 |
|
| JumpStart | ||
|---|---|---|
| Prev | Next | |
JUMP presents a general methodology for schema integration and data object integration.
The JUMP model meets these two critical requirements of data integration; it solves the naming problem and it eases determination of equivalence between data sets. The example provided below captures a single database table attribute in the structured language.
...
<<element>>
<<semantic>>author<</semantic>>
<<presentation>>book_author<</presentation>>
<<physical>>bkauth<</physical>>
<<description>>the full and proper name of the person or group that wrote the identified book<</description>>
<<property:value>>proper name<</property:value>>
<<property:type>>string<</property:definition>>
<<property:field>>VarChar(30)<</property:filed>>
<<property:format>>last name, first name<</property:format>>
<<constraint:required field>>
<<vertical tree:book:1>>
<<vertical tree:title:2>>
<<vertical tree:pub-co:4>>
<<vertical tree:pubdate:5>>
<<horizontal relation:type=2 rule:required/>>
<<relation=table:"publisher"/>>
<<relation=table:"publish_date"/>>
<</element>>
...
These descriptive tags provide a formally understood structure in which to express the metadata nomenclature, description, format, properties, constraints, and rules in a very intuitive language.
This allows semantic objects to be easily mapped and equated by anyone, even if they did not create the model.
