Build up, stupid.

Learning to fly by avoiding the ground

This will be the newsletter I use to document my progress in my work on /vanjacloud/. I already have a webpage which will act as a landing page / entry point for people mildly curious about the endeavour - more like a placeholder art project installation - but this page/newsletter I will use to stream-of-consciousness through some of the bigger challenges and ideas I’m exploring. I hope this ends up useful or interesting to people in the builder community, or anyone adjacent who might be wondering ‘how do you even end up with something like that?’

First thing first, the intro: my goal with this project is to end up with something cool and useful and usable by ‘building in reverse’. I don’t know what I’m building, and that’s part of the appeal. I like the creative and surprising aspect of building something and not being sure what form it will take in the end. But I also want something useful, so I’m trying some novel techniques to make it all happen.

1. Build up, stupid

A well known concept in software development is the idea of ‘dogfooding’. If you build something, you want to try get your own builders using the product so that they can really feel the pain understand the end user. There was a famous story of Microsoft creating a product for developers, but not using it internally for their own development, so the product naturally sucked and everyone hated it. A clever executive came in and declared that the project must ‘dogfood’ itself, and suddenly the whole org was in a lot of pain, dealing with garbage software. Funnily enough, the software got much better.

I’m building on that idea by dogfooding my own software, and I’m making it EXTREME by dogfooding software that doesn’t even do anything yet. What will it do? I don’t know. What I /do/ know is that I /will/ be using it, and to be considered a success I will want to be using it /a lot/, so the first thing I did was create a little electron menu bar app and bound it to a hotkey. Any time I press cmd-shift-\ my app pops up. Now I have a great place to put cool stuff I build.

The second and third thing I did was to put a similar app on my iPhone and in my CLI shell. Anywhere I am, my app is ready to handle doing something for me. I’ve recently started watching Marvel and I’ve found this strategy reminiscent of Ironman’s Jarvis.

(I also put a copy of my app on Azure cloud, and I started ‘connecting’ it to my services such as Notion and Twitter etc. As soon as I think of something neato, I’ll have a place to put my stuff and quickly iterate. This is IMO key to getting stuff done; fast feedback and fast deployment.)

2. Learning to fly by avoiding the ground

The second technique comes from a rationalist blog I enjoy, and the premise is that to be able to fly, you just have to practice not crashing into the ground. It’s a bizarre way to look at the world, but you can’t really fault it. Using a Test Driven Development style framework of achieving goals you end up finding ways to be 1 second off the ground, then 5 seconds, then 15… slowly iterating until you get to the point where you can actually fly.

Does this waste a lot of time building things that don’t scale? Sure. The solution for ‘flying for 1 second’ is very different to ‘general flying’, but at least it gets you out there doing stuff and building momentum. Momentum is good.

3. Three things are visually appealing

I have a certain taste for how I want things to function, so there’s a good chance I’ll just end up rewriting common tools I use into my own app until something kind of ‘sticks’ and is worth working on. As soon as that happens, I can spin it off into its own app and publish it to the app store or whatever. I’m only 10 years late to this goldrush!!

---

Well, that’s not a bad intro post to what I’m hoping will be a long and prosperous project. I know reading long form posts can be boring, so I’ll do my best to summarise below:

Start today, can’t you?
Finish tomorrow, no chance
Go do it anyway

Fin.