Re: creaping coupling......
From: Mark Nicholls (nicholls.mark_at_mtvne.com)
Date: 02/25/05
- Next message: Darius Zakrzewski: "Re: Application logic and Business logic"
- Previous message: Mark Nicholls: "Re: creaping coupling......"
- In reply to: kirkk: "Re: creaping coupling......"
- Next in thread: Thomas Gagne: "Re: creaping coupling......"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Fri, 25 Feb 2005 11:41:08 -0000
"kirkk" <kirk@kirkk.com> wrote in message
news:1109284712.501513.315590@g14g2000cwa.googlegroups.com...
> I see little wrong with the initial design, aside from the fact that C
> is coupled to B. Make C coupled to IB, which also indirectly couples it
> to IA. But C is decoupled from all the B's and A's in the world. It can
> be compiled separate from them, tested separate from them, and used
> only with those you desire.
now change something in IA.........you should logically only have to change
code that accesses a method in IA......but your tool will force you to
compile all references to IA......bad....incidental coupling.......I think
we all agree incidental coupling is the root of much evil.....
now deploy C.....oh dear you need to deply IA with C.....even though C never
uses it....
do this recursively....the example I have given is deliberately
trivial....real world system are several orders more complex.
>
> And if you place IB and IA in the same package or physical unit as C,
I don't do this, I have never done this.....I always deploy in a seperate
package to both client and server....the interface 'belongs' to noone.
It still applies.
> but separate the implementations, you'll gain even more flexibility.
> Put your interfaces close to the classes that use them. Separate your
> interfaces from the classes that implement them.
>
I do...and from the classes that use them......not to do so limits
you.........I don't want to revisit DIP, it doesn't apply.....(and I believe
it is at best suboptimal).
- Next message: Darius Zakrzewski: "Re: Application logic and Business logic"
- Previous message: Mark Nicholls: "Re: creaping coupling......"
- In reply to: kirkk: "Re: creaping coupling......"
- Next in thread: Thomas Gagne: "Re: creaping coupling......"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|