I've recently been playing with Uxn[1], a small virtual machine designed by 100 rabbits[2] with minimising software complexity as its highest priority. It is so simple that given the specification a hobbyist could reasonable implement the assembler and emulator on any new device in their spare time. As such, the programs are highly portable and will hopefully continue to be so as the era of personal computing moves on. It's a very interesting idea.
In that regard, it is much like gemini. Given the spec[3], a hobbyist could reasonably implement a client and server, requiring only minimal use of computationally complex software (the debate around what should be required is a discussion for another time i.e. cryptography).
Similarly, it is one of the reasons why I love the ed editor, despite it's absurd impracticality. Between Michael Lucas's book Ed Mastery[4] and the gloriously simple man page[5], I pretty much have the entire spec memorised and could, over a few months of evenings and weekends, implement my own version. Neovim is so complex it has required raising funds to pay a fulltime dev, and that's without starting on the current era of editors that all seem to require an entire Web browser to edit plaintext.
Ten years from now when the Web has inevitably been replaced with some new hot thing (metaverse, maybe...?), how likely is it that some group of hobbyists will still be maintaining some Web browser for us all to access the culture locked up in today's websites? I think it very unlikely. Modern websites are huge projects, only reasonably undertaken by companies with millions to throw at them. And that's without even mentioning the websites that we've already lost; culture of the late 90s and early 20s that is nothing but memories in the minds of those that were there.
And it's not just software. How many small-time films, TV shows, and music albums will be lost to history because their distribution methods are no longer accessible. Similarly, will today's cars ever become classics? How will you replace the parts when the computer says "no" because the manufacturer has gone bust? Will the museums of the 22nd century have anything to show but piles of dead, lifeless plastic?
I don't think the right-to-repair campaign is a silver bullet either. It still relies on the companies existing, to provide such parts. And it does nothing for software. The Web is an open, distributed platform, and yet I fear it is just as fragile as anything else. Open and distributed are necessary, but not sufficient. I'm beginning to think that we also need an aggressive and uncompromising pursuit of software minimalism to ensure that we are not beholden to the continued existence of profit-driven corporate entities for our culture to simply not vanish.
Last Updated: 2021-11-11