Verily

I spent the last few days running Debian, again. It was nice, like a comfy pair of shoes. The time felt familiar, and instructive, but ultimately the heaping piles of small issues, paper-cuts really, drove me back.

Some of the issues I encountered were easy enough to work around, even in a fully satisfactory manner. Others, like the excessive screen tearing, not so much. Eventually, these kinds of issues, plus the sort of uncertainty, and even reliance on the ArchLinux wiki for things that weren’t even for their distro was bothering me too much.

Consistency is a hell of a drug. I know that Windows is a crapshoot, and Apple seems to be fortifying their so-called “Walled Garden”, but damn if it isn’t at least internally consistent. It’s the consistency that I can walk up to pretty much every Windows or macOS device and know the how, the why and the methods to the madness, within a small margin of error. Walking up to any particular Linux (desktop) system, I can’t even be sure that it has the same init system, graphical layer, sound subsystem, packet filtering or even if the userland utilities are the same (busybox, vs gnu vs bsd).

I love Linux. I love what it stands for, what it can do, what it enables any half-way attentive idiot like myself to do with just an investment in time and energy. I also hate that part.

Covering this ground again feels repetitive, but time and time again I’m brought to account for it. I recently was working on my resume for a submission, and the only copy I had readily available was in LaTeX. Not ideal, still, I persevered and got it in shape. However, I know I had also been working on a Word copy that would have been absolutely mangled by LibreOffice (by no fault of its own) and been an ordeal to get working again. It’s this sort of setback and minor challenges that used to excite me. Now they just make me want to stop before I start.

I understand why Windows, and to a much lesser extent, macOS have the lion’s share of the market: Consistency. A fact that I have under-appreciated for far too long.