Just over a week ago, on the second day of ScienceOnline2013, I dragged myself out of bed, onto a bus, and eventually made it to the the coffee table at the McKimmon Centre in Raleigh, North Carolina. Jetlag, combined with a couple of late nights and early mornings, had finally hit me. Without a constant supply of caffeine, I wasn't sure I could keep my eyes open.

I realised a few minutes later that I really shouldn't have worried about staying awake. One of the speakers opening the proceedings that morning was Baba Brinkman. Imagine this song, turned up to 11, being played to hundred of sleep-deprived ScienceOnline attendees.

It was basically the perfect conference opener. And regardless of what you thought of his performance, it was hard to escape the idea behind this particular track: performance, feedback, revision.

The idea ran from the workshop I attended on Wednesday before the conference officially started, to the wiki for planning next year's sessions that was populated before most of us had even left Raleigh. Trying something out, seeing if it works and then trying to improve on it ran through the core of a lot of the week's conversations, and the conference itself.

I know the essence of "performance, feedback, revision" is not a new idea. But it is one that I find easy to forget, sometimes.

Scicurious and Zen Faulkes' session on "Blogging for the Long Haul" (Storify) showed that it's not just me who sometimes gets paralysed by their own perfectionism. Lots of people in the room seemed to share a nagging feeling that says the longer you leave your blog un-updated, the better your comeback has to be.

Bora Zivkovic had some good advice for anyone suffering this kind of perfectionism: post something small. A picture, a video, with a few lines of commentary. Make it something quick. Do this a few times and, hopefully, you'll get back into the habit. Your grand return doesn't have to be as grand as you think.

Even before the conference had really gotten going, the seed of "performance, feedback, revision" had been planted in my head during the workshop on rapid media prototyping run by John Pavlus and Rose Eveleth. If you follow this blog you'll have seen what came out of that session (from me).

What I took away from that afternoon was the idea that, most of the time, if you want to make something, you probably can. Don't worry about finding the perfect tool, or the perfect idea. Try something and see if it works. If it doesn't, never mind, move on to the next thing. If it does work: get feedback, make a new iteration. And then another.

Now the conference has ended, but, even as I type, people are editing the #scio14 planning wiki (now renamed the #scio13 debrief page) taking things that they learnt, or considered, or felt was missing at this years conference and putting together programme suggestions for next year's. And conference organiser Karyn Traphagen is thinking about how the conference format can be better. And, of course, I'm not the only one that is thinking and posting about what I learned at this years conference and how I might use it. Performance, feedback, revision...