Book a Demo

Author Topic: Metaconstraints and classifier (Theoretical questions)  (Read 4262 times)

Modesto Vega

  • EA Practitioner
  • ***
  • Posts: 1183
  • Karma: +30/-8
    • View Profile
Metaconstraints and classifier (Theoretical questions)
« on: April 02, 2020, 10:20:41 pm »
I have 3 theoretical questions regarding metaconstraints.

If I set a metaconstraint between 2 stereotyped elements - e.g.  A and B - in an MDG profile with a UMLRole of classifier,

Q1) do A and B have to stereotype the same base type - i.e., class, actor, or activity?
Q2) is it advisable/sensible to mix base types?
Q3) would it be more sensible to use a metarelationship with a metaclass or Generalization?


Eve

  • EA Administrator
  • EA Guru
  • *****
  • Posts: 8110
  • Karma: +119/-20
    • View Profile
Re: Metaconstraints and classifier (Theoretical questions)
« Reply #1 on: April 03, 2020, 08:02:36 am »
1) The most common would be to use Object and Class. The object stereotype has a classifier metaconstraint to the class stereotype.
However, there are some situations where EA doesn't use object for instances. eg. Component, Actor, Use Case, Node. In that case your two stereotypes would have the same base type.

2) The best thing to do would be to use the same instance type EA would normally use to a given classifier type.

3) Only if you actually want generalizations.