Iterate as fast as possible

Have an aptitude for serendipity

Get more parts on the table

Stay motivated with recurring small dopamine hits

Incremental correctness

Keep it simple, stupid

Measure output, not outcomes

Non-technical regrets at Spectrum

Solve your own problems

Design a process

 Make it easy to jump in

# Always be shipping

The more things I ship, the more chances serendipity has to strike one of them. ([[Have an aptitude for serendipity]])

If you never release that song you composed, how is a music label meant to discover you? If you never release that novel you wrote, how is a publisher meant to stumble upon you?

This doesn’t just apply to big projects, since I try to [[Iterate as fast as possible]] any small change counts, which also makes me [[Stay motivated with recurring small dopamine hits]].

> _”I like shipping things, and shipping consistently has helped put me at the forefront of a number of big shifts in technology over the past few decades." - Seth Godin_




# Incremental correctness

Incremental correctness is a concept I first heard from [Guillermo Rauch](https://twitter.com/rauchg) and internalised after [Max Schoening](https://twitter.com/mschoening) repeated it often enough. Essentially, start off with a minimum-viable product and then [[Always be shipping]] small, incremental updates. In theory, if you [[Iterate as fast as possible]] this the fastest way to build something great from scratch.




Always be shipping

Invest in force multipliers

Build a diverse network

Be as cross-functional as necessary

# Iterate as fast as possible

The more iterations cycles I can go through the faster I will arrive at a great result. ([[Incremental correctness]]) It also helps with [[Always be shipping]], if you have lots of iterations you have lots to ship. Iteration cycle speed is a force multiplier, and keeping it as fast as possible is crucial. ([[Invest in force multipliers]]) 

Some ways I improve my iteration speed are [[Invest in force multipliers]], [[Keep it simple, stupid]] and [[Be as cross-functional as necessary]].

For example at the micro level, I make sure my build setups are fast enough that I never have to wait to see changes in development. For example at the macro level, I try to give my users a quick and easy way to submit feedback and for me to see it.

Hunch: Anecdotally, I think the 80/20 rule applies to iteration speed bottlenecks. There seems to be bottlenecks which will speed up iteration cycles 80% without a lot of work, the hard part is figuring out where they are.

> _“Speed of developer iteration is the single most important factor in how quickly a technology company can move." — Paul Biggar, founder of Darklang and CircleCI_




@mxstbr's working notes

Incremental correctness

Links to this note

What do you think?