I’m playing with Gimp 2.99 and just discovered that the image “Precision” (in 2.10) has been renamed “Encoding”. While I agree that it is also be seen as an encoding (even if JPEG/PNG could also be considered encodings), that new moniker doesn’t hint that you don’t get as much data in 8-bit gamma than in 32-bit FP linear. And elsewhere I see “encoding” used mostly to describe the gamma/linear difference, not the size of the data this applies to.
My guess is that that’s the terminology used in babl, GIMP’s color conversion library (e.g. “CMYK u8” is an 8bpc CMYK encoding). I looked into the commit history and didn’t find a rationale for the renaming when it happened. Sorry!
This is a long-standing discussion in the GIMP team and the change happened in 2020. The idea behind that change was that it’s not just about precision but about TRC too (linear, non-linear…), what data is encoded (RGB, grayscale, indexed… in the future probably CMYK, CIE Lab…), how (integer, floating point, as an index to a table…).
So yeah, it’s all about how the data is encoded (the “representation” of color data), not only about channel precision.
By the way, I don’t really have any opinion myself. IMO both are fine and are kinda technical terms anyway. I’m just telling you the why, from discussions which have happened for years between maintainers and in which I never really participated to (mostly because no opinions myself on this, as said).