Future of Coding


Day 1

Jan 15, 2018

Wow, I’m not even sure where to start. I guess I can summarize the day.

I arrived at 1pm. Luke gave me a tour of the space, then I got lunch with Toby, Bret, Josh, and Luke. Back at the space, I walked around, reading the posters on the wall. That was fun and overwhelming – there’s so much content that I didn’t know about.

I was finally saved by Virgina who was working on projecting an image of survey results onto a table. We first had to figure out how to project an image onto a table, which is called a “supporter” and scale its size. Then we decided we wanted it to mirror itself across the table so that people on either side could see the table. In order to do that, we needed to change the way we were displaying the image to use the Illuminator library directly, instead of using an image helper function (which didn’t have rotation). It was a little fustrating to have to use the standard canvas that draws things from the top-left corner, with a context, etc.

I have a few project ideas:

I also want to see what ways I could help with fundraising. Here are some ideas:

Also, Luke said that it could be possible for me to host an event in the space on Sunday, my birthday. I want to double check with them tomorrow and will then see if there are people around. That could be a blast hackathon birthday!!

Day 2

Jan 16, 2018




There are a few really cool pages in Dynamicland:

I think it’d be fun to come up with more of these “meta-pages.”

One idea is to mimic some of the concepts found in programming langauges:

It seems like there are some basic Dynamicland operators you’d want:

And there are a few FP concepts that feel relevant:

Here’s a problem that a meta-page might be able to solve: when you dirty edit a piece of code, all it tells you now is that it was edited and on what line. It would be neat if you could:

1) temporarily revert the code to it’s non-dirty state 2) forever revert the code 3) print out the dirty version as a clean version 4) see the dirty code changes (without going to the computer)

Number (4) here is probably the most useful but might be a little tricky given the projector’s resolution. Might need to make it bigger so that it’s readable. Here’s an idea: it could be a cardboard cutout that’s the size of the editor but more mobile. It could also have a way to scroll it up/down and left/right. Maybe by moving a rock through a minimap?

I’m not entirely sure how state works in this system. So far everything I’ve defined has been a pure function of time.

Things I saw yesterday that I will look at for inspiration:


Day 2 was a little less overwhelming, but also jam-packed.

When I first got in, I spent some time reading as much as I could of the source code of the Realtalk engine (big board). This was fun and seemed like a good way to learn about things – as opposed to looking at random examples. Then we had lunch. I chatted with Luke about his prior experience starting two companies, and how he was intitally so excited about VR, but now, like other early-VR enthusiasts, he has trouble putting on the goggles. Additionally, he mentioned how far away the technology is from being anywhere near good enough quality.

Then I spent a lot of time with Josh, who explained how the macro/pattern matching system works, the backend architecture (Realtalk, the webserver, the text-editor, the filesystem), including printing out the compiled Lua code that is generated. He also showed me the “Addition Kit” which I LOVED, as well as Editor in the World, which was also really cool. He pointed at a few other “kits” that I’m excited to check out.

Then Luke and I spent an hour or so talking about fundraising. That was fascinating, however, I don’t feel comfortable publishing any of details there.

Day 3

Jan 17, 2018


I don’t have any particular inspiration on what to build yet. I still feel like there’s so much to see and tinker with. Here are the things that most interest me:


I got to the lab at 10:30 this morning. First I played around with various kits: the math kit, the fractal tree kit, the logic kit, Paul’s blocks kit, etc. Then I played with the RGB dial selector, but it was fustratingly difficult. Then I picked up poke with Toby and Bret.

Bret meets Alan

Over lunch, I asked Bret about how he got connected to Alan. Here’s approximately the story:

I made a website and one of Alan’s researchers showed it to him. Alan often dismisses things when he firsts sees them – “we were doing that back in the 70s at PARC” – but if you keep showing him somthing that he needs to see, he’ll eventually get it. They showed him my stuff a few times and we set up a time to meet. It kept getting pushed back because Alan had meeting with people like Barak Obama - this was in 2011 - but eventually I went to visit the lab. They were working on STEPS at the time. I stayed for a few weeks at the lab.

After lunch, I perused the books in the library. I felt a little overwhelmed, not really excited about any particular project. I decided that I’d made a simple “dot catching game.” That was fun and took 45 min or so.

Math Kit v2

I was then saved by Virgina again. She wanted a way to do “spreadsheet like things” to data in the system. So first we needed to figure out a way to represent numbers. What eventually worked was using a “poker chip” style, where you count the number of different colored chips on a page, greens are ones, reds and tens, and blues are hundreds. That was fun.

Then Luke showed us how to select a group of blocks, so it became clearer how to create aggregation functions. I made “sum” and “count”. Then I modified the Math Kit so that it works for division; it used to only work for addition but now it works for +-/*. I combined sum, count and division to get the average of numbers.

Things to think about
Things to definitely do:
Things to maybe do:

Day 4

Jan 18, 2018

I got in at 10:30am this morning as the same time as Omar. I showed him what I had been working on, the game and Math Kit v2. There were a bunch of volunteers in the space today making things.

Math kit progress

I spent the whole day on my math kit today. I spent a long time refactoring the code I wrote the day before to make things more stable. I refactored the numbers to the Math Kit library, so that you only need to claim that something is a number in order for it to work with the system. I also lowered the opacity on all the colors used in the system so as to less interfere with the dot detection for my “poker chip numbers.”

I also refactored sum_selection and count_selection to the Math Kit library. Then late in the day I started to make simple mapping over the numbers work. But before that, I spent 30ish minutes re-writing standard functional programming utilities so I don’t have to do dumb Lua for-loops anymore. Basic mapping works now, and you can do cool things, like map over the list and divide each item by the sum of the list.

The highlight of my day was when Omar used the “is number” convention that I made up to create his x-scrubber component. His component worked seamlessly with mine. For example, we hooked up him scrubber which goes from 0 to 1 to a *5 and we got a scrubber from 0 to 5. Amazing!

Math kit ideas for tomorrow

Other ideas for tomorrow

Take photos and videos of what I made!!!

This is really important but shouldn’t take more than 30 min.

Chat with Bret.

I asked him today if we could talk tomorrow and he said he was mostly free. What to chat about? Basically explain that I’m looking for my next move and would appreciate his advice on how to figure it out. I’m considering doing research, possibly hanging out here if you’ll have me, or I could help with other things like fundraising. The standard thing is to start a startup, but I’m not sure that’s for me. If I were to go down the researcher route, how would you reccomend I go about it? If I wanted to follow in your footsteps, what would my first steps be?

Day 5

Jan 19, 2018


I’ve been at the lab since 10am this morning. Being the first one here, I accidentally set off the alarm. That was a fun start to the morning! I spent the last two hours responding to emails and booking flights home.

My priorities today are:

  1. Chat with Bret. I think my notes yesterday are reasonable places to start.
  2. Get confirmation about my birthday in the space.
  3. Take photos and videos of what I made
  4. Make what I made practical somehow (with Virgina), including making a binder / documentation and refactoring the code to one page (or two one-pages)
  5. Add extra capabilities.

Extra capabilities

This is the least of my priorities, but it is the only one I can work on until Bret or Virgina get here. Of my list from yesterday, here’s what I want to work on, in order:


Today I accomplished all of my priorities as well as the “extra capabilities.”


Chat with Bret

I finally worked up the courage to chat with Bret. I think we spoke for an hour or so. I took some notes but will need some time to organize and proccess them. Hope to release soon.

Things for tomorrow

But the most important thing for tomorrow is: thinking about organizing activities for my birthday!!!


Convo with Bret

The conversation got off to a rocky start. I went in asking for very general advice, and Bret wasn’t sure how to help.

One thing he said: you can’t just decide to do what Nicky Case is doing. Nicky can’t not do what he’s doing. There’ just a manic energy there.

He also reccomended the documetary The Corporation and Noam Chomsky’s work for understanding his perspective on capitalism.

He explained his philosophy for prototyping leading to buildng a platform, and then prototyping on top of that platform.

He mentioned that Alan Kay is one of the greatest philosophers of all time, yet people seem to think he’s the OOP or overlapping windows guy. He explains Alan best at the bottom of Links 2013.

Probably the most helpful thing he said was:

Everyone I admire didn’t do anything impressive in their twenties. Papert in 50s. Alan in 30s. Tufte in 50s. Jack Goodie 60s. Twenties is time for breadth.

He also said to optimize for “feels important” or “there’s something here and I need to know what it is.”

What’s Alan Kay’s mission?

medium where children grow up to adults to take care of society by being fluent in systems

What’s Bret’s mission?

Ditto with Alan’s mission, plus continuous with human life, humane.

Another really important thing that Bret said:

You can’t just improve coding. You need a broader mission. Everyone who’s improved coding, Papert, McArthy, Alan, has done it with a broader goal in mind, and the technology fell out of that.

Great Alan quote:

Sure we can make the $100 laptop, but I can’t make 1000 great teachers to save my life.

His answer is AI. Bret believes that converstations should only happen with humans, not AI agent. Alan is into the agent. Bret is Ok with AI in environment, just not agent.

Bret also referenced “shit crayons” a lot. I’m glad I read that piece.

Kids were going to always make things dance. Like in a sandbox. We need an environment for teaching non-human universals like science. That’s why Etoys is clearly the best.