[Usability] A few questions re gnome3 usability



-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I haven't figured out how to test this on my primary environment (netbook) as all I have access to is unity and it's effectively unusable (and causes crashes).   So I'm asking
from the perspective of trying to figure out where to begin in these questions.   I'm currently working under kde unwillingly (it's heavy, it's high resource use, and it randomly
forgets things and crashes during suspend/resume)

1. keyboard-only interaction?   Can gnome3 be used with only keyboard interactions?    reason: touchpad is not very reliable.  Sometimes programs behave like gestures.
Sometimes they don't.  Touchpad is not much like a mouse, and very much not like a touch interface.   Also, moving pointers around interrupts workflow.
Also, is there any consideration for input tablets (Wacom etc) where there may be far more than 3 buttons and very very high precision.   For a group that started out working
with The GIMP, this should be an important item.   It makes me think that pointer support for 0+ pointers present (eg: no pointer, evil touchpad, mouse, trackball, digitally
precise tablet (and similar drafting tools), touchscreen, pen computing, motioning to a point in space in 3D immersion (and its little cousin kinect), etc).  However the one bit
of Internet Explorer I respect is that no matter what, you can get to every control on the screen with keyboard.   I have heard from people who work with interfaces for disabled
people that this is very important.   (and then there's the fun other extreme: touchscreens where no keyboard is available, which seems to be what Unity is made for)

2. Dual videochip systems (eg; ION2 such as I have : intel/nvidia switching).   This actually breaks into a more complex question - when is 3D rendering to be requested, and when
it is it not?   Ideally, it only switches to 3D mode for high-power requiring graphical applications, and remains in non-3D mode intel for rest of interactions.
The only environment that addresses this is actually Windows - which only activates the secondary chip for DirectX, so I've not much to go on for suggestions here.  "ironhide"
software handles the switching manually, and it's a bit of a bear to configure currently - but that's likely to change.   It's a very good approach for any system where power
management is critical.
This may cross several domains.   It's the kind of thing that's just not well handled in linux, and never has been.
How it functions is that the graphics chip can be enabled and disabled dynamically, and uses an overlay to communicate with the main chip.  It's a lot like 3dfx, actually - a
driver I worked on some back in the day.

3. session saving.  If it's missing especially on software that used to have it, it's a critical bug.

4. How can I make gnome 3 coexist with gnome2 until gnome3 is usable?   Since gnome3 is missing all of my dev tools, all of my console tools and everything I need to actually
function in my job, and it's all nicely handled in gnome2...   well it should be possible, yes?   (sadly, online docs aren't very helpful and my usual distro 'ubuntu' has become
as arrogant and rude as Apple.  So the advantage of working hardware (ubuntu after neuveau is purged) is negated by being stuck in an interface designed for an iphone)

5. If you're doing what unity is doing, it had better be usable in 360x240 (android) - since it certainly isn't usable anywhere except a phone.   Under Unity at least, the
desktop widgets all take up too much real estate on the screen.  Actually, while defaulting is fine, there should definitely be flexibility here.   Currently, from the web
capture reports I've seen, 1024x768 is still a good baseline.  This netbook is 1366x768, and it's brand new as of July/2011 with new graphics chipsets.  1680x960 and the like are
more desktop modes, and only if monitor + computer is new enough.

I could probably use some pointers at where to look at if any of these are pointing to the wrong list.   #1, #2 and #3 are all important from UX/interaction point of view at
least, #4 is probably elsewhere (anyone?) and #5 goes under UX policy.   My intense anger with unity has nothing to do with you all (I hope) - the thing's a complete violation of
user interface guidelines.   I've been programming since 1984-ish and with linux since 1993.   And I quite liked Gnome2 except that it didn't handle OpenGL or context switching
environments very well - and it certainly wasn't possible to embed gnome apps inside of an OpenGL 3D rendering engine.  (er call it a hobby project that I've been working on
across multiple platforms from Apple/6502 to modern)
- - Teunis
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk7VnbMACgkQllRoV5+NZ2bz+gCfbxUmvOCWa50TRSuQt6N2T23P
8asAnR+YmhDL4MlWXTmg40fn7qDg9lzs
=YdWN
-----END PGP SIGNATURE-----


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]