Organizing a database

I’m still a bit confused regarding the basic principles to use when organizing a database. Here’s a simple example to show what I mean.

My chief interests are writing and educational technology. Many files intersect these areas. Two possible filing strategies occur to me. If we assume that a new file concerning technology to use in writing education is to be filed, this is how things would work out:

  1. Have 2 main folders, (i) writing, and (ii) educational technology. Make a replicant, and file the 2 instances in the two folders.

  2. Have 2 main folders as above, but also make an educational technology sub-folder in the writing folder. The example file would go in that subfolder only.

The first strategy would tend to have lots of folders at a high level in the database and fewer levels of hierarchy. The second strategy would have a more elaborate hierarchy. In the first strategy, replicants would be the only way of showing intersections of subject areas; in the second, the database creator would be spelling it out through folder structures.

Is there a right or wrong way to do things? And even if it’s personal preference, are there more or less predictable consequences to choosing one strategy over another?

Another possibility is to…

  • create two main groups (“writing” and “educational technology”)
  • create an additional group for all the stuff which should be filed in both main groups
  • move this additional group into “writing” and finally replicate this group into “educational technology” too

This approach has some advantages:

  • Compared to your first suggestion, it’s only necessary to file documents once
  • Compared to your second suggestion, both main groups contain all related documents and each group has a unique name

It hadn’t occurred to me that groups could be replicated in this way. That opens up some exciting new possibilities.