Interface design
Interface design principle
In the design stage, in addition to the design algorithm, data structure and other contents, a very important part is the design of the system interface, which is the interface of human-computer interaction, including how to command the system and how to submit information to the user. A well-designed user interface makes it easier for users to master the system, thus increasing users’ acceptance of the system. In addition, the system user interface directly affects the user’s mood when using the system, and the following situations will undoubtedly make the user feel bored and at a loss:
The gaudy interface makes it difficult for users to understand its specific meaning and don’t know where to start.
An ambiguous hint
A long reaction (more than ten seconds)
Extra operation (the user intended to do only this, but the system did something else in addition to this).
On the contrary, a successful user interface must be user-centered, integrated and interactive.
Although graphical user interface (GUI,Graphical User Interface) has been widely used and supported by many interface design tools, due to the above reasons, interface design should be placed in a very important position in the process of system development.
The strategy for designing a user interface consists of the following [Peter Coad,Edward Yourdon] :
Classification of people: by carefully studying the people who use the system, classifying them, including according to the level of skills (beginners, senior personnel … ), according to the organizational level (managers, general employees … ), according to the identity (staff, customers … ). Through classification, we can find out the purpose of each type of personnel using the system, and then we can determine its corresponding human-computer interaction.
Describe people and their task scripts: after classifying people, identify the characteristics of each category of people, including the purpose of using the system, characteristics (age, education level, limitations, etc.), expectations of the system (must / want, like / dislike / biased), proficiency, use of the system’s task script (scenario). According to these characteristics, we can guide the human-computer interaction design of the system.
Design command layer: the design of the command layer includes three aspects of work, namely, studying the moral and criteria of the existing user interaction; establishing an initialization command layer; and refining the command layer. In the design process of graphical user interface, some formal or informal guidelines and meanings have been formed, such as menu arrangement (for example, in almost all MS-Windows applications, the first three first-level menu items are always “file”, “edit” and “view”, while the last two are “window” and “help”), and some operations (for example, Open files, save files, print) graphical metaphors, and so on. Follow these guidelines to facilitate users to familiarize themselves with the system more quickly. When refining the command layer, we need to consider the arrangement, the whole-part combination, the comparison of width and depth, the minimum operation steps and so on. One level of too “deep” command items will make it difficult for users to find, while too many command items will make it difficult for users to master.
Design detailed interaction: there are several guidelines for the design of human-computer interaction, including:
一致性 采用一致术语、一致的步骤和一致的活动；
操作步骤少 使敲击键盘和点按鼠标的次数减到最少；
不要“哑播放” 长时间的操作需要告诉用户进展的状况；
闭包 用一些小步骤引出定义良好的活动，用户应该感觉到他们的活动中闭包的意义；
Undo 人难免做错事，通常在这种情况下系统应该支持恢复原状，或者至少部分的支持。
减少人脑的记忆负担 不应该要求人从一个窗口记忆或者写下一些信息然后在另一个窗口中使用；
学习的时间和效果 为更多的高级特性提供联机参考信息；
趣味和吸引力 人们通常喜欢使用那些感到有趣的软件。
继续做原型 通过做原型系统，可以直接的了解用户对设计界面的反应，然后进行改善，使之臻于完美。
设计用户界面类 在完成上面的工作后，就可以着手设计用户界面类。在开发GUI程序时，通常已经提供了一系列通用界面类，如窗口、按钮、菜单等等，只要从这些类派生特定的子类即可。
根据图形用户界面进行设计 目前主要的GUI包括Windows，Macintosh，X-Windows，Motif等，基于它们开发应用软件可以使界面的设计简单化，但是事先要清楚其特性，如事件处理方式等等。
Elements in GIS Interface Design
For successful GIS software, a good interface is indispensable, in the design, we should also follow the above principles and steps. For GIS software, its interface needs to allow users to select and retrieve the corresponding spatial data, manipulate the data, and represent the results of the analysis. For basic data retrieval, operation and performance, which is consistent with ordinary software, the following elements should be considered in GIS:
Data selection
The filter used to select data may include spatial and non-spatial attributes, or a combination of both, for example (for a county administrative division data):
Retrieve all counties with a population greater than 100000
Retrieve all the counties through which railways pass
Retrieve all counties with railways and a population of more than 100000
Users can enter a command statement, select from a menu, fill out a form (Form), or select data through direct actions, such as using a mouse. Using command statements requires understanding the structure of the data table and extending the traditional SQL to support spatial filtering. Direct operation selection requires data to be displayed on the screen, and the combination of it and SQL queries is usually done indirectly. When some figures are displayed in a similar position, the direct operation selection will be ambiguous. In order to facilitate users to operate, it is necessary to provide some additional operations, such as roaming, magnifying display, zooming out display and so on.
Data performance
A good form of data representation is beneficial for users to operate directly for further analysis. compared with the standard relational database, more consideration should be given to its graphic display in GIS. The similarities or differences in graphic performance can be expressed to the user that these figure objects have some degree of similarity. Some variables shown in the graphics are:
Polygon outline: color, grayscale, black and white; linetype
Polygon fill: color, grayscale, black and white; fill mode
Lines: color, grayscale, black and white; linetype
Symbols: color, grayscale, black and white; shape; size, etc.
For graphic display, it needs to be carefully designed to correctly express the meaning of figure objects and be understood by users.
Data processing.
Data processing consists of a series of spatial and non-spatial operations, and a well-designed interface makes it easier to implement these operations. Compared with the standard relational database, the data managed by GIS has more object-oriented characteristics, so an object-oriented interface is conducive to the interaction between users and the system to complete data processing. In GIS software, object-oriented interface design includes the representation of geographic entities, such as points, lines, polygons and some operations as pictographic symbols, and users can achieve the corresponding data processing through simple click, drag and drop and other operations.
Some operations with graphical implications are listed below, and it is easy to associate the graphical representation with the corresponding operations.
Create: creates a spatial object based on a given graphical entity and corresponding attributes
Delete: delete a selected space object
Collection: forms a collection based on selected objects
Update: displays only the results of the last operation
Superposition: equivalent to a union operation in a collection operation
Intersection: equivalent to an intersection operation in a set operation
Difference: equivalent to a difference operation in a set operation
Conversion: scale, move, mirror, affine transformation and other operations on the selected figure
Checkpoint: set a checkpoint to which you can fall back when you are not satisfied with your later work
Echo: returns the status of the previous checkpoint
Submit: send all processing results to the database and update them.
SQL
Traditional SQL can not handle spatial query, which is due to the weakness of relational database technology. For GIS, it is necessary to extend SQL. SQL/MM, which is currently being developed, is mainly used in multimedia data, which contains a comprehensive set of GIS operations.
The adoption of the new SQL standard has brought about a change in concept. For traditional SQL, to achieve spatial operations, it is necessary to embed SQL commands into a programming language, such as C, while the new SQL allows users to define their own operations and embed them into SQL commands.
This extended SQL actually adds object-oriented support. In GIS, it is called GeoSQL,GeoSQL, which can be written as follows:
SELECT Soils.Map
From Soils,Parcels
Where Parcels.Value>6000 and Overlay(Soils,Parcels);
The query command of.
As it is still quite difficult to fully implement GeoSQL, the current implementation idea of most GIS software is to input standard SQL query statements and spatial queries respectively (figure 16-14), then deal with them separately, and finally merge the results, which is not flexible enough.
Figure 16-14: an interface that implements spatial queries, entering spatial relationships and standard SQL separately
Visualization
Because GIS is based on graphics, the results of its analysis and interpretation are usually shown in the form of visualization. Visualization refers to the process of generally expressing information in order to identify, communicate and explain patterns or structures. Spatial analysis needs to consider the perception of information patterns and spatial characteristics. For GIS, visualization can be described as a process of transformation from information to knowledge. For GIS, in addition to representing all kinds of information in a visual form, realizing the expression of WYSIWYS (What You See Is What You Get) is also an important principle of interface design.
GIS interface style
In the interface design of GIS software, there are three basic user interface styles, namely, command-based interface, menu-driven interface and workflow-based GUI interface. These three interfaces have their own advantages and disadvantages for implementation and use. In specific implementation, one or more styles can be supported at the same time.
GIS interface based on the command line (figure 16-15)
The command line is the simplest interface style and has long been used in a variety of operating system software. It uses only text language and requires users to know the options available, which requires memorizing various commands or constantly looking up help documents. The command-line-based interface does not provide any tips or suggestions, which makes users rely on printed documents to learn the system. Using the command line interface requires the development of a command line interpreter. In the command line interface software, the relationship between functional modules is relatively simple, often the output of one module is used as the input of another module, which is easy to develop and implement. Using batch command files or script files, you can perform multiple steps in turn, which is the advantage of the command line interface.
For GIS software, because it contains a large number of graphical operations, when using the command line interface, there needs to be a graphical window to display the results of the operation, so the command line interface acts as a console. Because batch commands and script files are supported, the command line interface can be used to achieve batch, procedural, and time-consuming data processing.
Figure 16-15: GIS interface based on the command line
Menu-driven GIS interface (figure 16-16)
After MS-Windows has become the mainstream operating system on PC, menu-driven user interface has been adopted in almost all application software. It lists all the actions provided by the system according to the hierarchy, and the user can select and perform an action through the keyboard or pointing device, usually the mouse. Each menu item has corresponding help information, which is easy for users to refer to at any time.
The biggest advantage of menu-driven interface is that it is friendly and easy for users to master the system. But for advanced users, it is often inflexible and inefficient compared with the command line interface. In GIS, batch data often need to be processed continuously, and it takes a long time to calculate. In this case, the menu interface becomes intolerable.
Figure 16-menu-driven GIS interface in 16:Windows environment
GIS graphical user interface using data flow diagram (figure 16-17)
In the graphical data flow interface, the operator controls the system through a “visual language” rather than strict text, in which the system uses graphical symbols to represent the functions it provides, called “icons”. Icons can represent not only operations, but also data or hardware devices.
In the user interface based on data flow graph, users can use the “Drag and Drop” operation to achieve the corresponding operation (for example, you can drag and drop the icon representing the interpolation calculation onto the icon representing isoline data to perform isoline interpolation), design and organize the data processing flow.
The data flow graph interface is suitable for systems with clear and relatively simple data flow, such as digital image processing software, but its disadvantage is that it is difficult to implement.
Figure 16-17: GIS graphical user interface with data flow diagram