I haven't accounted for that in my concepts, but we could add a way
to easily create custom lenses that the user could add their own
apps or widgets to.
I'm still not convinced that lenses are the way to go. Both iOS and Android support multiple screens with the swipe to left or right to view said screen. In both scenarios the screens are used to allow access to shortcuts, android allowing widgets as well. I know iOS allows the creation of folders for further grouping, not sure about android. It feels quite natural to swipe to the left or right to see further options, I know a lot of people who make certain screens their shortcuts for specific activities. While the option for custom lenses would provide this function, how would you navigate between them? Left and right swipe? If so then no need for icons along the bottom, maybe dots to show where you are in the virtual space. In all honesty, we could just use the ubuntu workspaces for this and workspace switcher. The metaphor for the interface is after all a desktop, while using the swipe to navigate which part of the desktop your looking at. I'm not sure what metaphor the lens incorporates.
Specifically in regards to widgets, iOS has shown that users can be
content without widgets. Possibly, we could allow developers to
widgets to the notification drawer?
Agree that iOS has shown that users don't need them, however that was only because iOS didn't originally support multitasking which I presume would be required for widgets, now I presume they don't support them because Android does and they don't want to do such a blatant copy. Another reason could be because integration with the desktop is difficult on a small form factor (tablets don't really have the problem) so sometimes the UI can appear pretty ugly with badly thought out widgets.
As for the notification draw for widgets... well in that case may as well just make them apps. the point of widgets is to allow access/interaction with information that the user deems so important enough to not use an application. Personally I like them and think they should be allowed, HTC's clock/weather widget is quite nice and I know a lot of people who were drawn to that phone purely because of that widget.
My main concern is that Unity will be shoved into the phone/tablet
arena. In my mind there should be development of key scenarios,
use cases, etc. A road map would be a good idea as well. Then when
mockups are presented then they can be compared against the key
scenarios and use cases to make sure they're suitable, or if they
require tweeking.
I'm not sure what you mean by a "Key scenario". Can you explain?
Scenarios (well I'm really referring to concrete scenarios) are part of means for capturing information in interaction design or HCI design. I suppose you could call it part of a methodology. Have a gander at:
Benyon, D. (2010). Designing Interactive Systems : A
comprehensive guide to HCI and interaction design (2nd ed.). Addison
Wesley.
Sharp, H., Rogers, Y., & Preece, J. (2007). Interaction
Design: Beyond Human-Computer Interaction (2nd ed.). John Wiley & Sons.
I forget which book talks about it the most, but it's a very good tool in creating real world scenarios of how people will use the system. Consider it a higher level use case which focuses on the person doing the activity. Both books are good, while they're not quite suitable for a lot of commercial software development (most companies would never wear the cost required for some of it), for new software solutions I think that large parts are very appropriate.
Cheers,
James