John Seddon's Bristol University talk

This was a very enjoyable talk from a very good speaker. There was quite a lot of overlap with John's 2010 Øredev conference keynote but it was well worth going. There's something about seeing someone speak in person. The occasional thing they've said before suddenly somehow makes more sense.

Here's what I jotted down, in no particular order:

• MBA - maybe best avoided.

• If you give it a name [eg Lean] then managers will think it comes in a box.

• [failure] demand went up. It should have been a signal.

• Telephone work is not a separate part of the system.

• Understand the problem - build understanding into the system.

• Cost is in flow - not in activity.

• If you measure cost, cost goes up.

• The only plan is get knowledge.

• Culture change is free [because] it's a product of the system.

• Experience is not the same as knowledge.

• Change is emergent. Don't think 'we can't make a change without a plan'.

• Trust is not a point of intervention. It's a consequence.

• You don't learn counter-intuitive ideas in a classroom.

• Taiichi Ohno - "he never explain".

• Don't persuade. Make them curious.

Those last three reminded me of the old chestnut about acting your way into a new way of thinking rather than vice-versa. John's point was that if you are trying to teach a counter-intuitive idea (a new way of thinking) then it's probably not such a good idea to start from the current way of thinking. That's trying to cause a big change in someone's behaviour (something counter-intuitive) by persuasion alone. Much better to get them to experience what you want them to experience directly and then they'll start to naturally think differently without any persuasion at all.

John also emphasized that service organizations can change fast because they don't make anything. That made me wonder to what degree his message applies to software-development (in isolation) since software developers very definitely do make something.