Re: Default Desktop Experience for 11.04 - User testing results

2011-04-19 Thread Matthew Paul Thomas
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Jono Bacon wrote on 16/04/11 20:05:

 On Fri, 2011-04-15 at 03:00 +0100, Matthew Paul Thomas wrote:
...
 Last week, Charline Poirier ran a user test of Unity, with 11
 individual participants. This week, I have helped Charline analyze the
 results.
...
 Wonderful work, and now very visible work:
 http://news.slashdot.org/story/11/04/16/0239213/5-Out-of-11-Crashed-Unity-In-Canonicals-Study
  ;-)

Up to Slashdot's usual standards, I see.

To correct the obvious from that post:

*   No, the results of the study have not been published. Charline
will do that soon.

*   No, my name is not Rick Spencer. (rickspencer4?)

*   The object of the study was, obviously, not to measure crashes.
Crashes are usually quick to find and fix, so any user test of those
would be weeks out of date when published. I mentioned them only as
a reminder that to users, bugs are indistinguishable from design
flaws, and vice versa. (For example, one test participant pressed
Ctrl Alt F1 apparently by accident, and ended up at a console. This
wasn't a crash, but it had exactly the same effect as one.)

 I think this feedback points to a series of design and engineering bugs
 that we need to resolve in 11.10. Have the design bugs been filed in
 Launchpad?

Charline has been working with John Lea on that today.

 I think it could be worthwhile to rate the prioritization of the design
 bugs based upon the level of success in your study. As an example, if
 1/12 achieved a task, it would be a high priority bug, as opposed to if
 10/12 achieved the task it would be a low priority bug.
...

I'll pass that on to John.

Cheers
- -- 
mpt
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk2tokcACgkQ6PUxNfU6ecr1TgCePApLBm/ySLUAasdjINBrWQHs
LKYAoNLXk9mfWvVzz9fF7h4rl3eBq9u3
=j4k8
-END PGP SIGNATURE-

-- 
ubuntu-desktop mailing list
ubuntu-desktop@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-desktop


Re: Thoughts about Unity and some ideas for improvement

2011-04-19 Thread Jan Claeys
Jo-Erlend Schinstad schreef op di 19-04-2011 om 15:56 [+0200]:
 2: workspaces as contexts
 
 Workspaces doesn't currently seem very useful to me. It can be nice if
 you have windows layed out in a certain way, but other than that, I
 don't think I'll use them very much. I think they would be very much
 more useful and user friendly if the super-w didn't display windows
 from other workspaces, but was limited only to the current workspace
 like alt+tab is.

You can fix this with CCSM by changing the keybindings for the compiz
Scale plugin if you want.


  I always use workspaces as contexts. That is, I
 have a workspace where I do general stuff like surfing and chatting
 with friends for no particular purpose. I also have one workspace that
 I use when I play guitar and sing, etc, I have one for managing my
 network of physical and virtual machines and one I use for
 development. I work with different files and websites in these
 contexts so I open new instances of Nautilus and Firefox. This is
 probably the only time I open more than one instance of these
 applications since both supports tabs very well.

 For my workflow, it would be extremely useful if the launcher and
 super-w was context aware so that pressing super-1 would open only
 the Nautilus relevant to my current work and super-2 would switch to
 my relevant Firefox. That is to say that if I'm in my general context
 with Firefox and Nautilus open, then switch to a new context, pressing
 super-1 and super-2 would open new instances of those applications,
 and then later switch between those instances in this context only.
 When switching between contexts, the default browser in the context
 you switch to should be set as the one to open links. (This has been a
 problem for me for ages). Opening a link from gnome-terminal in my
 Guitar context should never result in the link being opened in my VM
 Management context, for instance. This helps me stay focused on my
 current tasks.

+1000 from me on fixing that behaviour for Firefox etc.  :)

(I think it might require more workspace-awareness by the applications
themselves and/or maybe the GUI toolkits though?)


-- 
Jan Claeys


-- 
ubuntu-desktop mailing list
ubuntu-desktop@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-desktop


Re: Default Desktop Experience for 11.04 - User testing results

2011-04-19 Thread Jeremy Bicha
On 19 April 2011 15:13, Jan Claeys li...@janc.be wrote:
 Matthew Paul Thomas schreef op di 19-04-2011 om 15:55 [+0100]:
 *   The object of the study was, obviously, not to measure crashes.
     Crashes are usually quick to find and fix, so any user test of those
     would be weeks out of date when published. I mentioned them only as
     a reminder that to users, bugs are indistinguishable from design
     flaws, and vice versa. (For example, one test participant pressed
     Ctrl Alt F1 apparently by accident, and ended up at a console. This
     wasn't a crash, but it had exactly the same effect as one.)

 Maybe we need to add a line of text above the login prompt somehow, that
 tells the user what key to press to get back to their GUI?  (This might
 be difficult to do correctly with multiple logins etc. though?)

We could do like Fedora and have X on virtual terminal 1. Why do we
need 6 virtual terminals anyway?

Jeremy

-- 
ubuntu-desktop mailing list
ubuntu-desktop@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-desktop