Book a Demo

Author Topic: Generic relationships  (Read 7172 times)

Paolo F Cantoni

  • EA Guru
  • *****
  • Posts: 8626
  • Karma: +259/-129
  • Inconsistently correct systems DON'T EXIST!
    • View Profile
Generic relationships
« on: April 18, 2016, 05:51:52 pm »
I can't recall if the ArchiMate specification (or even UML) says this but for the sake of the discussion, let's assume that any arbitrary element metatype can be an aggregate of the same element metatype .


I can't actually specify that in Quicklinker, without specifying that each specific type aggregates to that specific type.  I can create (for example) a generic relationship definition  unstereotyped class to unstereotyped class but that would allow aggregation between different metatypes =- which may or may not be valid.


I thought one solution would be to allow the stereotype "<stereotype>" to mean an arbitrary stereotype - reuse of that extrinsic value in the same line would mean the same stereotype (and therefore the same metatype) and get around the problem and make [/size]Quicklinker definition files much smaller and easier to manage.


Thoughts?
Paolo
Inconsistently correct systems DON'T EXIST!
... Therefore, aim for consistency; in the expectation of achieving correctness....
-Semantica-
Helsinki Principle Rules!

qwerty

  • EA Guru
  • *****
  • Posts: 13584
  • Karma: +397/-301
  • I'm no guru at all
    • View Profile
Re: Generic relationships
« Reply #1 on: April 18, 2016, 07:01:20 pm »
How can something be aggregate of itself?

q.

Paolo F Cantoni

  • EA Guru
  • *****
  • Posts: 8626
  • Karma: +259/-129
  • Inconsistently correct systems DON'T EXIST!
    • View Profile
Re: Generic relationships
« Reply #2 on: April 18, 2016, 11:03:54 pm »
How can something be aggregate of itself?

q.
metatype, NOT instance.


This team is an aggregate of other teams... (for example)


Paolo
Inconsistently correct systems DON'T EXIST!
... Therefore, aim for consistency; in the expectation of achieving correctness....
-Semantica-
Helsinki Principle Rules!

qwerty

  • EA Guru
  • *****
  • Posts: 13584
  • Karma: +397/-301
  • I'm no guru at all
    • View Profile
Re: Generic relationships
« Reply #3 on: April 18, 2016, 11:31:52 pm »
That would probably be an easy improvement. +1

Don't ask me what I think of the QL handling in general. Waste bin and then design "something". It can't get worse.

q.

Glassboy

  • EA Practitioner
  • ***
  • Posts: 1367
  • Karma: +112/-75
    • View Profile
Re: Generic relationships
« Reply #4 on: April 19, 2016, 07:12:12 am »
I can't recall if the ArchiMate specification (or even UML) says this but for the sake of the discussion, let's assume that any arbitrary element metatype can be an aggregate of the same element metatype .

Code: [Select]
Furthermore, aggregation, composition, and specialization relationships are always permitted between two elements that have the same type.  Section 2.4 http://pubs.opengroup.org/architecture/archimate2-doc/chap02.html#_Toc371945148

Glassboy

  • EA Practitioner
  • ***
  • Posts: 1367
  • Karma: +112/-75
    • View Profile
Re: Generic relationships
« Reply #5 on: April 19, 2016, 07:13:44 am »
How can something be aggregate of itself?

To understand recursion, you must first understand recursion.   ;D

Paolo F Cantoni

  • EA Guru
  • *****
  • Posts: 8626
  • Karma: +259/-129
  • Inconsistently correct systems DON'T EXIST!
    • View Profile
Re: Generic relationships
« Reply #6 on: April 19, 2016, 09:24:32 am »
How can something be aggregate of itself?

To understand recursion, you must first understand recursion.   ;D
You might well think that, I couldn't possibly say that...   ;)


Paolo
Inconsistently correct systems DON'T EXIST!
... Therefore, aim for consistency; in the expectation of achieving correctness....
-Semantica-
Helsinki Principle Rules!