Re: SDL-News: UML vs SDL


Subject: Re: SDL-News: UML vs SDL
From: Rick Reed TSE (rickreed#tseng.co.uk)
Date: Fri Dec 04 1998 - 11:02:43 GMT


The originator of this message is responsible for its content.
-----From Rick Reed TSE <rickreed#tseng.co.uk> to sdlnews -----

At 13:37 +0000 3/12/98, Maha Boughdadi wrote:
>I received no answer concerning my request below, That's why I'm
>sendingit again.
>Thank you,
>Maha
>----------------
>I would like to know if some comparison work has been done concerning
>UML vs SDL for Real-Time system design (such as CORBA )

Please excuse just a brief reply as I am quite busy.

The view expressed by <Jacqueline.Floch#informatics.sintef.no> is shared by:
1. The ITU group responsible for the language standards (SDL, MSC, TTCN ...);
2. The SDL tool vendors;
3. Several industrial organisations.

Comparison between SDL and UML is not comparing like with like, as UML is a
set of notations and SDL is (essentially) one notation. A better comparison
would be between the UML notations and methodology and the corresponding
ITU notations and methodology.

The current ITU methodology (Z.100 Supplement 1) is to use object model
notation (as in OMT, now included in UML), with MSC for describing "use
seqeunces" and SDL for detailed behaviour description.

Work in progress within the ITU experts' groups for languages includes the
ITU Object Definition Language (Z.130) which is scheduled to be approved in
February 1999, and revisions of MSC and SDL scheduled to be approved
November 1999. One aspect of these revised standards is to support the use
of MSC/SDL together with CORBA. The SDL tool vendors support these changes.

Within this work there is also a specific study of harmonisation between
UML notation and the ITU languages. The main reason for this is because
within the ITU we do not want to invent yet another language for object
models.

Through my work as the SDL rapporteur, I understand that a number of
organisations have indeed made comparative studies of UML and the ITU
approach, and have concluded that the ITU MSC and SDL languages are largely
superior tothe corresponding UML notations: the ITU languages have well
defined semantics, have good tool support and a proven track record.
Unfortunately documentation on such studies seems to be confidential. The
ITU languages are also (of course) "Standardized" (with a captial S), and
therefore not the proprietary property of any one vendor (or "forum").

The result is that all those involved in MSC/SDL do not see the choice as
UML vs SDL, but rather either
a) ITU languages used in combination with some notation from UML;
b) UML in which some parts are instantiated by ITU languages.

The ITU has a further plan to update its methodology to incorporate the
languages approved in Novemeber 1999.

BTW: you should check any statement made by advocates of UML about SDL/MSC
to see if they are fully aware of the current versions of languages and
tools. SDL/MSC are living languages and therefore have developed over the
years (since the time when Jacobsen was actively involved over 12 years
ago, for example). On the other hand the standardisation process ensures
that investment in the languages is protected and that changes are
justifiable (rather than just fashionable).

--
Rick Reed, TSE Limited
13 Weston House, 18-22 Church Street
Lutterworth Leicestershire LE17 4AW United Kingdom
Tel +44 14 55 55 96 55; Fax +44 14 55 55 96 58
Mob +44 79 70 50 96 50
email: rickreed#tseng.co.uk
http://www.tseng.co.uk   ftp://ftp.tseng.co.uk/tseng/

-----End text from Rick Reed TSE <rickreed#tseng.co.uk> to sdlnews ----- For help, email "majordomo#sdl-forum.org" with the body of your email as: help or (iff this does not answer your question) email: owner-sdlnews#sdl-forum.org



This archive was generated by hypermail 2a23 : Sun Jun 16 2013 - 10:41:40 GMT