Comparing OMT and Booch's method
Helge Minken
Designs in our different instrument designs. We will to the
most extent use VC++ 4.
Two design methods have been launched: The Booch method and
the Object Modeling Technique (OMT). Can anybody advise me of
which (if any) is the best chose, or if any other technique may
be even better. What are their pro's and con's?
thanks for your effort,
Helge
John Nowak
You can get an outline of the new unified method proposed by G. Booch,
and J. Rumbaugh by sending your mailing address to Becky <bec...@rational.com>.
We are using Rational Rose - C++ and can design with either Booch or OMT. The
software will convert between the two notations so you can view the design either
way.
Good Luck,
John Nowak
John Nowak
Kenneth Lee
Hi,
check out the Unified Method or whatever the name will end up being.
This is the combined Booch and OMT, combining the best from both into
one. You can read more about it in the January edition of JOOP.
Gudjon (borrowing Ken's account)
Robert C. Martin
We are in the very beginning of introducing Object Oriented
Designs in our different instrument designs. We will to the
most extent use VC++ 4.
Two design methods have been launched: The Booch method and
the Object Modeling Technique (OMT). Can anybody advise me of
which (if any) is the best chose, or if any other technique may
be even better. What are their pro's and con's?
In a recent poll of Object Magazine and the C++ report, readers were
asked which methodology they used. Over 50% use Booch, and 31% - 34%
use OMT. Coad/Yourdon was next with 16%-20%...
Popularity is not necessarily a predictor of quality. But it is still
an interesting statistic.
In your case, the choice is somewhat moot. Booch and Rumbaugh (OMT)
have recently released a draft of their "unified" notation and method.
Booch and OMT have always been very similar. These two methodologists
recognized about a year or so ago that it would be to everyone's
advantage if they could resolve their minor differences and present a
common method and notation.
The new notation looks a lot like OMT. Most of the Booch
idiosyncrasies are gone. However, the new notation is much more
expressive than OMT was, having many of the extra things that
Booch-lovers enjoyed.
So, my recommendation to you, is to get hold of the unified method
documentation. (You can get it from Rational (408.496.3600) and start
from there.
--
Robert Martin | Design Consulting | Training courses offered:
Object Mentor Assoc.| rma...@oma.com | OOA/D, C++, Advanced OO
14619 N. Somerset Cr| Tel: (708) 918-1004 | Mgt. Overview of OOT
Green Oaks IL 60048 | Fax: (708) 918-1023 | Development Contracts.
Grady Booch
check out our web page at www.rational.com to obtain a copy of
the unified method document. currently, we are only shipping via snail mail,
but will eventually distribute electronic copies
--
egb
Bill Dugan
>Robert C. Martin (rma...@oma.com) wrote:
>> So, my recommendation to you, is to get hold of the unified method
>> documentation. (You can get it from Rational (408.496.3600) and start
>> from there. ^^^^^^^^^^^^
> Is this an IP address?
It looks like a U. S. telephone number.
Jeff Fulton
We are in the very beginning of introducing Object Oriented
Designs in our different instrument designs. We will to the
most extent use VC++ 4.
Two design methods have been launched: The Booch method and
the Object Modeling Technique (OMT). Can anybody advise me of
which (if any) is the best chose, or if any other technique may
be even better. What are their pro's and con's?
thanks for your effort,
Helge
============================================================
Helge,
This is a good and common question. I think that for all practical
purposes, Booch and Rumbaugh are essentially the same. Some users
like one, some the other.
Rumbaugh is conceptually simpler than Booch. This makes it easier to
understand for novices. Booch, on the other hand does a more complete
job of representing all entities that a typical software project
interacts with and so is more complete.
Currently, more students are coming out of school having learned
Rumbaugh, but Booch is also commonly learned.
So, all in all, it is a matter of taste I suppose.
If you are interested in looking at a design and reengineering tool
that supports both methdologies (and can reverse C++ code into them),
I might suggest looking at Graphical Designer from Advanced Software
Technologies. I have attached some information about the tool, as
anyone seriously considering using an OO methodology is best advised
to get automated support using it.
Graphical Designer is a tool that is very easy to learn and use but still
gives you the power to generate and/or reverse engineer legacy code (C &
C++ at present), support several OO design methodologies (Rumbaugh,
Booch, and Unified--coming soon), and is user-extensible--create/extend
your own methods, you may want to take a look.
We have a FREE evaluation offer. Via our web site you may access our ftp
site in order to download: http://www.advancedsw.com/infopage
Once downloaded, simply contact me via email and I can get you a softkey.
If you need media, please contact me.
You may contact me with any questions at: Jeff @ 800-811-2784 or my
email is je...@advancedsw.com
Hope this helps!
-Jeff Fulton
=====================================
Advanced Software Technologies Inc.
7800 S. Elati Street, Suite 205
Littleton, Colorado 80120-4456
Ph: (303) 730-7981
Fax:(303) 730-7983
e-mail:je...@advancedsw.com
=====================================
Corneil du Plessis
>Designs in our different instrument designs. We will to the
>most extent use VC++ 4.
>
>Two design methods have been launched: The Booch method and
>the Object Modeling Technique (OMT). Can anybody advise me of
>which (if any) is the best chose, or if any other technique may
>be even better. What are their pro's and con's?
>
>thanks for your effort,
>Helge
At this point in time it is pretty safe to consider either OMT or Booch.
However after having looked at the Unified Method 0.8 draft I would advise
newcomers to learn OMT because I think the impact on the diagramming side is not
going to be as great as to a Booch user, you can not ignore Booch because it has
just as great an influence on UM.
What is important for me as a Booch user is that I have found the Booch method
as a whole very valuable and have learned improve my process.
The diagrams is only a method of communication and thus only part of a greater
method, if THAT method is not implemented properly you will not deliver quality
software.
Regards
Corneil
Corneil du Plessis - cor...@netlunx.netline.co.za
Lexicon Business Systems, P.O. Box 176, Gallo Manor, 2052, R.S.A.
Tel: +27 11 803-8689 Fax: +27 11 803-8768
----------------------------------------------------------------
Confusious he say: He who experiments learns much,
but reboots often.
Corneil he say: Get Windows NT
Jens Coldewey
> We are in the very beginning of introducing Object Oriented
> Designs in our different instrument designs. We will to the
> most extent use VC++ 4.
>
> Two design methods have been launched: The Booch method and
> the Object Modeling Technique (OMT). Can anybody advise me of
> which (if any) is the best chose, or if any other technique may
> be even better. What are their pro's and con's?
> I think Booch is better suited for technical applications while
OMT has a strong emphasis on data modelling. But if you're not
too much dependent on tools, think about using their joint
'Unified Method'. It seems to collect all advantages of either
methods. As far as I know, Jacobson is participating too, so
there may also be a fairly good approach to use cases. But
there are no supporting tools by now. Contact rational.com, they
might have more info.
Jens
--
Jens Coldewey |s |d &|m | software design & management gmbh&CoKG
| | | | Thomas-Dehler-Str. 27
jens.c...@sdm.de | | | | 81737 Munich, Germany.