I don't feel strongly either way, but to be pernickety, if something has been lost, it surely can't really be lossless, that's all I'm saying
I'm a big fan of downsampling fwiw.
The way I look at it is that if the file contains all the data that the sensor is capable of recording, then it's lossless, and anything else is by definition lossy.
What matters however is whether the lossy image is *visibly* inferior.
By inferior I mean has it lost sharpness or detail, have the tones and colours become compressed to the detriment of the photo, and are there any artefacts.
The most practical way to analyse this is to take simultaneous stills images in RAW, C-RAW, JPEG and HEIF, and compare them side by side on a high quality monitor at 100%.
Personally, I can't tell the difference comparing RAW and C-RAW, but there may be circumstances (e.g. astrophotography or black cat in a coal cellar) where uncompressed RAW has an edge. I can *usually* tell the difference between simultaneously shot C-RAW and JPEG, as the latter always has less fine detail, and under certain circumstances the colour compression can lead to a contouring effect separating areas of slightly different tone. Sharpening (especially with Topaz DeNoise AI) can do a fair job of restoring detail and edge sharpness, but if colour information has been lost, there's no way to regain it.
I always prefer to err on the side of caution, so until recently I shot almost everything on RAW. Since I got my R5 a year ago, I've mostly shot C-RAW and only shot RAW in particularly difficult lighting conditions. I only shoot JPEG in situations when shooting RAW might fill the buffer, e.g. BIF bursts.