Having the core infrastructure for Linux distros be written with a coherent set of design goals instead of being 200 independent projects that can only interoperate via shell scripts parsing and piping output is good, actually
@mjg59 having simple composable tools that can communicate with pipes is good, actually.
@hyc as long as you're not having to pipe it through awk first because the tools all use entirely incompatible formats

@mjg59 I've never seen an init system that relied on awk to make all its scripts compatible, but even so, awk works.

On the other hand, software built with LLM-plagiarized code is *bad*, actually. https://mastodon.social/@kapualabs/116273857383339305

@hyc if you can look at sysv init scripts and think "This is the reasonable outcome of a thoughtful design process" then we are so far apart in matters of taste that there's no point in further discussion
@mjg59 use of AI isn't a matter of taste. https://mastodon.social/@hyc/116274100279140311
@hyc you appear to be responding to something I didn't say
@mjg59 @hyc just two posts upthread you suggested that Howard was promoting sysvinit, which had never been mentioned before that point, so you're one to talk.
@hatzka @hyc it's kind of the canonical system to rely on scripts so yes that's an assumption on my part but I think the same position applies to every other init system that relies on scripts
@mjg59 @hyc I thought Howard was using "scripts" in a broad sense that would include things like systemd units. it's a strictly incorrect usage but that interpretation came so naturally to me that it genuinely took me a bit to figure out where specifically shell scripts had entered the conversation at all
@hatzka @hyc Howard has generally struck me as the kind of person to use precise definitions and so I was responding with that in mind