In the guinea pig, the large head at birth is provided for by the carteliginous symphysis joint in the hips detaching. However unless the animal gives birth early enough (which always happens in the wild), they lose this capability and die if impregnated later. Some doctors thought it a good idea to try to emulate this in humans by cutting the cartilage there instead of doing a cesarian section, but this causes permanent problems, as in humans the joint does not reattach. Notoriously, for religious reasons some doctors decided to do so anyway, since cesarian section reduces the number of pregnancies a woman can have, which they regarded as more important than being able to walk easily and being continent.

When I was a kid, I got my tonsils removed "because they were useless and a source of illness".

I've recently heard that tonsil removal is now more disputed: it may collect filth, sure, but it may also prevent it from going deeper into the body, which may cause more serious illnesses.

Given its vast complexity, and the timeline of its creation/evolution, I remain skeptical over bold claims about the human body. It's really missing an "as far as we know." The ability to go beyond what is known is paramount to the progress of science, and historically attested with some intensity (e.g. Earth's shape, relativity with time/space & axiomatic geometry). Humility thus feels like a better posture.

Who would let a junior dev trim bits, or boldly modify a decades old codebase?

I still remember this bit from school and various pop-sci book, but is it actually true? Is there really some group of neurons in the brain somewhere that actively tries to restore the "raw" visual information that was blocked by the blind spot?

Thinking of ANNs, I felt it was more realistic that higher layers in the visual cortex are mostly only using the visual information to find patterns anyway, and that they're robust enough they can still find those patterns without the data from the blind spot locations. (As long as a pattern isn't fully contained within the blind spot regions of course)

An analogy would be a QR code reader that can still parse the encoded information if a part of the QR code is missing - but it won't actually "reconstruct" the missing sections to do this.

But I don't know if it really works like this.

I'm dreading the horror of genetic manipulation it would open. The gene editing craze feels like it is right around the corner.

In giraffes that nerve is several meters long.

[1] https://en.wikipedia.org/wiki/Recurrent_laryngeal_nerve