2026-02-10
Minimise lines of code to avoid being controlled, not to increase maintainability
Disclaimer
Main
- Multiple people seem to have a heuristic that says that open sourcing your code is not enough, and making it cheap to host is not enough, but you also need to minimise lines of code.
- I've increasingly realised there's two different schools of thought for why.
Some examples of people who minimise lines of code to avoid being controlled
Some examples of people who minimise lines of code to increase maintainability
- Linus Torvalds
- Jonathan Braid
Both lists are likely incomplete, biased etc
There is commonality between both goals. Both are about actually thinking long-term, like 100 years into the future, instead of just 6 months into the future. But I think the first group is more concerned with their project dying and no one restarting it. (Or in some cases, themselves literally being killed/imprisoned and no one restarting the project.)
I am personally more sympathetic to the first group's concerns. Ultimately though, I think it's possible this heuristic ends up not mattering at all when it comes to the ASI problem, and I am still holding on to it for deadweight reasons. I haven't yet made up my mind either way on this.
[1] Side Note
- I used to be more sympathetic to Moxie's criticism back in 2021 if I remember when I was actually leaving web3, but now I find it slightly more complicated.
- He is completely correct about the reasons why centralised walled gardens achieve product market fit way faster than open source distributed protocols.
- He is also probably correct that most of these people are probably Losers of History, in the sense that all their heroic efforts ultimately haven't so far changed the reality that basically only billionaires and politicians have power in society.
- Like Moxie, I too seem to like the Losers, for reasons I don't fully understand. I think I should atleast deliberately try to like Losers a bit less, if I actually want to Win here.
- See also: Deliberate status engineering is hard
- I'm wondering if our reasons for thinking they are losers is slightly different though. I'm saying moving slower than everyone else is a problem because ASI might be coming and that might permanently end the game.
- Otherwise, if ASI wasn't coming, it seems fine to me for the centralised walled garden to iterate faster and for the open source distributed version to iterate slower.
- For me, this is just how capitalism works. Obviously groups of people being paid to do a thing are generally going to move faster than people who get paid nothing.
- Eventually though the open source version could catch up, right? At some point you might run out of new creative ideas within some industry niche, and the centralised walled garden version of this final app will get built, and then later the distributed open source version.
- For instance, livejournal was launched in 1999 and substack was launched in 2017.
- It is not obvious to me that substack is that much better a product than livejournal. It is better, yes, but did this improvement really take 18 years to find? Or is it just building, breaking and rebuilding the same thing over and over again.
Subscribe
Enter email or phone number to subscribe. You will receive atmost one update per month
Comment
Enter comment