-6

It's an interesting idea. Basically, you define states for your current activity. In your base state, you could have certain plasmids showing in certain positions and sizes. In a state (maybe "work" state) you could have an additional music player plasmid and hide the clock and desktop view. Switching between states could have a tween animation of some sort (something simple like linearly changing size, position and rotation). The key here is that the plasmids would be the same for each state. Each state would have the same plasmids with the same settings, just the size, rotation, visibility, and position would change.

Suggestions are welcome of course XD.

This seems like a lot of effort for little or not benefit. You can just clone activities and then make the modifications you want to the clone. Under your proposal you could have two activities that don't have a single usable plasmoid in common yet are all considered part of the "same" activity. It also means you have a bunch of plasmoids taking up processor power that you can't actually interact with. This makes the user interface much more complicated without any benefit I can see over just cloning activities.

It makes more sense to look at this from the point of view of an activity manager, that allows you to have profiles in order to switch easily between one set of activities and the other.

I don't see why we need this "state" concept. In the end different states means different activities, so there's no need to add this strange artifact.

I could do as TheBlackCat said, and clone my activity. It would work spiffingly well - until I decide to change something that is. So, suppose I have a clock widget on my desktop. It's configured to have a minute hand and a seconds hand. I decide I want a few different configurations for my desktop and make three clones. I move my widgets into the right positions, delete a few I don't want to see, and add a few different ones to each. Later, I decide that I don't want my clock to have a seconds hand in any of the activities. That means I have to go to every single activity, and change the clock on each and every one to have the new settings.

And that's just a simple clock. Imagine changing multiple RSS feeds over various activity clones. It just seems a little unwieldy. Anytime you want to change something, you have to manually change it for all of those clone activities.

Now, maybe it's not so bad. I could probably learn to live with it. But it would be nice if some like-minded developer decided to implement it anyways...

Maybe a compromise would do? Perhaps just give us an option to use the same widgets in a few activities? I don't know enough about the plasma desktop's engine to make any great suggestions.

I think allowing to have certain plasmoids on multiple activities (similar to how you can have 'sticky' windows which are visible on all workspaces) would be a suitable substitute.
Typically with sticky windows, the window stays in the same place irrespective of what else is on each workspace, but your idea suggests that you intend to move/resize/rotate a single plasmoid to different positions on different activities (but want to maintain a single config across all activities that have that plasmoid) - that could be accomplished with using some kind of reference from the size/location information (stored per-activity) to the plasmoid config (stored per-plasmoid).
Before proposing the sticky plasmoids, do a search for similar ideas.

As for the animations - currently there are no transition animations between activities. I think it could be nice to add (optional) transition animations between activities (the usual things like cross-fade, pan in/out, whatever).