comp.lang.ada
 help / color / mirror / Atom feed
From: "Jean-Pierre Rosen" <rosen.adalog@wanadoo.fr>
Subject: Re: Popular Design Diagram Methods For Ada
Date: 2000/02/05
Date: 2000-02-06T09:56:59+00:00	[thread overview]
Message-ID: <87jghb$oj5$2@wanadoo.fr> (raw)
In-Reply-To: 389B1783.473576F5@quadruscorp.com

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2939 bytes --]


Marin D. Condic <mcondic-nospam@quadruscorp.com> a �crit dans le message :
389B1783.473576F5@quadruscorp.com...
> Jean-Pierre Rosen wrote:
> > If that's what you want, then you should *really* have a look at HOOD.
The
> > leading ideas  behind the method are "separate concerns" and "manage
> > complexity".
> >
> Well, I've always felt that it is very important to have some sort of
> overall "block diagram" of a large system so that you can talk about the
> big issues. You also need some ability to keep fleshing that out with
> more and more details until you can see where all the rivets and bolts
> are located. Having the ability to zoom in and out to various levels of
> complexity is something I've always wanted in a design tool, but in most
> of the things I've looked at, this is only supported weakly. Of course
> it has been a number of years since I last went through this exercise of
> trying to evaluate design methodologies and tools, so maybe things have
> improved some.
HOOD is for large designs, but you NEVER see all the details of the whole
structure. Remember that the 'H' in the name stands for "Hierarchical".
Either you see a few main blocks (but not what's inside), or you zoom into a
module and see its internals, but you don't see the outside anymore (except
for the connections to the outside world, but I won't give the whole
tutorial here :-)

> If I can find a document describing HOOD, I'll definitely read it over.
> It sounds like you are saying the methodology itself supports a layered
> view of a large system. I'm hoping that despite the name, it is not
> limited to describing only Object Oriented Design, since I've got to
> cover software that is not structured this way.
>
As I mentionned in the previous message, the official document is the HOOD
book ("HOOD: an Industrial Approach for Software Design", by HOOD User Group
and (hum) J-P. Rosen). Direct link:
http://pro.wanadoo.fr/adalog/hoodbook.htm. Note that you can view the
"overview" chapter of the book directly off this page, so it would give you
an idea.

The "OOD" in the name refers mainly to Booch-1 ("Software Engineering with
Ada") OOD, that is mostly composition-oriented, although there is now
support for inheritance,  but certainly not as a major direction. Over the
time, HOOD became less directive about the design process, so that it can
now be used with various design approaches.

(After reading you next message).
HOOD supports virtual nodes for the distribution of applications. You design
your application independently from distribution, then project it over a
logical structure of virtual nodes, which in turn get assigned to physical
nodes. Note BTW that this process matches very well with Annex E (although
designed independently)

--
---------------------------------------------------------
           J-P. Rosen (Rosen.Adalog@wanadoo.fr)
Visit Adalog's web site at http://pro.wanadoo.fr/adalog










  reply	other threads:[~2000-02-05  0:00 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2000-02-02  0:00 Popular Design Diagram Methods For Ada Marin D. Condic
2000-02-02  0:00 ` Ted Dennison
2000-02-03  0:00   ` Chris M. Moore
2000-02-03  0:00     ` Marin D. Condic
2000-02-04  0:00       ` Jean-Pierre Rosen
2000-02-04  0:00         ` Marin D. Condic
2000-02-05  0:00           ` Jean-Pierre Rosen
2000-02-03  0:00 ` Jeffrey D. Cherry
2000-02-03  0:00   ` Marin D. Condic
2000-02-04  0:00     ` Jean-Pierre Rosen
2000-02-04  0:00       ` Marin D. Condic
2000-02-05  0:00         ` Jean-Pierre Rosen [this message]
2000-02-07  0:00         ` Pierre Dissaux
2000-02-08  0:00           ` Marin D. Condic
2000-02-08  0:00             ` Jean St-Pierre
2000-02-04  0:00     ` Jeffrey D. Cherry
2000-02-04  0:00       ` Marin D. Condic
2000-02-07  0:00         ` Jeffrey D. Cherry
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox