I saw Catmull's keynote at SIGGRAPH 2008, and it was spectacular. It had some common elements to this one, but was different overall. What really struck me was how he is not quite a manager, nor a technical person, nor a creative person, but somehow an "organizer" for all 3 of the above groups to work together to create magnificent works of art.

I think in large part this comes from his technical background, which is simply incredible: student of Ivan Sutherland's (of Sketchpad fame), inventor or discoverer of z-buffering, antialiasing, texture mapping, subdivision surfaces, b-splines, and key developer of Renderman!

(For those not in computer graphics, any one of these would be sufficient to give you lifetime fame.)

This is a great talk and I recommend you watch it. Here are the cliff notes.

- Constantly review

- People and how they function are more important than ideas; ideas come as a result

- Do not let success mask problems, do deep assessments

- Organizations fall over by default, slowly enough that you don't notice it

- Everything you do has to be original, you can't repeat yourself, dig deeper, you're always missing something important

What strikes me is how they've internalized successive originality in each project. The latter half of their repetoire is something new in animation: Wall*E, Ratatouille, The Incredibles.

Correction: -- Organizations fall over by default, slowly enough that you can fix them if you care. -- Ed did NOT say that because organization fall slowly it's hard to notice. He implied that you have time to fix the failure... you just have to put conscious effort to notice the problem and then another effort to fix it (while organization is still successful overall).

He says "... organizations are inherently unstable, they will fall over, you have to work to keep them upright, but they fall slowly, most people don't notice it, they let success blind them." -- roughly minute 34.

I wouldn't want to misparaphrase him, but I think my notes are accurate.

22:20

"The other thing to note is that there is a confusion that people have, because the books and the press kind of work this way; We think about 'an idea'. We think about ideas for movies, ideas for products, and its usually thought of as some singular thing. The reality is these successful movies and these successful products have got thousands of ideas. There's all sorts of things necessary to make it and be successful and you have to get most of them right."

Pure gold.

Wow, this is great, thanks for posting!

The 1980s were a fascinating time in computer graphics, and there are few people who are better qualified to talk about it. Ed Catmull practically embodies the history of the field, starting out at the University of Utah, moving to commercial work and struggling for years before finally achieving the ultimate success at Pixar.

I also find him to be an excellent role model, as someone who worked over many years to achieve a real and lasting impact, very much different from some the lucky entrepreneurs of the last decade who receive so much undeserved praise.

This guy is so cool. He's like world class, a pioneer, a revolutionary. And yet he's talking about all this stuff as if he was digging ditches... that he's really into. He cares about it for its own sake.

The way he describes that story-teller brain-trust at the start (trust, necessarily honest) I think also describes him.

"make mistakes and learn from them".

So obvious, well-known and true... but it has a certain reassuring omph when someone at the absolute top of their field says it, with examples from their experience.

It doesn't need to be right the first time. In fact, it can't be right the first time, because there's things that you don't know - and can't know - til you make those mistakes. He said they're "failures, but that's not quite the right word". Another word is "experiment", but that doesn't capture how much these mistakes hurt. They are real.