5 ms·

Nonsense. Chesterton’s fence creates a clear criteria for eliminating fences and other accumulated junk

api 3y ago
What if you investigate why the fence is there and find nothing or a list of contradictory or nonsensical reasons? This is incredibly common in real life. If you must understand something to remove it you do end up with a lot of things that can never be removed. It’s a big reason that laws and regulations build forever.
- nerdponx 3y ago
  At that point, you give up on the search for understanding and must evaluate the fence given the information that you do have and can make sense of. It's really not any different from any other decision making process. The principle exists to encourage critical thinking, not to obstruct progress.
  - api 3y ago
    I actually like the principle as a tool for critical thinking, but like the precautionary principle which can also be a tool for critical thinking it has a tendency in practice to be converted into a thought-stopping and action-stopping cliche.
- latency-guy2 3y ago
  > What if you investigate why the fence is there and find nothing or a list of contradictory or nonsensical reasons? This is incredibly common in real life. Congratulations, you can now opt to remove the fence. This is also discussed in the article. Chesterton's fence is philosophical, and as always, there will be a value proposition to consider. E.g. building a new beltway that would connect two previously unconnected cities or developing trade routes completely overrides the value that a fence being used to keep wildlife away.
  - eviks 3y ago
    Nonsense, the response you'll get is to spend more time investigating: "To which the more intelligent type of reformer will do well to answer: “If you don’t see the use of it, I certainly won’t let you clear it away. Go away and think HARDER and investigate better. ""
- pixl97 3y ago
  > find nothing or a list of contradictory or nonsensical reasons Then you look at implementing process changes in the company at a higher level, there is 100% something absolutely wrong, but it may not be a problem with a physical system, but a human interaction one.
moring 3y ago
The criteria for removal are clear on paper, but impossible to meet, so the effect is the same. It is impossible to meet because the reason why the proverbial fence was put up is, more often than not, a mixture of misunderstandings, logical fallacies and political motives from both the one who put it up, as well as other stakeholders involved, and most importantly, none of those aspects are documented and none of the people are around anymore. You'd need nothing less than a time machine to understand why the fence was put up.
- nerdponx 3y ago
  As with all things in life, it's a matter of accumulating evidence up to an acceptance criterion. That criterion can never be 100% certainty, because nothing can be 100% certain. The difficulty lies in agreeing on what that point should be, and on the magnitude of any particular piece of information as evidence for or against any particular hypothesis. "The fence was a mistake" is a completely legitimate conclusion here. The principle exists to prevent us from jumping immediately to that conclusion out of convenience, arrogance, or ignorance. It should not be interpreted as preventing us from ever reaching that conclusion after some careful consideration.
- mcswell 3y ago
  "impossible to meet", "more often than not", "none of those aspects are documented": These are huge generalizations, and I very much doubt that they are generally true, or that anyone--including you--has done any kind of investigation into how general these conditions are.
- paulsutter 3y ago
  Fascinating, can you give a couple examples of pre-existing rationale that make it impossible to replace an unneeded fence?
  - haswell 3y ago
    Not OP, but I’ve encountered this in situations where legacy code involved in mission critical business functionality was nearly impossible to remove due to the potential risk of unforeseen impact. In other words, if the potential impact of an unforeseen breakage is high enough (costs us or the customer $$$), it’s not worth risking the change even if we can find absolutely no good reason for the current behavior to exist. Example: complex spaghetti that touches billing calculations. These things are better addressed by a full rewrite/redesign followed by a long period of running both things in parallel until we’re confident that we didn’t miss anything. Maybe this is just a more complex way of removing the fence, but I think it’s more like moving away from the ground the fence is built on.

api 3y ago

What if you investigate why the fence is there and find nothing or a list of contradictory or nonsensical reasons? This is incredibly common in real life. If you must understand something to remove it you do end up with a lot of things that can never be removed. It’s a big reason that laws and regulations build forever.

nerdponx 3y ago
At that point, you give up on the search for understanding and must evaluate the fence given the information that you do have and can make sense of. It's really not any different from any other decision making process. The principle exists to encourage critical thinking, not to obstruct progress.
- api 3y ago
  I actually like the principle as a tool for critical thinking, but like the precautionary principle which can also be a tool for critical thinking it has a tendency in practice to be converted into a thought-stopping and action-stopping cliche.
latency-guy2 3y ago
> What if you investigate why the fence is there and find nothing or a list of contradictory or nonsensical reasons? This is incredibly common in real life. Congratulations, you can now opt to remove the fence. This is also discussed in the article. Chesterton's fence is philosophical, and as always, there will be a value proposition to consider. E.g. building a new beltway that would connect two previously unconnected cities or developing trade routes completely overrides the value that a fence being used to keep wildlife away.
- eviks 3y ago
  Nonsense, the response you'll get is to spend more time investigating: "To which the more intelligent type of reformer will do well to answer: “If you don’t see the use of it, I certainly won’t let you clear it away. Go away and think HARDER and investigate better. ""
pixl97 3y ago
> find nothing or a list of contradictory or nonsensical reasons Then you look at implementing process changes in the company at a higher level, there is 100% something absolutely wrong, but it may not be a problem with a physical system, but a human interaction one.

moring 3y ago

The criteria for removal are clear on paper, but impossible to meet, so the effect is the same. It is impossible to meet because the reason why the proverbial fence was put up is, more often than not, a mixture of misunderstandings, logical fallacies and political motives from both the one who put it up, as well as other stakeholders involved, and most importantly, none of those aspects are documented and none of the people are around anymore. You'd need nothing less than a time machine to understand why the fence was put up.

nerdponx 3y ago
As with all things in life, it's a matter of accumulating evidence up to an acceptance criterion. That criterion can never be 100% certainty, because nothing can be 100% certain. The difficulty lies in agreeing on what that point should be, and on the magnitude of any particular piece of information as evidence for or against any particular hypothesis. "The fence was a mistake" is a completely legitimate conclusion here. The principle exists to prevent us from jumping immediately to that conclusion out of convenience, arrogance, or ignorance. It should not be interpreted as preventing us from ever reaching that conclusion after some careful consideration.
mcswell 3y ago
"impossible to meet", "more often than not", "none of those aspects are documented": These are huge generalizations, and I very much doubt that they are generally true, or that anyone--including you--has done any kind of investigation into how general these conditions are.
paulsutter 3y ago
Fascinating, can you give a couple examples of pre-existing rationale that make it impossible to replace an unneeded fence?
- haswell 3y ago
  Not OP, but I’ve encountered this in situations where legacy code involved in mission critical business functionality was nearly impossible to remove due to the potential risk of unforeseen impact. In other words, if the potential impact of an unforeseen breakage is high enough (costs us or the customer $$$), it’s not worth risking the change even if we can find absolutely no good reason for the current behavior to exist. Example: complex spaghetti that touches billing calculations. These things are better addressed by a full rewrite/redesign followed by a long period of running both things in parallel until we’re confident that we didn’t miss anything. Maybe this is just a more complex way of removing the fence, but I think it’s more like moving away from the ground the fence is built on.