I find it absolutely hilarious that most people with Strong Opinions about systemd — whether for or against — don’t seem to be able to clearly articulate the problem. If you’re pro systemd, you should be able to explain what problems it aims to solve that the SysV-style init system didn’t handle well. And if you’re opposed to it, maybe have a reason beyond “new thing bad”.

For example, I personally think systemd is a bad trade. Parallelization during boot is nice but insignificant for most users. But “let’s take a page from the Windows registry” and manage unrelated configs in one system is dubious at best and IMO violates Unix design principles. In exchange for these marginal benefits — which haven’t even been fully realized in practice — we get all new dependency hell, a system that’s harder to understand (and therefore troubleshoot) with a worse interface.

I wrote this piece about the challenges of #blogging in 2020:

https://andregarzia.com/2020/02/the-hard-part-of-blogging-is-not-writing-it-is-sharing-your-post.html

Everything I mentioned there is still true and worse than before.