The Process:
1) Decide what you want to achieve.
2) Try different ways of achieving it until you find one that works for you.
3) Do more of what works. Do less of what doesn’t.
4) Don’t stop doing it until it stops working.
5) Repeat.
It is both this simple and this hard.
When your working life rewards you, it’s easy to ratchet up the complexity: homes, cars, travel, possessions etc.
I have found that all that complexity comes at the sake of your most fleeting asset: your time. Instead of building things, all of a sudden you’re dealing with minutiae and logistics. Instead of talking mostly to engineers, you’re talking mostly to non-engineers. The building stops…the business of managing self inflicted complexity begins.
It’s worth noting that the best players in the game (Buffett, Elon) have kept their life extremely basic, almost monastic/nomadic, as success ratcheted them ever higher.
I think it’s the biggest secret hiding in plain sight:
When the world upgrades your status, downgrade your complexity.
For those unaware, SpaceX has already shifted focus to building a self-growing city on the Moon, as we can potentially achieve that in less than 10 years, whereas Mars would take 20+ years.
The mission of SpaceX remains the same: extend consciousness and life as we know it to the stars.
It is only possible to travel to Mars when the planets align every 26 months (six month trip time), whereas we can launch to the Moon every 10 days (2 day trip time). This means we can iterate much faster to complete a Moon city than a Mars city.
That said, SpaceX will also strive to build a Mars city and begin doing so in about 5 to 7 years, but the overriding priority is securing the future of civilization and the Moon is faster.
“There are 3 things that are import: Truth, curiosity and beauty, and if AI cares about those 3 things it will care about us” - Elon Musk @elonmusk
These are great qualities for all of us to appreciate and live by 🤩
https://t.co/7MXQppbemU
"Coding is like... creative problem solving, that maps very well into the real world. I think there is a huge correlation. " - Peter Steinberger @steipete, the creator of OpenClaw @openclaw
https://t.co/cqzWErEi4u
@bcherny could you look at working with @opencode? Their harness has been amazing and it's hard to blend back to the default #claudecode harness. Since they forked claude code could you not find a way to #configureclaudeharness where the features #opencode has are brought in?
@JordanFisherEzr Great insights @JordanFisherEzr especially the approach of timing a product idea with the arc of AI’s growth, almost as though we are building for the iPhone before it exists. I’m taking this as an encouragement to think bigger & imagine what true disruption looks like 👍 🤯
I have just started on working on the context problem with Claude Code CLI using a wrapper. Currently looking at reasonable local memory solutions to assist with the limitations & cost of CLIs connecting to AI APIs. Glad to see I'm looking at the same problem as @simonw .
I think context engineering is going to stick - unlike "prompt engineering" it has an inferred definition that's much closer to the intended meaning, which is to carefully and skillfully construct the right context to get great results from LLMs
“Truth … isn’t a 1:1 representation of reality , rather, truth is something that draws our attention to certain aspects of reality, while inevitably ignoring other aspects.” ~
@harari_yuval in his book #NexusBook. Massaging my mind, thanks for the great read 👌