The thing is that activities, at least to the extent to which I've been able to figure them out, just seem to be a way of changing the content of the Plasma desktop.
I never interact with my desktop - it's just a backdrop for whatever I'm doing if that happens not to cover the screen, which is uncommon. If I wanted an interactive window, I'd use an interactive window and get the benefit of the usual alt-tab behaviour, and other window management features. Hence I personally find the Plasma desktop to be a flashy gimmick of no real value or interest - so far no real examples seem to have been given of how it's used for anything but bling - so activities seem rather useless.
Activities and the move to context-oriented desktops
Posted May 28, 2009 20:39 UTC (Thu) by nix (subscriber, #2304)
[Link]
It *is* useful if you have a big screen so can avoid maximising your app
all the time, or if you bind a key to windowshade your app (trivial with
any decent wm). Then you can have a bunch of per-app widgets (generally
folder widgets) giving things like graphical file management (useful? yes,
if, say, you're working on something with images in it: instnt
thumbnailer!) on a per-project basis.
One thing I'm not clear on (because I haven't looked) is how much of
plasmoid functionality you can bind to keys. (Next time I'm near a system
running X I'll have to check, but I'm on the wrong end of a modem right
now.)