A place for developers to talk about building on Farcaster.

/dev

Looking for keystone reads on reducing complexity in software engineering. Blogs and essays, not books.

To me, reducing complexity means removing non-essential capabilities and/or adding graceful abstractions.

What’s the prior art on this?

Complexity is inversely correlated with control, which in this context means understanding or comprehension. What you want to look for are basically two areas. 

1) a universally useful way of framing a problem
2) a solid understanding of the core mechanics

As for 1) I would like to suggest Gravitropic Topologies: http://localhost:8080/Phenomena/Gravitropy. It would be great to get content there for the software industry. I am happy to collaborate on this if you are interested.

As for 2) this is use case specific and I think nobody can help you with that. You have to put in the effort in order to understand the problem that you care about really well.

in motion | founder, researcher, scout, storyteller | into computing, communication, complex systems | rewilding the web with subset.network | hq: msweet.net

Gravitropism looks pretty cool! Maybe complexity isn’t the right word for what to reduce, esp if I’m assuming one has a useful frame and understanding of the core mechanics. Perhaps thinking of increasing compression, instead.

This is the alt account of Sister Hazel. Writings at https://powerlaw.systems. Curating https://protocol-particles.github.io.

Gravitropism looks pretty cool! Maybe complexity isn’t the right word for what to reduce, esp if I’m assuming one has a useful frame and understanding of the core mechanics. Perhaps thinking of increasing compression, instead.
https://arxiv.org/abs/0812.4360

Gravitropics are pretty cool indeed. Thank you.

The terminology aside. We need to distill understanding in order to find the fundamental elements of a problem statement.

Gravitropy frames those elements as centers of gravity, which fits vaguely into the idea of compression progress.

Whether we reduce or compress, it's all the sides of the same coin.

Summer of Protocols

We can move to /sop if you like. I am also in Discord.

My take is, you wrangle complexity by increasing compression, to use your terminology. But that only works by understanding the dynamics under the hood. In my words, identifying the gravitational centers and their entropic interfaces provides you with the understanding of how to cut the cookie.

You asked initially about software. So there you define the interface. And once you know the system, you can replicate an algorithmic version according to your use case.

Gravitropic topologies have gravitational centers, which are the points at which behaviour is exercised and influenced. Those centers of gravity do then also have lifecycles in software terminology. Understanding lifecycles helps greatly to define software interfaces. All of this would be cool to add to https://protocol-particles.github.io.

Okay, but practically—imagine you have a product or entity to steward—how do you wrangle complexity or increase compression?

Beginning to think that it comes down to having better *interfaces* to the underlying system, rather than trying to intervene in it directly.

*cough /sop *cough