It's easy to just plunge into GIS work, doing what has to be done to get a map produced. But without a sound framework for managing your data and GIS projects, this approach can create serious long-term problems. Here are some guidelines for ensuring that your GIS work stays understandable and efficient - particularly when one GIS person leaves and a new person has to take over.

The best strategy is to organize your working projects and your data into two separate systems. Here's how:

PROJECT MANAGEMENT

Create a standard directory structure for each project you undertake.  GIS "projects" are more than just the actual project files that GIS software creates. By having a standard directory structure you can ensure that it's easy to find what you need and keep track of where you are in your project.  Here's a good starting place for project organization:

(add graphic of project directory)

Each project has its own directory ("MartinProject" in the example). You can create major groups of projects (Acquisitions, Fundraising, Conservation Plan, etc. - here shown as "TopicName") on top of the project level. Within the project folder, this is the framework:

• GIS projects go into folders based on the software in which they are created.
  
• Data created just for this project goes into the \data directory (but data you reuse stays in the data structure described below).
  
• General memos, reports, instructions and other documents are filed in the \docs&memos directory.
• If there are databases involved or you are creating illustration software (Illustrator, Freehand) files, they go into those directories.
• Images you create of your maps and any associated graphics go into the \images directory.
• Plot files (EPS, RTL files, etc.) go into \plots and are compressed when you are done, to save space.
• Any tables that you bring in or create are saved in that directory.

Even if you only use some of these directories, it's very good practice to always put your entire project directory structure into each project.

It's also a good idea to prepare a memo describing your project when you are mostly done with it, so that another person can pick up where you might have left off.

DATA MANAGEMENT

GIS data can grow very complicated. By placing all of your base data into a set structure you can balance complexity with clarity. Here's one example of a useful data directory structure:

(add data structure graphic)

• In this sample, data is organized by topic. In some cases, it may be important to organize first by jurisdiction (county or state), then by topic within each jurisdiction.
  
• If you regularly work with one jurisdiction (e.g., a county), you might end up adopting their data structure to facilitate data exchange.
  
• Regardless of the actual approach taken, it's important to take the time to document each GIS layer that you add to your data structure. Metadata (information about data) is a key element in any data management strategy. The simplest way to provide metadata is to include a short text (.txt) file with each GIS layer you have, describing what it is, what projection it is in, what its key attributes are, where it came from and any limitations about it.  A more robust approach is to create standard metadata files (use ESRI's ArcCatalog or another metadata generator) that are saved with the basic data. For more on metadata, visit the Federal Geographic Data Committee.
  
• If you have an existing data structure that you want to change, a good strategy is to simply start a new structure, leaving the old one intact so that projects done with it will still function.
  

Return to Advanced GIS Techniques <<<