Hacker News Reader

Comment by api | Hacker News Reader

> it is important to know what those reasons are before attempting to modify the things.

People misinterpret Chesterton's fence as an argument against changing things, but it's actually just about what I quoted above. It has no verdict on whether or not something should be changed. It just argues that you should make a strong effort to figure out why it is how it is before attempting to do so.

I'd argue that if you don't at least do the exercise your attempts to change things will probably fail, since whatever you try to do will likely fail to capture something necessary in the old system.

Keep in mind though that it doesn't always work. Sometimes the reason for the fence is forgotten. Sometimes the reason is deeply perverse and there never was a good reason. Sometimes the reason is obsolete.

View on HN

Yup. I've often seen Chesterson's Fence trotted out as an argument to never change anything. Or at least place onerous burden of proof on the new change to prove itself.

However, some times the circumstances that gave rise to the fence are gone! You still need to do the work to show that this is true, or at least some best effort. But if after a cursory check there is no compelling reason, retorting "ah but there may still be some use, go search the wisdom of the ancients further" is a terrible response. That is how you calcify debt and become unable to adapt to changing circumstances.

There is a balance between preserving out of an abundance of caution and tearing it all down because it was inconvenient or "new is better."

We already have plenty of innate status quo bias. We don't need to heap more on top without good reason.

The problem with Charleston's fence is that most of the time the reason the fence needs to be removed isn't "I don’t see the use of this; let us clear it away.", but that the fence is actively a problem.

Unfortunately, I've seen this play out too many times. There is a trivial fence in the way of solving an expensive problem. Instead of taking action, months, or years pass trying to over- analyze the different ways to take down the fence. Since no one takes any action, the fence falls down on its own, causing severe outages and thousands, or millions of dollars in collateral damage.

I think plenty of people err on either side of this question. It's an eternal controversy, and as you say, there is a balance.

That having been said, I also think you're underestimating the degree to which it's easy to verify the value of legacy practices.

https://slatestarcodex.com/2019/06/04/book-review-the-secret...

Search for the phrase "And then there’s manioc. This is a tuber native to the Americas." Read the extended block quote afterwards and the first few paragraphs after the block quote. If anything, Chesterton's Fence is insufficient--if a certain practice works, it sometimes works for reasons that even the practitioners can't explain.

Furthermore, I can point to numerous examples of catastrophic failure directly caused by insufficient status quo bias, like tearing apart cities because it's the 20th century and we need to build freeways straight through the middle of them, or communism. So it's a lot more complicated than you're making it out to be.

All those anecdotes are cute, but has there been any attempt at systematically determining whether the net result of all instances of a particular culture’s practices which are justified by “do it this way because it has always been done this way”? This seems difficult to analyze, but it’s critical to the validity of any argument that invokes these cute anecdotes. Isn’t it plausible that generations of repressing people’s curiosity in experimenting with food preparation had a net negative effect?

Interesting point. I think you could characterize "being conquered by the Portuguese in the first place" as a negative outcome, so there's a plausible argument that the Portuguese were more willing to embrace innovation, which is how they gained the ability to conquer the Tukanoans in the first place.

On the other hand, if you dig into the history of colonization, indigenous cultures had a massive home field advantage, and it was common for Western colonies to utterly fail, either starving to death in the same lands that indigenous people were able to survive in or in some cases, completely defecting from their original culture and joining indigenous populations.

Okay but then what is your version of Chesterton’s fence actually saying? It just sounds like “don’t do something unless you think it’s okay to do.”

some years ago I had to take a break from my startup to consult, partner had a consultant come in to do some stuff, consultant wanted to redo a bunch of things to their liking - one of things he redid was to redo how products were found to be based on navigating by brand - without evidently taking time to consider that many sources of data we had, had a very poor understanding of what constituted a brand and that a brand might very well be some escaped xml and unreadable unicode glyphs, the letter A, the exact same text as the product name, the word Shirt, or a number of other things, including in some instances the actual brand.

In other words he did something because he thought it was okay to do it, he did it because he didn't like the way it was done, but he did not consider that in this case there might have been a reason why it was done the way it was and not the way he wanted it.

There were many things he could have changed, and improved things, but since he didn't bother finding out why things were they way they were changing things for the better was at best a random outcome.

I most commonly apply Chesterton's Fence in the context of new employees / managers. In that instance, it's more about naivety or ego ("I know better"), and less about needing to read the ancient runes to fully understand something.

I've seen so many people jump into a business, want to prove themselves, and start changing things without understanding the context in which that thing was built. (Similarly, though less often, businesses expanding into new products or markets.) "Tell me what will change if you stop doing X" is thus added to "Tell me what will change if you start doing Y".

To be fair, citing Chesterton's Fence just make me sound wise (and/or a little curmudgeonly); the better story for the point I'm making is the hospital bed where someone dies every Thursday morning at 10am ... only for video footage to reveal the cleaner unplugging machinery to plug in their vacuum cleaner every week.

> It has no verdict on whether or not something should be changed.

Hard disagree. The difficulty of discovering the reasons for Chesterton's Fence and the impossibility of ever being certain that you have discovered them all means that it actually is a general prescription for caution / against change.

Of course, as always, it comes down to a judgement call. Do you understand the reasons for Chesterton's Fence well enough? Unfortunately, that's exactly the same situation one was in before the Chesterton's Fence metaphor was brought up, so either the metaphor is completely useless (by way of being disclaimed into oblivion) or it is a general prescription for caution / against change.

I think the way you're conflating "for caution" and "against change" is a bit of sleight of hand though. I would definitely agree that it's a prescription for caution, but not necessarily against change.

the bias you detect is actually the bias of the changers; from experience we learn that the people who agitate for change are overwhelmingly likely to not look into the reasons things are the way they are.

It isn’t against change, it’s about making the correct change and correct amount of change to achieve the desired result, which can often be no change.

It’s against wanton and unwise change.

Considering nuance in qualification “disclaimed into oblivion” is needlessly derisive.

Reply on news.ycombinator.com