- current vision, ideas, concepts and terminology: are we thinking the same thing when we say a term?
- identify user scenarios, what do you want to do, why and in what situation
in the machine (example: being at work on a document while discussing about it simultaneously on email, chat and audio call)
- what in current suituation of things matches with this?
- what doesn't?
- Try to answer the problem without looking at other answers..
"because it works on OSX" or even "because it works on Plasma Desktop" are not valid answers ;)
- Thinking instead for a moment about the other solutions: why the solution of other system does work/why it doesn't work?
High level "manifesto" points
- natural feeling ui
- few user interruptions
- settings always correct for what i'm doing now
- right level of mono/multitasking in different situations (complete
dedication to a task vs listening mucic/glancing over news items while working)
- working on a document while talking about it
- doing a presentation
- doing a call
Define sub task
Define sub tasks of the "manifesto" points, more deltailed, but still not regarding implementation
- realistic light effects.
- High performance, but highly usable, UI when "candy" effects are disabled. (I.e., Compositing yes, Effects no.)
- "One-Handed Desktop" via Mouse Shortcuts on platforms which can support mice. (NOTE: Windows appears to provide very limited capabilities, in comparison to Linux, Unix, OS-X, and even Blackberry.)
- HIG for Gestures, and HIG for Button-based shortcuts.
- Must be usable with just a keyboard (i.e., with no Pointer/Mouse device.)
- Is our on-screen keyboard widget "good enough"?
Efficience in doing tasks
- think in steps required to do a task: not just in "launching applications" that is just an aspect of it
- what is and what should be the boundary between what is an app and what is the workspace
- generic "information flow" of everything that i see on the screen: includes applications switching, not limited to it
- what are the scenarios and the problem they are designed to solve
- what are the ui-wise problems in them right now?
- A possible one: gulf of execution (Info)
- How to reduce it?
- when to turn on/off bluetooth, audio level, camera?
- what to do if i attach a screen? a beamer?
- what to do if i disconnect to office wifi and connect to home wifi?
Notifications and sys tray items
High level thinking
- What are they for?
- What are the different use cases, what are the different "types"?
- What interaction is needed?
- What does the user need to be aware of?
- Identify each category of current notification (i.e needs to persist)
- A mockup of how notifications should work
- A HIG of when and how items should appear in the sys-tray, and what notifications should be enabled by default. How often can they be emitted (i.e KTp should not emit 5 notifications when the network drops)
A Plamoid Desktop HIG
- A HIG that is _finished_
- If there's a HIG for active, we need a HIG for situations with a mouse and keyboard, where interaction is (and should be) different, but with a very large overlap.
- Based on current plasmoids, not a big shakeup, but define consistency whilst things are being rewritten in QML.
Planning for implementation
- Write an updated version of what is now "ways of the plasma", ui guidelines etc
- identify where the work has to be done for an abstract task item:
- qml components
- tasks applet
- actual coding/documenting/coding tasks up for grab ;)