The finished thing is the least informative part of a project. It tells you what worked and nothing about why the six other approaches didn’t — which is the part you’d actually have needed.
What makes mitxela’s write-ups worth reading isn’t the photo at the end, it’s that the dead ends are still in there: the revision that didn’t fit, the measurement that turned out wrong, the bit where he admits he got lucky. A build log tells you how someone thinks. A product page tells you how someone markets.
There’s a selfish argument too. Writing it down while it’s still warm is how you find out whether you understood it. The gaps only show up when you try to explain them, which is roughly the same reason these live as notes you keep rewriting instead of posts you finish once.
The cost is that it’s slower and it exposes you a bit. Publishing the failed attempt means publishing the part where you were wrong — which is exactly why it’s useful, and exactly why most people skip it.