The concepts of lifecycle planning can be applied to projects of lesser scale and scope, particularly to those pursued in undergraduate and graduate research. This does not mean that every project will move through every step outlined above. Some steps such as benchmarking and system selection may be irrelevant in a setting where the researcher must make do with whatever equipment and software is on hand. But lifecycle planning should not be viewed as a series of boxes on a checklist, it is a process of careful planning and problem solving. It is this process of careful planning that should be emulated regardless of the scope or scale of a project.
This point is not always understood. Some researchers reject the methodology of project planning because it seems overly formal and stringent given their modest research goals. Instead, they improvise a GIS solution. But improvised solutions are always a risk. Attention to the process of careful planning can waylay such risks. Perhaps the essence of this process can be summarized in three points.
1. Think ahead to how the GIS will be used, but keep in mind what sources are available.
Designing an effective GIS involves setting clear goals. The temptation is to rush ahead and begin digitizing and converting data without establishing how the system will be used. Even for small GIS projects, it is wise to engage in a modest functional requirements study. This allows the user to gain an idea of exactly what data sources are required, how they will be processed, and what final products are expected. Without clear-cut goals, there is too great a danger that a project will omit key features or include some that are irrelevant to the final use.
2. Exert special care in designing and creating the database.
Again, it is easy to rush ahead with the creation of a database, and then find later that it has to be reorganized or altered extensively. It is far more economical to get things right the first time. This means that the researcher should chart out exactly how the database is to be organized and to what levels of accuracy and precision. Attention to (and testing) of symbolization and generalization will also pay off handsomely.
3. Always develop a prototype or sample database to test the key features of the system.
No matter the size of a project, the researcher should aim to create a prototype first before moving toward full implementation of a GIS. This allows the researcher move through all of the steps of creating and using the system to see that all procedures and algorithms work as expected. The prototype can be a small area or may be confined to one or two of the most critical layers. In either case, testing a prototype is one step that should not be overlooked.
This point is not always understood. Some researchers reject the methodology of project planning because it seems overly formal and stringent given their modest research goals. Instead, they improvise a GIS solution. But improvised solutions are always a risk. Attention to the process of careful planning can waylay such risks. Perhaps the essence of this process can be summarized in three points.
1. Think ahead to how the GIS will be used, but keep in mind what sources are available.
Designing an effective GIS involves setting clear goals. The temptation is to rush ahead and begin digitizing and converting data without establishing how the system will be used. Even for small GIS projects, it is wise to engage in a modest functional requirements study. This allows the user to gain an idea of exactly what data sources are required, how they will be processed, and what final products are expected. Without clear-cut goals, there is too great a danger that a project will omit key features or include some that are irrelevant to the final use.
2. Exert special care in designing and creating the database.
Again, it is easy to rush ahead with the creation of a database, and then find later that it has to be reorganized or altered extensively. It is far more economical to get things right the first time. This means that the researcher should chart out exactly how the database is to be organized and to what levels of accuracy and precision. Attention to (and testing) of symbolization and generalization will also pay off handsomely.
3. Always develop a prototype or sample database to test the key features of the system.
No matter the size of a project, the researcher should aim to create a prototype first before moving toward full implementation of a GIS. This allows the researcher move through all of the steps of creating and using the system to see that all procedures and algorithms work as expected. The prototype can be a small area or may be confined to one or two of the most critical layers. In either case, testing a prototype is one step that should not be overlooked.
No comments:
Post a Comment