Comments on: Systems fight back

Systems fight back

chavenet — Mon, 15 Sep 2025 01:48:36 -0800

This lineage of (mostly) working systems is easily forgotten. Instead, we prefer a more flattering story: that complex systems are deliberate creations, the product of careful analysis. And, relatedly, that by performing this analysis – now known as 'systems thinking' in the halls of government – we can bring unruly ones to heel. It is an optimistic perspective, casting us as the masters of our systems and our destiny. from Magical systems thinking [Works in Progress]

By: HearHere

HearHere — Mon, 15 Sep 2025 02:00:03 -0800

The first electric grid was no more than a handful of electric lamps hooked up to a water wheel in Godalming, England, in 1881 [godalmingmuseum:] Streets had been lit up in Paris, London and San Francisco before Godalming's lights. On the 18th September 1879 Blackpool promenade pier was lit by six Siemens lights and the electric light had been introduced on 28th October to the British Library reading room

By: lalochezia

lalochezia — Mon, 15 Sep 2025 02:09:10 -0800

Lots buried in here(ironically), but peeping through the banal disparate examples is the cry of "deregulation, get out of the way and let us build", which, Given "us" in the context of WIP is not a neutral political position. Especially when sold with a plaintive cry for "humility" at the end (!). One might argue that this essay was constructed deliberately around this philosophy, constructed to be a trojan horse.

By: johnabbe

johnabbe — Mon, 15 Sep 2025 04:08:46 -0800

Appreciating the pointer to Systemantics! (groan) I had not heard of the book, even though I knew that systems thinkers were talking about these sorts of realizations by the 1970s (also see second-order cybernetics, ideas like double-loop learning). The article seemed to end suddenly, without ever acknowledging that systems thinking/thinkers have now had 50+ years of working with the understanding that systems "fight back" or have a mind or entity-ness of their own. A part two, digging into that history (and present), now that would be interesting! Permaculture comes to mind as one of many systems domains which place centrally things like observing, learning, and working with what is already in the field. Ward Cunningham's classic question, "What's the simplest thing that could work?" would also fit in a part two. Indigenous perspectives. Might need a part three...

By: late afternoon dreaming hotel

late afternoon dreaming hotel — Mon, 15 Sep 2025 04:13:58 -0800

I don't have the reading that the comment above expresses. I imagine context is key. I'm a regulatory toxicologist with 20ish years of dealing with the regulatory systems that, in the most charitable interpretations, have only been intentionally built at the smallest scales and have never been designed intentionally in a holistic way. The way medicines reach us is, to put it simply, it humble at all. The system is lost and self-sustaining unless crises (like the pandemic, mentioned in the article) provide the political will for momentary, targeted approaches that improve some parts of it. Every single human being working in the regulation of medicines has not stopped talking about the positive changes ushered in by the pandemic's political, public necessity for the system to work better and faster. It's like we've seen what could be, and yet we know that it's not a common experience to witness such a total shift in an approach to regulation. I hope that there are new leaders in thought brewing today ideas of leadership that have been cultivated by the humility (and by the success) of that recent history.

By: jy4m

jy4m — Mon, 15 Sep 2025 06:03:08 -0800

This is a somewhat novel marriage of two seemingly opposed conservative premises: that solving complex problems with top-down interventions will always (always!) result in failure, and that we can let the market run roughshod over public programs and it'll certainly work out. I think it kind of slips up with the examples. Yeah, the French Revolution was a meticulously planned affair, and that's what was wrong with it.

By: BungaDunga

BungaDunga — Mon, 15 Sep 2025 07:20:37 -0800

HealthCare.gov was designed to simplify access to health insurance by knitting together 36 state marketplaces and data from eight federal agencies. Its launch was paralyzed by technical failures that locked out millions of users. This is a weird example because the problems were fixed and it works fine now.

By: BungaDunga

BungaDunga — Mon, 15 Sep 2025 07:26:51 -0800

like, Healthcare.gov absolutely was not a futile attempt to rationalize a system that had its own irrefutable logic, it was an IT project with a failed rollout that got repaired and went on to success. America's healthcare is not provided by a "simple working system" that was slowly accreted into a functional but messy reality. And Healthcare.gov was not an attempt at centralizing it all or a Project Cybersyn or whatever. And you can tell because it worked eventually!

By: mhoye

mhoye — Mon, 15 Sep 2025 08:07:32 -0800

I feel strongly about this, but mostly from a personal, psychological preparedness standpoint. People can change systems from within, and absolutely should, but: you have to prepare yourself for that to be a long, drawn out struggle, without a lot of clear wins or obvious success checkpoints.

By: Aardvark Cheeselog

Aardvark Cheeselog — Mon, 15 Sep 2025 08:39:51 -0800

After looking at the thread I should have known to give TFA a miss but I clicked anyway. He's not wrong that systems analysis has historically had a hubris problem, and there are some systems that nobody is a good enough analyst to break them down into logical manageable pieces. But after that one fiasco example of a broken rollout that was later fixed, I don't think any evidence he offers for anything can be trusted. TFA is woo-slinging at best, and an argument for libertarianism with the blade guards off at worst.

By: drowsy

drowsy — Mon, 15 Sep 2025 09:06:45 -0800

Also very happy to learn about Systemantics and of course in some situations we should perform endruns around existing systems. But the vibe of the story doesn't give credit to why things get complicated. What does that evolution look like that turns a simple plan into an unwieldy one? How can you manage when legislation is a necessary mess? Those are the questions I am left with, and glad to have them, thanks OP. Now I am hungry for pointers. Warp Speed and vaccine discovery for C19 was a good thing. (The rollout maybe not.) And we are seeing now what happens when Executive orders are treated as better than legislation. DOGE etc. What we do with these lessons matters a lot.

By: clawsoon

clawsoon — Mon, 15 Sep 2025 09:55:26 -0800

Systemantics is a fun little book. I have it on my shelf beside books like The Peter Principle and Parkinson's Law.

By: mhoye

mhoye — Mon, 15 Sep 2025 12:08:39 -0800

But the vibe of the story doesn't give credit to why things get complicated. What does that evolution look like that turns a simple plan into an unwieldy one? Bug fixes.

By: clawsoon

clawsoon — Mon, 15 Sep 2025 12:57:03 -0800

Bug fixes. Also not-invented-here syndrome. I've noticed in both computer animation production management and software development that people generally don't understand complex systems unless they rebuild those systems themselves. This leads to parts of systems being rebuilt in inconsistent ways with different visions, different levels of ambition, and different levels of follow-through.

By: chromecow

chromecow — Mon, 15 Sep 2025 13:22:13 -0800

Can speak from game industry experience, "Comps" (competitive comparisons) are one of the worst offenders for...incoherent...systems design. A producer might come in after the launch of a competitors game (hypothetical for instance), and champion the inclusion of some of their systems. Systems that were not designed to work within your ecosystem of systems. Maybe you aren't given sufficient time to redesign the systems ecosystem to accommodate the new bullet-point-on-the-box. Maybe you attempt triage by fencing those systems off from the rest of the game so they don't break...everything. If you've every played a game and thought, "That feature seemed bolted on," or "What is the point of this? It doesn't seem to affect any other part of the game?", thank a designer that did their best to avoid a more dramatic level of systems incoherence. While I am solidly anti-libertarian, which is to say, I am a human being with empathy, I have some sympathy for the idea that systems that accrete complexity over time need some mechanism to pay off what we would call in the game industry design debt, or tech debt. I don't have good solutions, especially in an adversarial environment with dishonest actors. Getting mad and kicking everything over (massive deregulation) hurts so many people. Usually the answer is...do the work, make the incremental changes, devote some percentage of your time to paying off the tech debt by revisiting old rule sets and seeing what incremental changes would improve the lives of those that interface with those rules.

By: chromecow

chromecow — Mon, 15 Sep 2025 13:56:22 -0800

Ok, I read TFA, and I find it sloppy. Starting with germanium transistors and working your way up to giant fabs factories = good Iterating on legislation to achieve public good like healthcare exchanges = bad? Bypassing bureaucracy to build giant-from-scratch rocket program = good Building giant healthcare system when needed = bad? I agree that this sounds like a general call for deregulation, which...libertarian? Neo-liberal? GOP? Who isn't the party of deregulation in America? Every designer wants to build perfect systems., The great designers don't let the perfect be the enemy of the good.

By: The_Vegetables

The_Vegetables — Mon, 15 Sep 2025 14:07:42 -0800

IDK, I think Republicans get too much credit for being the party of 'deregulation' while Democrats sheepishly stand in the corner and say 'nationalized healthcare'? Like the health care system is of value in and of itself, no matter how well-designed and complex. It's not. It's in the way to people actually getting health care. I'm glad the Healthcare.gov worked did an amazing feet to combine the backends of 37 different health care systems - it's what was necessary - but is it actually a good idea? No. One health care system would be far better. Incremental change for problems that aren't out of control is a great idea. But once the system is out of control and half the population is suffering - slowly nattering about the edges helps James Jones living at 1137 Edgewater Lane, but it leaves behind 37,000 other people.

By: chromecow

chromecow — Mon, 15 Sep 2025 15:23:03 -0800

The_Vegetables I think that is a totally valid critique! A great counter example (counter to incrementalism) would be The New Deal. Yes please, more of that! I'm definitely gun-shy around big, fast-moving government initiatives in general (and specifically, when billionaires are firmly at the wheel). Both parties are (currently) Billionaire fluffers, imo. More firebrands like AOC and Sanders, please! And by please, I mean that's where I'm putting my money this time around. Also, I see Healthcare.gov as a systems success, and as a failure of political leadership. And a failure of the people to hold leader's feet to the fire. We have a lot of problems that a rich country could solve with a modicum of vision, and the political will to act on it. Instead, we get...[waves hands at everything]. The bad news is...[waves hands at everything], the good news is, maybe this is the end of the era of political apathy in the US? I'm out protesting every weekend, and I don't see that changing if Democrats somehow dislodge TFG from the people's office next cycle. You are correct. The problems are too big for incremental change at this juncture.