Re: OOA, OOD, design all first, how to go about it all?
From: Joseph Jolic (jjolic_at_bigpond.net.au)
Date: 04/25/04
- Previous message: Universe: "Re: Aggregation vs composition"
- In reply to: AndyW: "Re: OOA, OOD, design all first, how to go about it all?"
- Next in thread: Phlip: "Re: OOA, OOD, design all first, how to go about it all?"
- Reply: Phlip: "Re: OOA, OOD, design all first, how to go about it all?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sun, 25 Apr 2004 09:44:44 GMT
This gells with me as well. there needs to be some sort of design to have
direction, and then the iteration. Other techniques are probably more
specialized and not what i really want to start now.
"AndyW" <foo_@bar_no_email.com> wrote in message
news:6gkm80lvtj1vgomijp0sq715icdnll8qdk@4ax.com...
> I believe that the general principle of software development has
> always been to do some level of analysis and a modicum of design at
> the higher level. At least enough so that you understand the problem,
> enough to know what ballpark you are in and to understand what game
> you are actually playing. You may even chose to do a small amount of
> development to test that you understand the rules of the game.
>
> Once you have achieved the goals of knowing the problem, having a
> tentative idea of the solution and understanding that your approach
> will at least work, then you can go down and use any technique you
> like. Whether it be a big bang software approach or small iterations
> it really doesnt matter.
>
> The main thing is to understand that you do need to reduce risk during
> the main development phase. Ideal ways to do this are to communicate
> with all parties often, dont do too much at a time, and prove that
> what you have is always at a working state.
>
> These principles have been around before the age of software
> development and are nothing new.
>
> The agile (or hack methods as I call them) seem to all be targeted at
> the small $10 projects or at least, try to convert the large scale
> systems into a collection of $10 projects. While they do often
> attempt to solve many of the problems during the main software cycle,
> I have often observed that people are quite often in the wrong ball
> park in the first place.
>
> Getting in the wrong ballpark and/or playing the wrong game is the
> general problem that has always dogged software development. There is
> no magical hack method that will ever solve this for you. But I can
> say that if you really do understand the problem and have a solid idea
> that your solution will work, then it is highly likely that you will
> succeed so long as you keep the risk to a minimum.
>
> Andy
- Previous message: Universe: "Re: Aggregation vs composition"
- In reply to: AndyW: "Re: OOA, OOD, design all first, how to go about it all?"
- Next in thread: Phlip: "Re: OOA, OOD, design all first, how to go about it all?"
- Reply: Phlip: "Re: OOA, OOD, design all first, how to go about it all?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|