Question about the way of specifying relationships

In the GLUE Specification v. 2.0, in the association end list, relations are always specified with the attribute they have to link with. (E.g. Service.ID) This is not correct because of several reasons: - In the Entity-Relationship models, you just specify relationship between entities/classes/objects, without any knowledge of each of the attributes in them. - To know which attribute you want to link with from the related object, you should have to take a look of its definition to get the key fields. - The key fields could be more than one. Sometimes the key of an entity is a combination of several fields, so the current way of visualizing relations will be broken. I suggest just taking out every attribute in all the Association End list to comply on how to use this kind of diagrams, to avoid problems in the future with multiple key fields. Regards, David -- David Horat Software Engineer specialized in Grid and Web technologies IT Department – Grid Deployment Group CERN – European Organization for Nuclear Research » Where the web was born Phone +41 22 76 77996 http://davidhorat.com/ http://cern.ch/horat http://www.linkedin.com/in/davidhorat

glue-wg-bounces@ogf.org [mailto:glue-wg-bounces@ogf.org] On Behalf Of David Horat said:
In the GLUE Specification v. 2.0, in the association end list, relations are always specified with the attribute they have to link with. (E.g. Service.ID)
As I said before, in glue the ID attribute *is* the unique key which identifies the object, so that is the thing which should be used in relations. The schema is not just some generic entity-relation model, it has very specific properties, and many of those are intended to make it easy to implement in LDAP (e.g. no tables). Actually the relational implementation has more problems, given the multivalued attributes and many-to-many relations which are easy in LDAP but harder in SQL. Stephen -- Scanned by iCritical.
participants (2)
-
Burke, S (Stephen)
-
David Horat