Subject: SDL-News: SAVE: Request for additional feedback
From: William H. Skelton (W.Skelton#SOLINET.com)
Date: Thu Jan 08 2004 - 13:04:09 GMT
Dear John, Jacqueline, Richard & Laurent (and others interested in the save
construct),
Thank you for your interest in the SDL Task Force; here are some comments
to clarify the background and request more feedback from people using the
save construct or otherwise feel it is important.
Firstly, anyone not yet registered, please register on the task force home
page (www.SDL-Task-Force.org) before sending emails. Otherwise the
mail-server will reject your email and we will not receive notification of
this (although you will). I have asked the server administration to add
John, Jacqueline & Susanne (Richard should already be registered) to
prevent further emails being missed, but others will need to register.
Also I would like to point out that the SDL Task Force is identifying and
defining the 'simplest, useful SDL subset'. At the same time we are
identifying the simplest, useful enhancements to allow SDL to be used for
test requirements. The task force is a self-funding group of volunteers,
recognised by the SDL-Forum; the task force will report its results at the
SAM'04 workshop later this year (http://www.site.uottawa.ca/sam04/).
All of the language elements in the draft specification (Version 1.1) on
the web-site (Version 1.2 coming soon) have been filtered and fully
justified as essential for the implementation or validation of state
machines. The starting point is 'What is a state machine?' - a large
amount of work has been put into this and it has been taken very, very
seriously.
Please understand that it is not important for the justification of
language elements how much experience you have or what your position is;
the only thing that counts is whether the language element is essential for
(or greatly simplifies) the implementation and validation of state machines.
Although I fully respect and acknowledge that many people are established
members of the SDL community, the SDL Task Force cannot consider emails
using the author's credentials as a reason to add save to the SDL subset.
The task force members do not promote themselves or their technical
achievements in the discussions, technical comments have equal weighting
regardless of the authors experience.
So far, no example has been presented that shows why save is
essential. Some examples have come close; for example, Andreas Prinz, Qing
Li, Alkis Yiannakoulias, Rick Reed, Susanne Graf and others have presented
examples addressing specific ways save seems to be useful. These are in
summary:
1) To handle the processing of signals that may arrive in an arbitrary
order -> under discussion if other constructs may be more suitable.
2) The deferred processing of signals not covered by a procedure -> under
discussion from the view point that the way procedures are used may not be
clean.
3) The buffering of application data, while waiting for an event to happen
(i.e. activation of a data link) -> under discussion if SEQUENCE OF may be
a better way to do this.
A problem has been identified that there is no atomic way to flush signals
that have been saved. For example, if a signal E1 is saved, then E2
arrives causes a reset of the state machine, then another E1 arrives, there
is no way to flush at the time of the reset the saved E1 without losing the
new E1.
In conclusion, although there is a strong interest in save, unless there is
a technical justification presented as an essential need for a specific
situation, it will not be part of the subset. In particular it would be
very helpful and much appreciated, if those people with such a broad
experience, such as Laurent Doldi, could identify the key points why save
is needed, and contribute to the other issues at the same time.
Best wishes for the New Year - looking forward to comments...
William
>Reply-To: <sarallo#appairent.com>
>From: "John Sarallo" <sarallo#appairent.com>
>To: "Floch Jacqueline" <Jacqueline.Floch#sintef.no>,
> <members#sdl-task-force.org>,
> "'William H. Skelton'" <W.Skelton#SOLINET.com>
>Cc: <sdlnews#sdl-forum.org>
>Subject: RE: SDL-news: Experience report: Save is widely used - also by
>new SDL users
>Date: Wed, 7 Jan 2004 13:52:01 -0500
>X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
>Importance: Normal
>X-RCPT-TO: <W.Skelton#SOLINET.com>
>
>All,
>
>I realize I'm a little late with this response, but I just wanted to say
>that I agree strongly with the comments made by Jacqueline
>(below), Richard, Susanne, and Laurent.
>
>I am a long time user of SDL, modelling protocols such as UMTS, 802.11,
>and 802.15.3. Automatic code generation was used to create real-time
>embedded software from these models.
>
>I have also formally taught SDL in a classroom setting, so I have first
>hand experience with new users learning and using the Save construct.
>
>Regards,
>
>John Sarallo
>
>-----------------------------------
>The Answer is Appairent!
>-----------------------------------
>John Sarallo
>Senior Protocol Engineer
>Appairent Technologies, Inc.
>150 Lucius Gordon Drive
>Suite 211
>West Henrietta, NY 14586 USA
>Phone: (585) 214-2468
>Mobile: (585) 727-2014
>Fax: (585) 214-2461
><mailto:john.sarallo#appairent.com>mailto:john.sarallo#appairent.com
>-----------------------------------
>-----Original Message-----
>From: owner-sdlnews#sdl-forum.org [mailto:owner-sdlnews#sdl-forum.org]On
>Behalf Of Floch Jacqueline
>Sent: Friday, December 19, 2003 4:17 AM
>To: 'members#sdl-task-force.org'; 'William H. Skelton'
>Cc: 'sdlnews#sdl-forum.org'
>Subject: SDL-news: Experience report: Save is widely used - also by new
>SDL users
>
>Dear All,
>
>As a contribution to the debate "SDL in danger", I would like to report
>some of my experience.
>
>I have been working with and using SDL the last 15 years in various ways:
>
>- I have developed various embedded and telecom systems using SDL.
>- I have developed several SDL runtime systems. Among them some are used
>in embedded systems with limited memory space.
>- I have introduced SDL as as part of a model driven development approach
>in several Norwegain companies.
>- I have developed a flexible code generation approach from SDL to various
>programming languages (C, C++, OCCAM, Plex, Java) and execution platforms.
>
>My long and wide experience with SDL leads to the following observartions:
>- save is easy to undertand and to use.
>- save is essential - both for modelling the postponing of signal
>processing and for modelling
>the coordination of concurrent initiatives.
>- save can be easily supported by run-time systems.
>- save is widely used by SDL users in Norway (note they are many SDL users
>in Norway).
>
>In my opinion, the removal of save will not simplify the language, but
>rather add complexity to modelling.
>
>Regards,
>Jacqueline Floch
>
>
>Jacqueline Floch
>Dr.ing. - Research scientist
>______________________________
>SINTEF Telecom and Informatics
>NO-7465 Trondheim, Norway
>Phone: +47 73 59 30 12
>Telefax: +47 73 59 29 77
>
------------------------------------------------------------------------
William H. Skelton, Engineering Dept.
SOLINET GmbH Solutions for Innovative Networks
Mittlerer Pfad 26, 70499 Stuttgart, Germany
Tel +49 711 1398 1377, Fax +49 711 866 1240
Mobile +49 171 247 6688
W.Skelton#SOLINET.com, www.SOLINET.com
Dear John, Jacqueline, Richard & Laurent (and others interested in
the save construct),
Thank you for your interest in the SDL Task Force; here are some comments
to clarify the background and request more feedback from people using the
save construct or otherwise feel it is important.
Firstly, anyone not yet registered, please register on the task force
home page
(www.SDL-Task-Force.org)
before sending emails. Otherwise the mail-server will reject your
email and we will not receive notification of this (although you
will). I have asked the server administration to add John,
Jacqueline & Susanne (Richard should already be registered) to
prevent further emails being missed, but others will need to
register.
Also I would like to point out that the SDL Task Force is identifying and
defining the 'simplest, useful SDL subset'. At the same time we are
identifying the simplest, useful enhancements to allow SDL to be used for
test requirements. The task force is a self-funding group of
volunteers, recognised by the SDL-Forum; the task force will report its
results at the SAM'04 workshop later this year
(http://www.site.uottawa.ca/sam04/).
All of the language elements in the draft specification (Version 1.1) on
the web-site (Version 1.2 coming soon) have been filtered and fully
justified as essential for the implementation or validation of state
machines. The starting point is 'What is a state machine?' -
a large amount of work has been put into this and it has been taken very,
very seriously.
Please understand that it is not important for the justification of
language elements how much experience you have or what your position is;
the only thing that counts is whether the language element is essential
for (or greatly simplifies) the implementation and validation of state
machines.
Although I fully respect and acknowledge that many people are established
members of the SDL community, the SDL Task Force cannot consider emails
using the author's credentials as a reason to add save to the SDL
subset.
The task force members do not promote themselves or their technical
achievements in the discussions, technical comments have equal weighting
regardless of the authors experience.
So far, no example has been presented that shows why save is
essential. Some examples have come close; for example, Andreas
Prinz, Qing Li, Alkis Yiannakoulias, Rick Reed, Susanne Graf and others
have presented examples addressing specific ways save seems to be
useful. These are in summary:
1) To handle the processing of signals that may arrive in an arbitrary
order -> under discussion if other constructs may be more
suitable.
2) The deferred processing of signals not covered by a procedure ->
under discussion from the view point that the way procedures are used may
not be clean.
3) The buffering of application data, while waiting for an event to
happen (i.e. activation of a data link) -> under discussion if
SEQUENCE OF may be a better way to do this.
A problem has been identified that there is no atomic way to flush
signals that have been saved. For example, if a signal E1 is saved,
then E2 arrives causes a reset of the state machine, then another E1
arrives, there is no way to flush at the time of the reset the saved E1
without losing the new E1.
In conclusion, although there is a strong interest in save, unless there
is a technical justification presented as an essential need for a
specific situation, it will not be part of the subset. In
particular it would be very helpful and much appreciated, if those people
with such a broad experience, such as Laurent Doldi, could identify the
key points why save is needed, and contribute to the other issues at the
same time.
Best wishes for the New Year - looking forward to comments...
William
Reply-To: <sarallo#appairent.com>
From: "John Sarallo" <sarallo#appairent.com>
To: "Floch Jacqueline" <Jacqueline.Floch#sintef.no>,
<members#sdl-task-force.org>,
"'William H. Skelton'" <W.Skelton#SOLINET.com>
Cc: <sdlnews#sdl-forum.org>
Subject: RE: SDL-News: Experience report: Save is widely used - also by new SDL users
Date: Wed, 7 Jan 2004 13:52:01 -0500
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
Importance: Normal
X-RCPT-TO: <W.Skelton#SOLINET.com>
All,
I realize I'm a little late with this response, but I just wanted to say that I agree strongly with the comments made by Jacqueline (below), Richard, Susanne, and Laurent.
I am a long time user of SDL, modelling protocols such as UMTS, 802.11, and 802.15.3. Automatic code generation was used to create real-time embedded software from these models.
I have also formally taught SDL in a classroom setting, so I have first hand experience with new users learning and using the Save construct.
Regards,
John Sarallo
-----------------------------------
The Answer is Appairent!
-----------------------------------
John Sarallo
Senior Protocol Engineer
Appairent Technologies, Inc.
150 Lucius Gordon Drive
Suite 211
West Henrietta, NY 14586 USA
Phone: (585) 214-2468
Mobile: (585) 727-2014
Fax: (585) 214-2461
mailto:john.sarallo#appairent.com
-----------------------------------
- -----Original Message-----
- From: owner-sdlnews#sdl-forum.org [mailto:owner-sdlnews#sdl-forum.org]On Behalf Of Floch Jacqueline
- Sent: Friday, December 19, 2003 4:17 AM
- To: 'members#sdl-task-force.org'; 'William H. Skelton'
- Cc: 'sdlnews#sdl-forum.org'
- Subject: SDL-News: Experience report: Save is widely used - also by new SDL users
- Dear All,
- As a contribution to the debate "SDL in danger", I would like to report some of my experience.
- I have been working with and using SDL the last 15 years in various ways:
- - I have developed various embedded and telecom systems using SDL.
- - I have developed several SDL runtime systems. Among them some are used in embedded systems with limited memory space.
- - I have introduced SDL as as part of a model driven development approach in several Norwegain companies.
- - I have developed a flexible code generation approach from SDL to various programming languages (C, C++, OCCAM, Plex, Java) and execution platforms.
- My long and wide experience with SDL leads to the following observartions:
- - save is easy to undertand and to use.
- - save is essential - both for modelling the postponing of signal processing and for modelling
- the coordination of concurrent initiatives.
- - save can be easily supported by run-time systems.
- - save is widely used by SDL users in Norway (note they are many SDL users in Norway).
- In my opinion, the removal of save will not simplify the language, but rather add complexity to modelling.
- Regards,
- Jacqueline Floch
- Jacqueline Floch
- Dr.ing. - Research scientist
- ______________________________
- SINTEF Telecom and Informatics
- NO-7465 Trondheim, Norway
- Phone: +47 73 59 30 12
- Telefax: +47 73 59 29 77
This archive was generated by hypermail 2a23 : Thu May 09 2013 - 16:05:50 GMT