💾 Archived View for tilde.pink › ~slondr › will-i-outlive-netbsd.gmi captured on 2023-06-16 at 16:43:07. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2023-04-19)
-=-=-=-=-=-=-
I was recently reading a detailed account of Apple UNIX, a Macintosh-compatible UNIX operating system published by Apple for Macintosh computers with m68k CPUs. The account appeared to have been written in 2001, based on one screenshot of the author's PC desktop. The account concludes by recommending that operators of m68k-based Macintosh computers who wish to run UNIX on their machines find an alternative to Apple UNIX for a number of reasons. They recommend NetBSD.
I understand in my human brain that NetBSD is one of those BSD flavors that came around in the 1990s. I know, consciously, that Linux is about as old. It still felt jarring to see this recommendation appear in a 22 year old account of 1980s software systems, though. I've used NetBSD systems, and in using them they feel "modern" as they exist during modern (or at least current) times. The idea that 22 years ago it would have been reasonable to reccomend running a modern-and-current-in-2023 operating system on then-ancient (and now paleolithic) hardware feels like an anachronism of the most severe type.
Clearly NetBSD and other software projects like Linux and KDE (which the author's screenshot showed him using) have quite the staying power. Apple UNIX, even then-contemporary Mac OS and OS/2 operating systems are just gone now. However, I've always imagined that primordial versions of modern-and-current software must feel alien. KDE 2 screenshots certainly look unusual to me as a KDE 4/5 user; I'm vaguely aware of a lot of modern conveniences and recent software creations like Zsh that I use a lot, so there's no way I'd be familiar with the 1990s versions of these software systems. Right?
Oh. Zsh came out in 1990; it's older than Linux. Huh.
I don't have a good word for the group of software that exists which is both older than me and also lively, actively developed, and incredibly current.
There's lots of software that is older than me and existed coterminously with myself but is now dead; all of that stuff feels "correctly" ancient.
But there's an even smaller group than either of those two groups. Software which is older than me, and is still current and updated, but I don't really interact with that often. This tilde server is running NetBSD so maybe it shouldn't qualify... FreeBSD is probably a good example. FreeBSD is older than me, and I don't run it on any of my machines. This makes it feel like ancient software, but it's not! It's very modern-and-current, it's actively being updated quite a bit. There is a chance that it will outlive me, depending on how soon I die and how much momentum the project maintains in the future. This chance exists despite the fact that it came first.
Either possibility feels weird to me. These 1990s open source software projects are fixtures, existing in a recognizable form despite a world and computing universe that has changed unimaginably since the 1990s, so clearly they will never die. However, 29 years is a very long time for any single project to exist; it really shouldn't be so unimaginable that (hypothetically; I have no idea of the internals of any of the BSD projects of course) some core group of passionate spearheaders may retire or slowly drift away from the project and eventually it lacks the steam to fix critical bugs or what have you, and the project dies.
The best example I can think of of something like that happening is Crunchbang Linux, which I actually used for a short period of time as my main daily driver distro back in 2012 or so. That project died a fast death due to the creator abandoning it; there are some clones around, but none of them seem to have the same energy. I never used any of the clones, and I bet that's true for many other users as well.
As awesome as it was, Crunchbang wasn't an operating system - it was basically a highly configured custom flavor of Openbox on Debian, and as such it could be held up reasonably by one person for a few years, dying when that person left. A full operating system like FreeBSD, NetBSD, etc obviously needs a lot more manpower going into the project to stay viable, and once that happens and the system joins the relatively short list of operating systems viable for production use, maybe (maybe!) there's a snowball effect which leads to increased manpower dedicated to projects which get onto that list that keep those projects alive. I have no idea how reflective of reality this is, though.