (bonus points if the two haven't necessarily corresponded much ahead of time, either in the development of the system or in preparation for the presentation)

[interlock]

The Fish Bowl format may work well some discussions. Hopefully the talks are documented somehow with video/audio/wiki/time capsule/etc.

[brycebaril]

I'm definitely interested in this Fish Bowl format idea.

One note about a lot of these debate or discussion formats is that a format (e.g. the traditional speaker/audience format) where one person gets to talk uninterrupted is a lot easier for introverts or people who aren't as comfortable in conversations. Once they muster the courage to talk it can be defeating if someone can interrupt them.

[mikeal]

panels, debates, etc.

I've tried to play with this format in tech events and have found limited success. The critical problem is that two people may differ on one specific idea or approach but otherwise mostly agree. The more people added to the panel/debate the closer the participants are to each other.

This is probably the best debate I've ever seen. I've watched it several times and even once read the transcript. It's from 1971 and is a debate between Michael Foucault and Noam Chomsky.

https://www.youtube.com/watch?v=3wfNl2L0Gf8

A few things I would steal that I've never seen at a tech event.

• Forget audience participation. The audience tends to ask questions they already know the answer to. The questions also lack any linearity which means you just get a scatter-shot of ideas. Just leave it out, even if people finish early don't ruin it by having the audience ask questions.
• The moderator should prepare a linear collection of questions to ask and be prepared to cut questions if they end up being addressed by prior answers.
• Only two participants and keep the scope of the debate to something they actually differ on, that way you can dig backwards through their reasoning.

[mikeal]

I don't know if anyone has noticed but I've been tearing apart and rebuilding new conference formats for some time now, so I have a lot to say on this matter as well as a long list of both successful and failed experiments.

One thing that will help tremendously is not to consider or kick around any aspects of format until you identify and articulate the goals of the event, and potentially reduce the number of goals to something more accomplish-able.

For example: the goal of NodeConf 2012 (PDX at the Imago Theatre) was to expand everyone's understand of what Node was and what Node could do. Just throwing talks at people that are outside the realm of what they had previously considered would not accomplish this but I knew that if I was able to create some kind of narrative it would roll out more like a story and by the time we got to talks about robots the audience would all have arrived there together and it wouldn't seem at all out of bounds. That lead to the layout of content in to "acts," reducing the talk time to 20 minutes in order to have talks rely on each other more, the use of a stage play theatre where people would watch it like a play, and not posting a schedule (stolen from Funconf) so that the story would unroll in front of people and they would be forced to attend every talk as not to miss any part of the narrative.

For NodeConf 2013 I wasn't satisfied with people just understanding the breadth of Node, I wanted them to actually feel enabled to do the things people were speaking about. This lead to me throwing out talks entirely and moving to a hands-on workshop format. That goal has remained the same and this year as well as next year are basically iterations on what worked and didn't work from the previous year.

So, I guess what I'm asking is: what are the goals?

[mikeal]

Semi-relevant prior experiment: TenConf (10 minute talks, everybody speaks).

The idea being that everyone who attends is a speaker, you can talk about anything you are passionate about.

The first and most obvious problem is the number of talks. If you have no transition time you're looking at 6 talks an hour, so even with 10 hours of content over two days you can only do 60 people.

The event was a success in that the talks were amazing, the people who were there participated in each others talks in a really great way. It was a failure in the sense that it only sold 30 tickets and still used basically all the time that had been allotted for twice as many talks.

The fact that everyone brought a talk did some pretty amazing things. People interjected and participated in the talks in a way that's hard to get even in unconf formats like the "fish bowl." I've been to a lot of unconferences and even at FooCamp you rarely get 50%-60% of the people in a room actually participating but TenConf was close to 100%.

[wraithan]

This is more of a traditional format thing, but I think it could be good as a contrast. @djspinmonkey gave a talk at the New Relic engineering offsite, as well as at CascadiaRuby titled Be Awesome by Being Boring where he talks about using established tools and not reinventing the wheel. I think as part of NISConf it could be good to have a countpoint keynote. He makes a lot of great points in his talk, that can be good to keep in mind when reinventing wheels.

TL;DR: let's have traditional talk about not reinventing wheels as a keynote, Nothing is Sacred, even NothingIsSacredConf.

[junosuarez]

Along with that last point, it's hard to reinvent wheels when you're also reinventing axles. Are we explorers or simply adventurers? Is our reinvention in the course of a more targeted inquiry or are we tearing at things aimlessly? If we accept that our energy and our cognitive capacity is finite, then we need a way of prioritizing and focusing our experimentation and (re)work.

Being boring in certain aspects of our work allows us to be creative, critical, and free-spirited in others. This affects not only our own work but how we communicate it to other people.

[mikeal]

@jden you basically just explained why I own a Mac and not a Linux laptop :)

[djspinmonkey]

Say his name and he shall appear!

@jden That's a pretty good summary of the talk, actually. :-) I want people to make crazy beautiful awesome new things, but I think those things will work better if you build them on top of the most stable, well understood foundation possible. There's a whole bit at the end of the talk where I basically just wave my hands and get excited about that, but there aren't any slides for that part, since I wanted to talk more directly to the audience.

Anyway, I know this is still in super-early speculative stages, but if folks do end up having any interest, I'd be happy to come give the talk again. I think a video of it will soon appear here, if you're interested: http://www.confreaks.com/events/cascadiaruby2014