What I have learnt, said out loud so it can be tested.
A guide is for looking something up. A talk is for putting an idea in front of people who can interrupt, and finding out what it is worth. Each one here presents as a deck and reads as an article afterwards, so the version you hand on is the same object the room sat through.
What a talk is here
Each one takes something I have spent a while learning and says it plainly enough to be disagreed with. There is a claim, the evidence behind it, and a point where the subject turns from the material to the people listening and what they might do with it on Monday. Every number traces to a source. Where a finding is contested, the talk says so, because a room that leaves persuaded of something unsound has been made worse off.
Each is one file that presents as a deck and reads as an article. The version you hand to someone who was not there is the same object the room sat through, and it can be marked up, argued with and passed on.
Why I give them
Two reasons, and the selfish one comes first. Nothing sharpens a half-formed idea like explaining it to people who can interrupt. Building a talk forces me to find out which of my convictions have anything behind them, and standing in front of a room finds the rest. We learn best by teaching. I would rather learn that way, in the moment, than in print a year later.
The second reason is that the ideas are not mine to keep. Most of what is in these talks came from somebody else’s research and somebody else’s mistakes. Passing it on with the sources attached is how a field gets to think collectively rather than one person at a time. The best sessions end with the room correcting the speaker, and the next version of the talk is better for it.
So this list will grow. If you were in the room for one of these and have a disagreement I have not heard yet, I would like to hear it.
Available
The talks
Each opens in its own tab. Press R at any point to stop presenting and start reading. It is the same document either way.
Good work does not speak for itself
Most work is reported in the order it was done, and readers need it in the order they will use it. Communicating the work is not packaging added at the end. It is the last stage of the work, and a piece is unfinished until it can be stated claim first, for a named reader, at a stated level of confidence. Eight moves, from naming the reader’s decision to checking what they actually heard.
Open the talkDesigning for learning
Ease of use and ease of learning are different things, and they often pull against each other. Product instinct trained on shopping, streaming and feeds will mislead a learning system unless it is checked against how people actually learn. Seven things that are true about learning, where each collides with design instinct, and a method for turning evidence into principles a team can defend.
Open the talkWhat design is, and what designers do
Design is a category of decision that sets what a system makes easy, hard, and impossible — and most of it is made by people who do not hold the job title. What the decisions cost when nobody owns them, what the evidence says design is worth, where that evidence falls apart, and the question each role around the table is actually responsible for.
Open the talkZero is an average
Forty years of research finds that educational technology has an average effect of roughly nothing. The average is the least interesting thing about it — the spread around it is enormous, and design accounts for a good deal of the spread. What that means for what to build, what to measure before it ships, and what to stop citing.
Open the talk