I've been a part of a few successful rewrites, one definitely larger than what Sonos did.
We should let go of some of the JoS dogma. Failed rewrites are the result of mismanagement, not because rewrites are inherently bad.
We should let go of some of the JoS dogma. Failed rewrites are the result of mismanagement, not because rewrites are inherently bad.
And a lot of “bad code” comes from starting to write clean code and then coming across a variety of assumptions that prove incorrect, and real-world considerations that add complexity.