Re: MI ammunition : linked lists

[Marius Amado Alves <amado.alves@netcabo.pt>]

On Tue, 2003-11-11 at 18:38, Georg Bauhaus wrote:
> amado.alves <amado.alves@netcabo.pt> wrote:
> : <<Inheritance should reflect an "is a" relationship. Do you fill comfortable stating that a middle node is a first node and a last
> : node at the same time?...>>
> : 
> : I do saying a Node_Pointing_Both_Ways is a Node_Pointing_Forward and a Node_Pointing_Backward.
> : 
> : But I agree with the general advice. Thanks.
> 
> Maybe there is another clear picture, if you think of slightly
> more complex structure than a list. You can "ask" a node, "do
> you have a predecessor/successor?", but what question are you
> going to ask a graph node?
> For example, if a star is a node with indegree >= 5, does it
> make sense to employ any kind of ihneritance to reflect an
> indegree of 5?

The main question to be a asked a graph node is "what nodes are attached
to you?" The reply is a iterator over the corresponding nodes.

(Note another reply in this thread made the same claim for MI for linked
lists regarding names with different names than mine but with the same
exact meaning.)