Loading ...
Sorry, an error occurred while loading the content.

UML Profile for Data Modeling - Model Types and allowed Stereotypes

Expand Messages
  • k.drahmann
    Hi everyone, I read the article from Scott W. Ambler at http://www.agiledata.org/essays/umlDataModelingProfile.html I asked myself why the stereotype
    Message 1 of 3 , Feb 3, 2007
    • 0 Attachment
      Hi everyone,

      I read the article from Scott W. Ambler at

      http://www.agiledata.org/essays/umlDataModelingProfile.html

      I asked myself why the stereotype <<PK>> can't be applied to an
      attribute/column in a conceptual or logical model. When drawing an ER-
      diagramm with the Chen-Notation (which I think is a conceptual model)
      there can be primary key attributes. Similarly there can be foreign
      keys in logical models (say IDEF1X key based models), but again the
      defined mapping is only for physical models.

      What are the reasons why these mappings are defined this way?

      I don't know if this has been discussed before - I have not found
      anything. If there has been an earlier discussion about that please
      point me to the right direction. A book with a detailed explanation
      is also welcome.

      Thanks in advance,
      Kai Drahmann
    • Scott Ambler
      ... http://www.agiledata.org/essays/umlDataModelingProfile.html ... Yes, you can do that if you like. The general consensus within the data modeling community
      Message 2 of 3 , Feb 4, 2007
      • 0 Attachment
        --- "k.drahmann" <k.drahmann@...> wrote:

        > Hi everyone,
        >
        > I read the article from Scott W. Ambler at
        >
        >
        http://www.agiledata.org/essays/umlDataModelingProfile.html
        >
        > I asked myself why the stereotype <<PK>> can't be
        > applied to an
        > attribute/column in a conceptual or logical model.
        > When drawing an ER-
        > diagramm with the Chen-Notation (which I think is a
        > conceptual model)
        > there can be primary key attributes.

        Yes, you can do that if you like. The general
        consensus within the data modeling community is that
        issues such as primary keys are pretty much a physical
        modeling issue whereas candidate keys are pretty much
        a logical modeling issue. The profile reflects that.

        You can do whatever you like, however. The UML
        Modeling Police have only as much power as you give
        them. ;-)


        > Similarly there
        > can be foreign
        > keys in logical models (say IDEF1X key based
        > models), but again the
        > defined mapping is only for physical models.
        >
        > What are the reasons why these mappings are defined
        > this way?

        It reflects common industry practice.

        In the end, you can do whatever your toolset is
        able/willing to support.

        - Scott

        Scott W. Ambler
        Practice Leader Agile Development, IBM Methods Group
        http://www-306.ibm.com/software/rational/bios/ambler.html

        __________________________________________________
        Do You Yahoo!?
        Tired of spam? Yahoo! Mail has the best spam protection around
        http://mail.yahoo.com
      • Kai Drahmann
        ... I just don t want to reinvent the wheel and I think you ve spend much time and work in creating this profile. I want to use this profile for extending an
        Message 3 of 3 , Feb 4, 2007
        • 0 Attachment
          Am 04.02.2007, 13:48 Uhr, schrieb Scott Ambler <scottwambler@...>:

          >
          > --- "k.drahmann" <k.drahmann@...> wrote:
          >
          >> Hi everyone,
          >>
          >> I read the article from Scott W. Ambler at
          >>
          >>
          > http://www.agiledata.org/essays/umlDataModelingProfile.html
          >>
          >> I asked myself why the stereotype <<PK>> can't be
          >> applied to an
          >> attribute/column in a conceptual or logical model.
          >> When drawing an ER-
          >> diagramm with the Chen-Notation (which I think is a
          >> conceptual model)
          >> there can be primary key attributes.
          >
          > Yes, you can do that if you like. The general
          > consensus within the data modeling community is that
          > issues such as primary keys are pretty much a physical
          > modeling issue whereas candidate keys are pretty much
          > a logical modeling issue. The profile reflects that.
          >
          > You can do whatever you like, however. The UML
          > Modeling Police have only as much power as you give
          > them. ;-)
          >
          >
          >> Similarly there
          >> can be foreign
          >> keys in logical models (say IDEF1X key based
          >> models), but again the
          >> defined mapping is only for physical models.
          >>
          >> What are the reasons why these mappings are defined
          >> this way?
          >
          > It reflects common industry practice.
          >
          > In the end, you can do whatever your toolset is
          > able/willing to support.

          I just don't want to reinvent the wheel and I think
          you've spend much time and work in creating this profile.
          I want to use this profile for extending an open source
          UML tool to build SQL scripts from class diagramms that
          are using this profile. If I create my own profile (or
          derive it from yours) and yours is turning into a
          standard all my work was (perhaps) for nothing.

          Perhaps you could think a minute about the stereotypes
          <<PK>> in a conceptual data model so that it represents
          the same information as an ER-Diagramm in Chen-notation
          and <<PK>> and <<FK>> in logical data models for the
          equivalence to IDEF1X logical diagrams. Just to cover
          the case that an IDEF1X model designer wants to switch
          to UML.

          Kai Drahmann
        Your message has been successfully submitted and would be delivered to recipients shortly.