[kde-guidelines] Phrasing of selection teasers

[Diego Moya]

On 4 September 2013 20:54, Thomas Pfeiffer <colomar at autistici.org> wrote:

> On 04.09.2013 13:44, Diego Moya wrote:
> Do you have any data supporting that users are indeed confused?
> I haven't done any usability tests on the classic dual-list pattern yet,
> but with one full list and one empty list, I don't see how users
> would think they should interact with the empty list first.


Not hard data, but anecdotal evidence watching users struggle with the flow
to add and remove items, and being confused as to which list of elements if
the one being used. Some of the confusion may be dependent on the actual
application design, but the dual design definitely requires some time to
learn.

As for interacting with the empty list - finding an interface "intuitive"
depends on subtle visual cues more than on rational analysis (no shit,
Sherlock). In the dual list design proposed by the guideline [1] both lists
have the same visual design, which makes them symmetric and therefore
intuitively have the same weight; one has to actually 'think' about which
one to use (oh, the lists on the right is empty, I have to start with the
other one); there is no confusion, but you have taxed the user attention a
little bit.

If you provide an asymmetric design instead, giving more weight to the
thing that users have to interact with first, the first impression already
directs attention to the [Avaliable items] list without needing any
conscious thought. The Five Second Test [2] is a good way to measure
whether users will find the basic structure of a design intuitive and can
make sense of it without much effort; you might want to give it a try to
compare the various design alternatives.


[1] http://techbase.kde.org/Projects/Usability/HIG/DualList
[2] https://pidoco.com/en/help/ux/5-second-test



On 4 September 2013 20:40, Heiko Tietze <heiko.tietze at user-prompt.com>
 wrote:

> Your discussion is a nice example of the advantage of Personas: either he
> or
> she wants to have a very efficient tool (reduce the number of clicks) or
> prefer
> learnability.
>
> I agree, Personas make it easier to find out when it's better to recommend
separate interfaces for different use cases, and when it is possible to
include a one-size-fits-all widget.


However, building good Personas for an open source project is a daunting
task. I see that there's an initiative to build Personas for KDE4 [3], can
you tell me what's its overall state?

[3] http://techbase.kde.org/Projects/Usability/Principles/KDE4_Personas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-guidelines/attachments/20130905/72c3e8a2/attachment.html>