From Meaningful Data
The context of a simple data element is the approach used to address it for handling.
Providing Context[edit | edit source]
There are four basic ways of providing a context for a simple data element:
- 1. Hierarchical - generally the approach used in object oriented languages, JSON, XML etc.
- 2. Relational - data is organized into rows and columns within tables, and tables relate to each other through foreign key columns. - generally the approach used in relational databases.
- 3. Hypertext - sort of like this wiki, free text pages contain words or phrases that are links to other pages.
- 4. Characterization - the meaning of the context of data is represented with one or more endemes.
- Relational and Hierarchical approaches usually result in tight coupling of fields to each other. With hierarchical approaches, members of classes are tightly coupled with each other in a class. Tight hierarchical coupling begins to provide implementation headaches when developers begin to use inheritance.
- With relational approaches, columns of tables are tightly coupled with each other in a table. Relational databases tend to have trouble dealing with high levels of complexity.
- Characterizing data removes or reduces the tight coupling problem, at the cost of speed and precision, but with the benefit of focusing on what is meaningful to the user.
Hard Coded Context[edit | edit source]
Relational programming hard codes the context. I see, set based programming is like relational rather than procedural The main problems with handling meaning in a program are tight field coupling and hard coded context tight field coupling means that the fields (columns in RDB) are tightly coupled to each other in a table In an object oriented language the members (fields) are tightly coupled to each other in a class