You could have invented Fenwick trees

Tangential to the algorithm itself, this is about naming:

> Fenwick trees, also known as binary indexed trees

Every time I read something like this, I'm reminded of https://willcrichton.net/notes/naming-conventions-that-need-... . "Fenwick tree" makes it seem like some unknown strange entity, while "binary indexed tree" immediately makes it a lot more accessible and gives you a better handle on the concept too.

I'm reminded of a town in New Mexico named after the railroad paymaster because everyone said "I'm going to see Gallup to get paid" -> "I'm going to Gallup to get paid" -> "I'm going to Gallup".

Short, memorable names will always dominate precise, descriptive ones.

Finding short, memorable, precise and descriptive names is why "naming things is hard".

Meh, I agree in spirit but in practice it’s quite hard to get enough unique names using descriptive terms. I also think there’s some benefit in learning history that comes from naming things after people.

In practice, the most it gets is people thinking "I guess some guy named Fenwick was involved somewhere along the way", if even that. I think there's a lot of benefit to learning the history of how things were discovered, but that should be done consciously by making it part of the teaching material. The names make barely any difference.

It's not trivial to come up with descriptive names, but the difficulty is often overstated because people underestimate its importance. The names don't have to be perfectly descriptive, we don't need twenty syllable names like organic chemical compounds, but every step taken towards making it descriptive helps a lot. The reality is that the names get chosen basically arbitrarily in journals or textbooks, with hardly any thought given to how it will affect generations of students, working professionals, and even future researchers in remembering the concept and looking it up in their mind among the hundreds or thousands of other things they've learnt.