I was wondering if anyone had some insight on this?
I understand that deprecating a function usually means you'll be removing it at some point, which goes against the rule "Evolution is addition only", but is it that bad to try to steer users towards a better API, or am I missing something particularly bad about that deprecation attribute?
As said in one of the YouTube comments for the video you linked here, it's not so much feature branches that are the problem, but long-lived feature branches.
If you want to add a complex feature to a project, you could do it via several branches, one after the other. For instance, the first branch could introduce a feature flag and the second branch implement part of the new feature, and so on.
As pointed out by some of you, this looks more like permutation of sequences. Also, since a symmetric cipher is used, I'm surprised the author didn't mention Black and Rogaway [1].
Their algorithm is a permutation (int -> int) that works on a domain of any size up to a limit. A typical application for this is encrypting credit card numbers so that the ciphertext still looks like a credit card number (non-trivial because the size of domain isn't a power of two) or efficiently shuffling sequences, randomly in appearance but repeatably if you know the seed.
For instance, this is used by Masscan to randomize the order in which IP addresses and ports are scanned [2]. I've built a Python package that could help you use this algorithm [3] (mostly for fun, but maybe that's useful, let me know :)).
I was wondering if anyone had some insight on this?
I understand that deprecating a function usually means you'll be removing it at some point, which goes against the rule "Evolution is addition only", but is it that bad to try to steer users towards a better API, or am I missing something particularly bad about that deprecation attribute?