Login | Register
My pages Projects Community openCollabNet

Discussions > Notification of Issues > Re: [argouml-issues] [Issue 744] Changed - Composition or Aggregation with 1..1 or 0..1 multiplicity should have a critic

argouml
Discussion topic

Back to topic list

Re: [argouml-issues] [Issue 744] Changed - Composition or Aggregation with 1..1 or 0..1 multiplicity should have a critic

Author "bob dot tarling" <bob dot tarling at ntlworld dot com>
Full name "bob dot tarling" <bob dot tarling at ntlworld dot com>
Date 2002-03-11 02:05:14 PST
Message Hi Markus

I think I prefer my option 1. Have an associations toolbar containing drop
downs for the styles at each end and a toolbutton to make it happen. Option
2 would be difficult to represent the different properties on each button.
I'll add option 1 as a seperate enhancement request. It can be discussed
further from there.

My original reasons for wanting this was not to do with critics but to do
with speed of drawing a diagram. I'd tend to draw many associations between
classes and then have to go back over them editing their properties. It
would be quicker to set the stanadrd properties first and then draw all
lines of that type.

As far as your suggested critic's concerned we'll have to agree to disagree
but I'm not going to push the issue. If others want it then fine.

I'd be happier when critics are plug-ins and the plug-in documentation makes
its purpose clear. I think standard critics in argo should just tell me when
I have broken a standard or suggest a pattern or approach I should take. Not
one that tells me a valid diagram has suggestions of things todo. I think
that's what you're asking for is an automated todo item which to me seems
different to a critic although the difference is blurred by the fact theat
they would be triggered in the same way..

Maybe critics could be catogorised by a short list of agreed descriptions
rather than just high medium low. eg Breaking Standard, Suggested Approach,
Todo Item etc. A user could then choose to turn on/off entire categories or
items in that category with a better understanding of the impact.

What do you think?

Cheers

Bob.

----- Original Message -----
From: "Markus Klink" <mkl at baba9 dot com>
To: <issues@argouml.t​igris.org>
Sent: Monday, March 11, 2002 9:16 AM
Subject: Re: [argouml-issues] [Issue 744] Changed - Composition or
Aggregation with 1..1 or 0..1 multiplicity should have a critic


>
>
> bob.tarling wrote:
>
> >I was thinking of suggesting one of these 2 features for Argo.
> >
> >1. Have drop down lists available on the toolbar for setting relationship
> >type (aggregation/composition/none) and multiplicity for each end.
Whenever
> >an association is drawn between 2 classes the association takes this
> >default.
> >
> >2. Have a configurable toolbar for which the user can add association
> >buttons of his preferred style.
> >
> >Either of these would remove the problem of the default 1..1
multiplicites.
> >
> >I don't like the idea of critics being there at all unless they genuinely
> >report a problem. The user can ignore the critic only if they have
decided
> >they know better in this case.
> >
> >The problem is for UML newbies lke myself. If I see a critic I often
think
> >the system knows better than me and my thinking is wrong.
> >
> But that does not have to be the case. It is a matter of explaining the
> critic better when it fires up. There is no reason to say "your design
> is wrong", but the critic could and should be used to help newbies
> instead of confusing them. It is a neat guide when employed correctly.
> If the critics convey the impression that they know better, than the
> critics must be reformulated, since they work on a pretty abstract
> basis, without knowing about the designer, his experience or even the
> intention of the system being modelled.
>
> In my opinion it still beats to have a dozen different buttons for each
> and everything. We should rather focus on making the ones we have work :-)
>
> Regards,
> Markus
>
>
> >
> >
> >----- Original Message -----
> >From: <issues@argouml.t​igris.org>
> >To: <issues@argouml.t​igris.org>
> >Sent: Monday, March 11, 2002 3:44 AM
> >Subject: [argouml-issues] [Issue 744] Changed - Composition or
Aggregation
> >with 1..1 or 0..1 multiplicity should have a critic
> >
> >
> >>http://argouml.tigri​s.org/issues/show_bu​g.cgi?id=744
> >>
> >>*** shadow/issues_4/744 Sun Mar 10 07:51:12 2002
> >>--- shadow/issues_4/744.tmp.16285 Sun Mar 10 19:44:09 2002
> >>***************
> >>*** 41,43 ****
> >>--- 41,65 ----
> >> Car 1 <---> Sunroof 0..1
> >>
> >> These seem quite valid to me.
> >>+
> >>+ ------- Additional Comments From mkl at tigris dot org 2002-03-10
> >>
> >19:44 -------
> >
> >>+ Bob, thanks for your valid comments. Your examples are all very valid
> >>+ however somehow in my opinion special cases where the critic should be
> >>+ snoozed away.
> >>+
> >>+ In my way of thinking aggregations/composition still tend to deal with
> >>+ whole/part relationships (whereas this is not always true) where part
> >>+ is something existing multiple
> >>+ times. It happened to me very often that I modelled something being
> >>+ more in the analysis phase were I do not tend to put a lot of
> >>+ multiplicities on the model. And invariably my
> >>+ hHouse --> room (1..*)
> >>+ Car --> tyres (4)
> >>+ Shop --> Shelve (1..*)
> >>+ Car --> Passengers (1..*)
> >>+
> >>+ relationships end up having 1..1 multiplicities (ArgoUML default). I
> >>+ would see this as
> >>+ an issue where a carefully crafted critic could be of help to remind
> >>+ me setting the multiplicities right. Someone who doesn't like it,
> >>+ could still disable the critic completely.
> >>
> >>------------​--------------------​--------------------​-----------------
> >>To unsubscribe, e-mail: issues-unsubscribe@a​rgouml.tigris.org
> >>For additional commands, e-mail: issues-help at argouml dot tigris dot org
> >>
> >>
> >
> >
> >----------------​--------------------​--------------------​-------------
> >To unsubscribe, e-mail: issues-unsubscribe@a​rgouml.tigris.org
> >For additional commands, e-mail: issues-help at argouml dot tigris dot org
> >
> >
> >
>
>
>
> --------------------​--------------------​--------------------​---------
> To unsubscribe, e-mail: issues-unsubscribe@a​rgouml.tigris.org
> For additional commands, e-mail: issues-help at argouml dot tigris dot org
>


--------------------​--------------------​--------------------​---------
To unsubscribe, e-mail: issues-unsubscribe@a​rgouml.tigris.org
For additional commands, e-mail: issues-help at argouml dot tigris dot org

« Previous message in topic | 4 of 5 | Next message in topic »

Messages

Show all messages in topic

[argouml-issues] [Issue 744] Changed - Composition or Aggregation with 1..1 or 0..1 multiplicity should have a critic issues at argouml dot tigris dot org issues at argouml dot tigris dot org 2002-03-10 19:44:09 PST
     Re: [argouml-issues] [Issue 744] Changed - Composition or Aggregation with 1..1 or 0..1 multiplicity should have a critic "bob dot tarling" <bob dot tarling at ntlworld dot com> "bob dot tarling" <bob dot tarling at ntlworld dot com> 2002-03-11 01:09:18 PST
         Re: [argouml-issues] [Issue 744] Changed - Composition or Aggregation with 1..1 or 0..1 multiplicity should have a critic mkl_foo Markus Klink 2002-03-11 01:16:12 PST
             Re: [argouml-issues] [Issue 744] Changed - Composition or Aggregation with 1..1 or 0..1 multiplicity should have a critic "bob dot tarling" <bob dot tarling at ntlworld dot com> "bob dot tarling" <bob dot tarling at ntlworld dot com> 2002-03-11 02:05:14 PST
                 Re: [argouml-issues] [Issue 744] Changed - Composition or Aggregation with 1..1 or 0..1 multiplicity should have a critic mkl_foo Markus Klink 2002-03-11 02:03:42 PST
Messages per page: