Pylimitics

Simplicity rearranged

unmonetizable content since 1997


Resilience and Fragility

Distributed systems are more resilient. Natural systems, from organisms to ecosystems, are distributed. There’s a lot of redundancy; even redundancy that doesn’t appear, at first glance, to be redundancy.

Centralized systems are fragile. They have a single point (or a small number of points) that can bring down the whole system if they fail. I should say, instead, when they fail. Because everything fails eventually. Breaks. Wears out. Stops. Is faced with unexpected and overwhelming conditions. There are a million ways to say it because there are a million ways failure can happen, which is a rhetorical flourish but is really an understatement.

The US government was designed to be distributed, and thereby resilient, and so far it seems to have worked, even if sometimes it has felt like it was on the brink. The US economy wasn’t designed in the same sense; there is no written economic constitution. But the distributed nature of the governing system, and the heterodoxy of the population, have contributed, I think, to an economy that has been relatively distributed.

Now, the US economy is certainly subject to cascading failures; we’ve had something on the order of fifty recessions and depressions, and if you look at smaller scales than “the whole country,” there have been many more. Drive through the rust belt and count the economic failures, if you can. But the system itself has been resilient, even if it doesn’t function in a steady state. Maybe, for all we know, human collective action is inevitably cyclic, and there are necessarily ups and downs. I don’t know, and for all the defensive claims that economics is a “science,” nobody else really does either.

There’s a looming problem though, and we’ve just recently seen a few effects. The Internet, which was originally designed to be a distributed, resilient system (that could survive a nuclear war), has been centralized. Not in one place, but in far too few places. The economy has become pretty centralized too, with AI investment (which is nowhere near paying off) being the single point of failure. That centralization is not just about money; it’s about the small handful of rich white guys holding the reins. You know who I mean. And that’s the problem: you know who I mean. There are far too few of those would-be geeks, and they all sound pretty much the same.

When only one set of opinions holds sway over a human system, it’s going to crash. That’s not speculation; that’s history. Data. The tech bros claim to like making data-driven decisions, but if you examine the decisions they actually make, they’re driven by cherry-picked factoids, not unbiased data. They also claim, some of them explicitly, that they’re smarter than everyone else, and hold up all their riches as evidence. They’re not. They’re just believing their own press releases. The only things they might excel in are personality traits like domineering, bullying, and a certain flexibility in the ethics department.

We seem to have unwittingly arrived in the midst of a whole set of major, global-scale systems that are no longer resilient, and getting more fragile all the time. What’s going to crash next? Banking system? Transportation? Communication? It’s only a matter of time before there’s a satellite collision in orbit, and the debris could cause a cascading chain of orbital destruction. Not necessarily, but don’t ignore Murphy’s Law. That would impact all the systems I just listed.

My conjecture is that systems most likely to fail next have to do with connectivity via the Internet. That’s already passed the red warning line. Maybe start carrying cash again.



Leave a Reply

About Me

I’m Pete Harbeson, a writer (among other things) located near Boston, Massachusetts. In addition to writing my own content, I’ve learned to translate for my loquacious and opinionated pup Chocolate Bossypaws. No surprise, she mostly speaks in doggerel. You can find her contributions tagged with Chocolatiana.

Check out my newsletter, Artisanal Scribbles, where I’m grappling with writing, authentic human authorship, integrity, and the craft of beauty in the AI age. You can also find some of my old software projects at GitHub. Nothing very impressive. I mostly wrote tiny utilities in Python and shell scripts. I’m mostly retired from coding nowadays.

I find myself suddenly de-corporatized (their choice, not mine). To help keep the lights on, buy me a coffee!