[a8000]

I don't understand this attitude, if something is incredibly complex and very hard to get right, this can indicate that there is a mismatch between the tools you use to implement it and what you should use. With the right language it might be much more pleasant to implement: Consider for example https://github.com/GaloisInc/halfs a filesystem implemented in Haskell, I think it has ~10000loc

It can also mean that existing implementations are poorly documented and research papers on the subject place no importance on implementation details. Of course you should not use your first implementation in production, but that is as obvious as not letting someone do brain surgery who has no prior training in it.

[lttlrck]

I don't understand the point you are making with the filesystem. LOC as a single point of reference is useless.

[userbinator]

10kLOC of Haskell for a filesystem seems like an awful lot, especially when considering how much more concise and higher-level Haskell is supposed to be compared to languages like C. Maybe filesystem code just don't fit too well into the functional paradigm?

[a8000]

That number was an upper bound estimate obtained by visually integrating the commit history, it is entirely possible that it is significantly lower.