---
title: "The Myth of Bloat"
date: 2021-06-04
type: "post"
tags: ["Linux", "Rant"]
---


*Originally posted on my [old blog](https://git.exozy.me/Ta180m/blog/src/branch/main/_posts/2021-06-04-myth-bloat.md)*


If there's one thing that hardcore Linux users are ridiculously obsessed about, it's the vague and scary concept of *bloat*. You gotta keep package counts low, use [WMs](/posts/why-wms-suck), and compile [suckless.org stuff](suckless.org/). Visit any Arch or Gentoo forum or chat, and this philosophy seems to be epidemical. But... there's a gaping problem: package count is totally irrelevant, WMs are painful to configure, and suckless.org software really sucks. Yes, it's all a load of nonsense.

Let's start with the term "bloat". What does it even mean? Wikipedia offers a few definitions, such as [code bloat](https://en.wikipedia.org/wiki/Code_bloat) and [software bloat](https://en.wikipedia.org/wiki/Software_bloat) which are definitely legitimate issues. However, this obsession with bloat seems to be about something different; an allergic aversion to any software that's sufficiently complex.

What about package counts?

![Wise words](/img/package-count.png)

It's actually really easy to get a very low package count: just install Windows and you can brag about having zero packages installed. Beat that!

Enough said about package count; The issue with WMs and other supposedly "minimalistic" software is that *it's simply worse than its alternatives unless you use are purely optimizing for metrics like lowest number of lines of code.* Holistically, they're just... worse.

Well, that was quite the rant.