Command Line Interface (CLI)
At the beginnings of the
computer era, CLIs were the only interface, which lead to the term “shell” as a
synonym for those kind of input types. Command line interfaces feature a single
line, in which users can, with help of a keyboard, type in specific commands to
start certain functions like copying, pasting, browsing, and so on. The whole
interface is exclusively represented by lines of text and therefore has a steep
learning curve and is hard to understand, especially for non-tech-savvy users.
Figure 1 shows a typical CLI, as known e.g. from DOS or
Windows Command Prompt.
 | |
| Figure 1: Windows Command Prompt |
Though the main interfaces of almost all operating systems shifted towards different types (see below), they still offer Command Line Interfaces as an additional input form for very specific commands and hence for more experienced users.
Graphical User Interface (GUI)
The concept of GUIs was
invented during the 1970s in California at Xerox PARC, but achieved its breakthrough
with the introduction of Apple’s Macintosh in 1984, the first affordable
computer on the market, which featured a graphical user interface. At first,
GUIs were very slow compared to their text-based ancestors and lacked useful
applications. Furthermore, the machines, which were able to show decent
graphics and had enough memory to avoid the annoying constant switching of
disks were still very expensive. But with increasing computing power and
display resolutions, more complex concepts could be realized while at the same
time prices decreased. Today, GUIs are the dominant form of representing
information and controls in operating systems, on personal computers as on
smartphones.
Via graphical elements and
controls, GUIs help the user to better control the operating system of his
computer and its applications. Those graphical elements are often metaphors of
real life (e.g. a recycle bin), so that the user intuitively knows what
function it serves because it looks familiar to known objects. Usually, a
computer mouse is used for pointing, selecting and manipulating visual object
on the screen, whereas the keyboard remains the primary text input device. Figure 2 shows the File
Explorer of Windows 7 with the typical windows layout and icons for
representing different functions/applications.
 | ||||
| Figure 2: Windows Explorer with highlighted Recycle Bin |
 |
| Figure 3: Windows Title Toolbar with minimize, change window mode and highlighted close button |
In those windows, the actual
functionality and data of the application is shown. Those contents also can
be enriched by certain GUI elements likebuttons, toolbars, scrollbars, or other
symbols. Furthermore, dialog boxes are
usually used when asking about user input or error boxes if some error
occurred. While designing those boxes, certain aspects like e.g. preciseness,
explicitness and understandability must be kept in mind to enhance the user
experience of the interface. For further coverage of Error Message Design, you should have a look at the blog posts of
Team 10, Team 17
and Team 20 about this very topic.
Natural User Interface (NUI)
Recent developments lead into a more human form of
interacting with a computer, i.e. the user doesn’t have to adapt his behavior,
so that the computer “understands” his commands. The user rather behaves
towards the computer like he would when interacting or talking to another
person. We want to pick Touchscreen and Gesture Recognition Interfaces as
examples for the first and Voice User Interfaces as examples for the latter.
Touchscreens
Touchscreens make it possible
for the user to directly interact with the surface of the displaying device. Whereas
former touchscreens used a stylus for input, recent touchscreens mostly got rid
of artificial input devices and rely on finger input (for further information
about receptive and capacitive touchscreen, see our former blog entry).
This not only solves the common problem of losing the input device, but also
allows for more intuitive and natural styles of interaction, especially
gestures. Gestures for doing certain commands follow gestures known from real
life, like e.g. swiping over the screen to turn a page, or crossing out a list
entry to delete it. Other gestures like e.g. pinch-to-zoom on the iPhone’s
multi-touch screen don’t have an equivalent in real life but provide a feeling
of interacting straight with the target object. Watch the following video for an example of the gestures in
the photo app of iOS, the operating system of the iPhone.
With the fingers as input
“device”, the graphical user interface has to be designed in a different way as
compared to GUIs which are operated with a computer mouse. Because the finger lacks
the preciseness of a mouse pointer, buttons, checkboxes, etc. have to be
bigger, and while interacting with the device, some parts of the screen are
covered and therefore not visible for the user at this certain point of time.
These are just two examples of a variety of issues designers have to address
when thinking of a touchscreen interface. In turn, using the fingers in
combination with a multi-touch screen allows for more flexible, complex and at
the same time more intuitive and simpler controls, where e.g. buttons are
omitted in favor of gestures.
Gesture Recognition Interfaces (GRI)
As we already mentioned gestures for executing certain commands on a touchscreen, we now want to get rid of the surfaces on which those gestures are done and concentrate on spatial gestures. A very well-known example of that technology is Microsoft’s Kinect together with the Xbox 360.
As we already mentioned gestures for executing certain commands on a touchscreen, we now want to get rid of the surfaces on which those gestures are done and concentrate on spatial gestures. A very well-known example of that technology is Microsoft’s Kinect together with the Xbox 360.
Kinect features a specific hardware device consisting of a
range sensor, a multi-array microphone and a camera (see Figure 4) whose data is
interpret by a proprietary software to gain a spatial image of the environment
and to track movements, facial expressions and voice of up to six users. This information is then used to control certain objects or
maybe a digital representation of the user, a so-called avatar, on a displaying
device.
 | ||
| Figure 4: Microsoft Kinect |
Initially, Kinect was designed solely for Microsoft’s
Xbox360 and games, but short time after its market launch, users wrote drivers
to make it also work on Windows 7 and Linux. Moreover, Kinect enjoys great popularity
among scientists and researchers. Microsoft also realized the potential of this
technology beyond games and is now heavily researching on new gesture
interfaces (see the video below) and even published an SDK for developers
to write their own Kinect applications.
Voice User Interfaces (VUI)
Another very interesting
development is the emergence of voice user interfaces. Coming from a
microphone, the underlying software recognizes the user’s voice and interprets
his commands. That is why VUIs usually come without any hierarchical menus as
known from GUIs and allow the user to directly express his wishes.
While the voice recognition
and interpretation software has to be error proof and able to deal with
different phrases and wording, the user also must know, how to communicate with
the device, that is, e.g. which commands and also languages the VUI can understand
and interpret.
In order to get an impression how voice user
interfaces can be implemented, you should check out Siri on Apple’s iPhone 4S,
as featured in the following video clip.
Conclusion
In the beginnings (ref. CLI), the user had to
adapt to the computer in order to communicate with it, like e.g. following strict
interaction guidelines, specific input languages and syntax or predefined input
fields. With new forms of user interfaces and input styles like voice and
gesture recognition, the communication between user and computer becomes more
human and resembles the interaction between people. It should therefore be
simpler and more intuitive.
However, there are also drawbacks. Because those
interfaces allow the user to interact in an open manner, they are more
error-prone then former types of input. The programs and algorithms behind
those interfaces must therefore not only follow design guidelines, but must
also process and correctly interpret huge amounts of data and at the same time be
as error proof as possible. As all of us already have experienced misunderstandings
while communicating and interacting with other people, this part seems to be a
tough challenge for future user interfaces.
Do have references for your blog posts!
ReplyDelete