Now that I know this, it is not a big deal. But it was confusing. Why not make it work when you click on the grey area? I realise that this area contains rotate buttons and star ratings, but most of it is inert.
I think this is a valid design decision once one understands and gets used to it. In fact I had to go and re-try it to compose this post whereas when I'm actually using LR/Classic I don't even think about it anymore, My hand just seems to go to either the image or the canvas based on what I'm trying to do without me even thinking about it.
One could argue that it is not documented well, but that depends on who's documentation you choose to read.
It is by design that dragging on the grey frame area does nothing?
If I had a painting on my wall and grabbed its frame and took it into another room, I would kind of expect the painting to come with it!
The grey frame (or canvas as you call it) has three pieces. I had a look around the Adobe help and could not find names for them. Let’s call them the title bar, the frame body, and the bottom title. Adobe calls the image itself the thumbnail. The title bar has the photo number and some other information. The frame body has a flag button. The bottom title has rotation buttons and the star ratings.
The thumbnail and the frame body have the same context menu. The title bar has a different context menu, mostly concerned with what information is displayed there. The bottom title has no context menu.
Weirdly and wastefully enough, clicking on the title bar has the same effect as right-clicking on it. Also, this does not select the photo. Clicking on the frame body or on the thumbnail selects a photo.
I get your point that having the thumbnail and frame(s) behave differently might be a powerful usage paradigm. But I would argue that it currently is not, and is counter-intuitive for Windows users. The analog for Lightroom Classic grid view is icon view in the familiar-to-everyone Windows Explorer, which has no concept of a frame around its icon representations. It does frame (or outline) the currently selected icon(s). I was intrigued by your comment about multiple selection, but the same effect of changing the active single-selection in a multiple-selection is achieved on Windows by just doing a control-click So this different paradigm has no new functionality and as far as I can see is just confusing.As I said in my original post, we can easily learn a work-around. Humans can learn any bizarre interface, eventually. It doesn’t mean that we should, though. I honestly thought for months that LR was broken – not a good way to introduce it to a novice user with long Windows experience.
Personally I think it is a good feature to be able to make an image the active image and depending on where I click either keep the other selected images selected or de-select them. I would like to see the context menu's stay the same where ever I click, and I'd like to see a few other choices in the context menu. But to each their own and I can understand why you may consider this a poor design choice.