It's an encoding, not a cipher. The goal of an encoding is not to secure a text from being read, because the algorithm is all you need to know in order to decode it (and security by obscurity is sure not something desireable).
A cipher would have a key, so that the knowledge of the algorithm isn't enough to decrypt it and you can use different secrets for different messages (cracking one message won't help cracking all of the others immediately too, this way you can have different persons send messages to using different keys, without fearing that one person might be able to read the messages you send to the other)
Criticising null encoding for being easy to decode is the same as criticising base64 for being easy to decode. It doesn't make sense at all. An encoding should be easy to encode and decode.
A ciphertext does not have to be encrypted the way you define encryption (assymetric or symmetric keys). A substitution cipher is a weak encryption and one of the first described forms of creating cipher texts in history along with the caeser cipher.
Substitution cipher of the same word, but the cleartext is not "easily" known without knowing the alphabet/substitutions:
fpggrr
viuu66
fbppee
I'm sure you are well aware of all this, it's just for reference to other readers
The "null byte encoding" described is a fixed substitution, but it still is basically a substitution cipher and at first deciphering the method of cracking is the same as when the algorithm is unknown.
One reason compare the described encoding to a substitution cipher is TS's comment on "hiding"/securing text for people using the encoding.
