News for hackers. Stuff that matters.
Re:Chesterton’s Fence: A Lesson in Second Order Thinking by charlieflowers on Tuesday March 10, 2020 @08:11PM (#22541500)
Am I the only one that finds it irritating that the article starts out saying the same thing over and over?

First as a parable, then a restatement, then a bold restatement? Why not repeat it again, but underline it this time?

I feel like an 8th grader did it for homework and tried to stretch it to the maximum amount of words.

Re:Chesterton’s Fence: A Lesson in Second Order Thinking by charlieflowers on Tuesday March 10, 2020 @08:30PM (#22541619)
Just to clarify ... it is the start of the article that I’m criticizing. Once it gets going it makes some decent points.
Re:Chesterton’s Fence: A Lesson in Second Order Thinking by andrewflnr on Tuesday March 10, 2020 @09:15PM (#22541954)
Interesting. It was actually the back half where I thought it was getting repetitive.
Re:Chesterton’s Fence: A Lesson in Second Order Thinking by sandoooo on Tuesday March 10, 2020 @08:31PM (#22541637)
Legibility is a bell curve. The author could have stopped earlier, sure, but at the cost of reaching a smaller fraction of the audience.

Words are cheap. No trees were harmed. No need to be stingy.

Re:Chesterton’s Fence: A Lesson in Second Order Thinking by TheOtherHobbes on Tuesday March 10, 2020 @09:49PM (#22542150)
However, a number of readers were irritated and left unconvinced.

Chesterton's Fence sounds like Pascal's Wager - it's perfectly convincing and rational until you realise it's misleading and oversimplified.

Of course you shouldn't get rid of things just because you want to change them. But unless a fence-builder left an explicit record explaining why the fence is there, your belated estimate of their reasons is at least as likely to be wrong as your desire to remove the fence.

And that doesn't matter, because the original reasons are irrelevant.

The fence may have been built for a good reason, and the reason may no longer apply. It may not have been built for a good reason, but a good reason now exists.

Skip straight to the problem, ask stakeholders (fenceholders?) if you can find them, run predictive modelling, maybe some A/B testing if it's Schrodinger's Fence, make a backup just in case, and then do what you need to do.

The original fence builder is likely long gone and may not want to talk to you anyway.

(Although if it all goes horribly wrong, maybe try looking on LinkedIn?)