I recently watched a guest lecture Alan Kay gave to Paul Pangaro’s students in Detroit (while on a long bus ride from Toronto to Montréal). I started watching while noting down words but ended up transcribing pretty much the entire talk - every sentence felt like it had a special weight, and could be unpacked in many ways. I think my notes will make much more sense if you actually take the time and watch the talk - it is really worth it.

  • distinction between internal and external tool

  • distinction between prosthetic and an amplifier

  • don’t design an interface if you don’t have a curriculum. The interface will suck otherwise.

  • Smalltalk’s desktop is a full media system - any object from any desktop can be moved around. That’s a good model for what the Internet ought to be.

  • all great ideas in UX came from people who were doing supercomputing, because they were building both hardware and software.

  • RAND Corporation researchers were staying at work late hours and going through people’s trash bins to see what their waste paper showed they had trouble making (and that was flowcharts). So they came up with Grail (GRAphic Input Language) that did exactly that.

  • Manhattan project was first time engineers and scientists worked together prioritizing progress over anything else.

  • “If you change the context then you boost the IQ by 80 points.”

  • “Cosmic goodness intuition” on how the idea for Dynabook came to him.

  • What is the key to inventing the future? The Wayne Gretzky effect (”you lose all the shots you don’t take). Here is Alan’s guide for us:

  • Chase the “cosmic goodness intuition”.

  • Find the favorable exponential (Moore’s law) - something that gets faster by doing one thing (it was making things smaller in that case)

  • Take the cosmic intuition out 30 years

  • Ask the question of whether it will be ridiculous to not have it?

  • If the answer is YES - then bring something back from that to about 10-15 years. What can we really do in that time?

  • Spend money to make flexible super-things that can simulate and explore multiple conditions. You get to do so many user experience simulations without a lot of cost (for Kay it was Smalltalk)

  • If we then want to optimize, we can make a tool for the future.

  • Be against the “release immediately” culture. Even if something is successful, it is very tough for it to become better.

And some highlights from the closing remarks:

  • Quality of funding equals quality of results (and there has not been good funding since the PARC days)
  • “The key problem of our society is growing better adults”
  • “In America being good is to same to being rich.”
  • We should still direct our efforts into building technology for making education-oriented environments for children.
  • Why not have the AI teach a child to read, but have an AI play music for you? (that’s my own rhetorical addition).

As I said above, there is so much to unpack. I am trying to keep a balance of reading, writing and actually starting to make things, which is tough. My personal interest lies within the context of the Internet as a medium, and whether we can apply Kay’s principles to a better web. The reason why I like listening to the creators of the field is because I need to know when things went wrong - and whether we can revive some of the original ideas and ideals in the post-net neutrality world of information.

link to video