> the authors have also learned of several other hash tables that make use of the same high-level idea in different settings [7, 9].
At least part of the result was already known, and the fact authors didn't know about it mostly goes to the large corpus of knowledge we already posses.
But the core inspiration came from looking at another recent research paper "Tiny Pointers": that is totally against your premise.
If Krapivin was a software engineer looking to implement this solution as optimization for a particular problem, he would have done so without ever thinking of making a research paper to prove it formally, but mostly relied on benchmarking to prove his implementation works better.
Now, it has always been somewhat true that lots of existing knowledge limits our creativity in familiar domains, but you need both to really advance science.
At least part of the result was already known, and the fact authors didn't know about it mostly goes to the large corpus of knowledge we already posses.
But the core inspiration came from looking at another recent research paper "Tiny Pointers": that is totally against your premise.
If Krapivin was a software engineer looking to implement this solution as optimization for a particular problem, he would have done so without ever thinking of making a research paper to prove it formally, but mostly relied on benchmarking to prove his implementation works better.
Now, it has always been somewhat true that lots of existing knowledge limits our creativity in familiar domains, but you need both to really advance science.