superfluous is a node.js component framework build around page delivery. it’s main feature is that it supports isomorphic rendering, as well as bigpipe style delivery and just-in-time multiplexed assets that get stored in LocalStorage instead of the browser cache.

The goal with superfluous is to make building an app as easy as possible while delivering the page at break-neck speeds.

On the whole, superfluous has served me well for 5+ years, but I’ve recently been moving away from the node.js ecosystem because of how much churn there is. I want to write an app that works today, tomorrow and 10 years from now. With node.js, sometimes an upstream packages breaks, but I don’t find out until I try to deploy a server from scratch, since once I have a server running, I rarely fiddle with it. This can lead to a program being uninstallable for weeks or months until I learn about the breakage and fix it. (I know package pinning would work here and I do use npm lockfiles, but that leads to security problems)

To that end, I’ve been working on pudgy components - trying to re-create the experience of superfluous in python.