Hector Martin

As opposed to a scenario where Y must do as X commands and has no choice in the matter, which is something that would normally be termed a "slave".

The thing is "power dynamics" are rarely much of a thing in technology, given that usually there is no free will involved. Things usually do what they're programmed to do, so there is no "choice" distinction, right?

They do what they're programmed to do, yes. But in a distributed system some entities have a control relationship over others. Some parts execute any command they're given while others think about it, reject some according to policy, etc.

What i'm really getting at is, don't blindly replace one word with another unless you understand all of the subtle implications that has to the engineers reading the documentation. There likely isn't a 1:1 replacement for any problematic term.

Well, as I expressed in a previous thread we're already abusing master/slave in contexts where it has nothing to do with control, so the meaning is already all but totally diluted :-) E.g. for I²C "initiator/responder" makes a heck of a lot more sense.

Exactly - but if you're picking new terms, pick ones that make sense *in the context of the system you're using them in*.

Yup. Do you have an example of a case with more of a control relationship? I guess something like JTAG would qualify, but for that one we're already using "debug target".

SPI slaving of a microcontroller, where the main UC controls the slave UC?