Re: UML & Ada

Re: UML & Ada

[Anonymous]

On Tue, 16 Jun 1998 08:29:14 -0700, Dennis Miller
<Dennis_Miller-P28917@email.mot.com> wrote:

> Although Ada is considered an object-based language (not including Ada
> 95) as opposed to a OO language, is it possible to construct code in Ada
> from UML notation and vice versa?  Better yet, is it possible to get a
> high level of functionality from an OO notation system and implementing
> it with a object-based language?
> 
> Dennis Miller
> 

If you define "object oriented" as "implemented using inheritance and
dispatching", then Ada 83 was not an OO language and Ada is. This is a
common "definition" of OO; however, it makes no mention of objects, so
it is not a definition of OO.

If, instead, you define OO as "structured around problem-domain objects;
directly representing these objects in software; encapsulating data
structures with their operations; providing a visible interface and
hiding both the definition of the data structures and the implementation
of the operations", then Ada has always been an OO language.

You can, and have always been able to, implement a high-quality OO
design directly in Ada; it follows that you can represent such an
implementation in a notation that describes such a design.

In Ada 83, you must use the technique known as "composition": You define
your most concrete types first, then create composites of such types to
create more abstract types. If you require dispatching operations, you
must explicitly create such operations. This requires writing a little
more code than if inheritance and dispatching features of a language are
used, but has the benefit that the resulting code is much more easily
read and understood than code that uses inheritance and dispatching,
making it easier to modify.

Ada now has inheritance and dispatching features, but the benefits of
composition argue against using them.

Jeff Carter  PGP:1024/440FBE21
My real e-mail address: ( carter @ innocon . com )
"We call your door-opening request a silly thing."
Monty Python & the Holy Grail

Posted with Spam Hater - see
http://www.compulink.co.uk/~net-services/spam/

Re: UML & Ada

[Michael Stark]

Dennis Miller wrote:
> 
> Although Ada is considered an object-based language (not including Ada
> 95) as opposed to a OO language, is it possible to construct code in Ada

But Ada 95 _is_ Ada, just as FORTRAN 77 superceded previous versions
of FORTRAN to become _the_ standard.  Given this, I see no reason that
you can't use UML with Ada, unless you are dealing with Ada 83 legacy
code, in which case the answer is a resounding "it depends" ;)


> from UML notation and vice versa?  Better yet, is it possible to get a
> high level of functionality from an OO notation system and implementing
> it with a object-based language?
> 
> Dennis Miller

Mike
-- 
Michael Stark
Goddard Research & Study Fellow
University of Maryland, College Park
e-mail: mstark@cs.umd.edu
phone: (301) 405-2721
"Soccer is life -- the rest is just details"  (GO USA!!!)

Re: UML & Ada

[Richard D Riehle]

In article <35880034.68F8@cs.umd.edu>,
	Michael Stark <mstark@cs.umd.edu> wrote:

>Dennis Miller wrote:

-- snip snip snip
>
>But Ada 95 _is_ Ada, just as FORTRAN 77 superceded previous versions
>of FORTRAN to become _the_ standard.  Given this, I see no reason that
>you can't use UML with Ada, unless you are dealing with Ada 83 legacy
>code, in which case the answer is a resounding "it depends" ;)

-- snip snip snip

It would be nice if UML and Ada were a comfortable, easy fit.  This is 
not the case.  UML assumes that a class is a module and vice-versa much
as one would expect of Eiffel or C++.  In Ada, a module is a package. There
is not straightforward support in UML for child packages.  UML is at
the wrong level of granuality to support Ada's model for encapsulation 
and separation of specification from implementation.  There seem to be
a lot of disconnects between UML and Ada. 

What we need, for Ada, is an Ada-specific notation for object modeling, 
one which takes advantage of the power of Ada rather than the limitations
of C++.  I recall that Ed Colbert was working on a modeling notation that
would map to Ada.  I wonder if he is still pursuing this.  

Richard Riehle  
AdaWorks Software Engineering

Re: UML & Ada

[John McCabe]

Richard D Riehle <laoXhai@ix.netcom.com> wrote:

>What we need, for Ada, is an Ada-specific notation for object modeling, 
>one which takes advantage of the power of Ada rather than the
>limitations of C++.  I recall that Ed Colbert was working on a modeling
>notation that would map to Ada.  I wonder if he is still pursuing this. 

As a matter of interest, have you checked out HOOD (Version 4)? I have 
been led to believe that the mapping to Ada is quite good, but whether 
HOOD has become more Object-Oriented rather than Object Based in its 
transition from 3.1 to 4 I don't know.

Also, TNI do a tool called STOOD which is supposed to have code 
generation targeted to GNAT (or something like that).

-- 
Best Regards
John McCabe

=====================================================================
Not necessarily my company or service providers opinions.
=====================================================================

Re: UML & Ada

[Jean-Pierre Rosen]

There is a book about HOOD4 (which, BTW, does support inheritance).
You can find info about it at http://perso.wanadoo.fr/adalog/hoodbook.htm

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

John McCabe a �crit dans le message <6malik$42b@gcsin3.geccs.gecm.com>...
>Richard D Riehle <laoXhai@ix.netcom.com> wrote:
>
>>What we need, for Ada, is an Ada-specific notation for object modeling,
>>one which takes advantage of the power of Ada rather than the
>>limitations of C++.  I recall that Ed Colbert was working on a modeling
>>notation that would map to Ada.  I wonder if he is still pursuing this.
>
>As a matter of interest, have you checked out HOOD (Version 4)? I have
>been led to believe that the mapping to Ada is quite good, but whether
>HOOD has become more Object-Oriented rather than Object Based in its
>transition from 3.1 to 4 I don't know.
>
>Also, TNI do a tool called STOOD which is supposed to have code
>generation targeted to GNAT (or something like that).
>
>--
>Best Regards
>John McCabe
>
>=====================================================================
>Not necessarily my company or service providers opinions.
>=====================================================================
>
>

Re: UML & Ada

[Pierre DISSAUX]

John McCabe wrote:
> 
> Richard D Riehle <laoXhai@ix.netcom.com> wrote:
> 
> >What we need, for Ada, is an Ada-specific notation for object modeling,
> >one which takes advantage of the power of Ada rather than the
> >limitations of C++.  I recall that Ed Colbert was working on a modeling
> >notation that would map to Ada.  I wonder if he is still pursuing this.
> 
> As a matter of interest, have you checked out HOOD (Version 4)? I have
> been led to believe that the mapping to Ada is quite good, but whether
> HOOD has become more Object-Oriented rather than Object Based in its
> transition from 3.1 to 4 I don't know.

It was one of the aims of this release of the method.

There is now an explicit concept of HOOD4 Class, which describes an
Abstract module (Type + primitive ops). Its behaviour may be described
by a State Transition Diagram, and it may be annotated graphically to
map Ada constructs:
- A passive HOOD4 class produces a package with a tagged type.
- An active HOOD4 class produces a package with a task type.
- A passive HOOD4 class with Read-Only or Read-Write labels on its
operations produces a package with a protected type.

There is a similar mapping for modules containing data (Objects):
- A passive HOOD4 object produces a package.
- An active HOOD4 object produces a package with a task.
- A passive HOOD4 object with Read-Only or Read-Write labels on its
operations produces a package with a protected object.

You can mix Classes and Objects, in order to explicitly specify where
you decided to use static instances (HOOD objects), and where you let
the ability to use dynamic instances (HOOD classes).

Generic classes and objects are also supported at high level.

There is a support of inheritance and polymorphism ("abstract" and
"inherited" operations).

....

Graphical notations are very close to those of UML.

> 
> Also, TNI do a tool called STOOD which is supposed to have code
> generation targeted to GNAT (or something like that).

You are right, but we also target ObjectAda from Aonix (It was recently
demonstrated at DASIA'98 in Athens).

Pierre DISSAUX
TNI
http://www.tni.fr/

Re: UML & Ada

[Pat Rogers]

Pierre DISSAUX wrote in message <358A24F7.2D85@tni.fr>...

<snip>

>> Also, TNI do a tool called STOOD which is supposed to have code
>> generation targeted to GNAT (or something like that).
>
>You are right, but we also target ObjectAda from Aonix (It was recently
>demonstrated at DASIA'98 in Athens).
>
>Pierre DISSAUX
>TNI
>http://www.tni.fr/


I went to the web site but could not find price info for STOOD.

Re: UML & Ada

[Stephen Leake]

Dennis Miller <Dennis_Miller-P28917@email.mot.com> writes:

> Although Ada is considered an object-based language (not including Ada
> 95) as opposed to a OO language, is it possible to construct code in Ada
> from UML notation and vice versa?  Better yet, is it possible to get a
> high level of functionality from an OO notation system and implementing
> it with a object-based language?

Yes; Ada 83 is equivalent to a Turing machine, so it can be done. But
I suspect this may not be what you meant by "functionality".

On a more practical note, why are you excluding Ada 95? It is
certainly easier to get an Ada 95 compiler than an Ada 83 compiler for
most machines! (see http://www.gnat.com/ and http://www.aonix.com/).

> 
> Dennis Miller

-- Stephe

Re: UML & Ada

[Dennis Miller]

Stephen Leake wrote:

> Dennis Miller <Dennis_Miller-P28917@email.mot.com> writes:
>
> > Although Ada is considered an object-based language (not including Ada
> > 95) as opposed to a OO language, is it possible to construct code in Ada
> > from UML notation and vice versa?  Better yet, is it possible to get a
> > high level of functionality from an OO notation system and implementing
> > it with a object-based language?
>
> Yes; Ada 83 is equivalent to a Turing machine, so it can be done. But
> I suspect this may not be what you meant by "functionality".
>
> On a more practical note, why are you excluding Ada 95? It is
> certainly easier to get an Ada 95 compiler than an Ada 83 compiler for
> most machines! (see http://www.gnat.com/ and http://www.aonix.com/).
>
> >
> > Dennis Miller
>
> -- Stephe

I excluded Ada 95 on the principle that it is more OO than its predecessor
(Ada 83).

Dennis

Re: UML & Ada

[Pierre DISSAUX]

Dennis Miller wrote:
> 
> Although Ada is considered an object-based language (not including Ada
> 95) as opposed to a OO language, is it possible to construct code in Ada
> from UML notation and vice versa?  Better yet, is it possible to get a
> high level of functionality from an OO notation system and implementing
> it with a object-based language?
> 
> Dennis Miller

HOOD (Hierarchical Object Oriented Design) method provides high level
constructs that can be easily translated into Ada code.
Last version of the method (HOOD4) is closer to Ada95 and C++, but
remains fully compatible with Ada83 and C.

For further details: http://www.hood.be/


-----------------------
Pierre Dissaux
TNI

Re: UML & Ada

[John McCabe]

Sorry to digress a little, but on this subject, has anyone here used UML 
significatnly with Ada and/or can they recommend a good book which 
encompasses this subject.

Alternatively just a good general UML book.

TIA

-- 
Best Regards
John McCabe

=====================================================================
Not necessarily my company or service providers opinions.
=====================================================================

Re: UML & Ada

[Alan Gauld]

> Alternatively just a good general UML book.

I'm finding the Wrox press "Instant UML" quite useful.
UML Distilled gets good rviews too.

On the original thread - look at Booch's first edition of his OOD book.
The last case study is an OO system in ADA.

Also the OMT book by Rumbaugh gives general guidance in implementing
OOD in non OOP languages. (I think from memory, that the Coad/Yourdon
OOD book does also)

Although the notations are different they all translate directly and easily
to UML.

Alan G.

UML & Ada

[Marin David Condic]

The last time I went informally cruising through anything to do with the
UML, I seem to recall that it lacked any means of representing Ada tasks.
Through interaction diagrams, you kind of had a way of saying how one thread
of control would interact with another, but when it came to representing
something that should be translated to a task, complete with rendesvous,
etc, there didn't seem to be any way of expressing this.

I hear rumors from Dr. Dobbs that there is some sort of revision going on to
come up with a v2.0 of UML. Has anybody heard if there is any intention to
support tasks with some kind of diagram symbol? UML seems to be bent on the
assumption that all software will be built via OOD & I've had a hard time
ibn the past trying to make it fit something that was not OOD, so I'm
wondering if they are at all considering making it more general - or at
least making it possible to describe Ada programs more thoroughly.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/

Re: UML & Ada

[Jeffrey Carter]

Marin David Condic wrote:
> 
> The last time I went informally cruising through anything to do with the
> UML, I seem to recall that it lacked any means of representing Ada tasks.
> Through interaction diagrams, you kind of had a way of saying how one thread
> of control would interact with another, but when it came to representing
> something that should be translated to a task, complete with rendesvous,
> etc, there didn't seem to be any way of expressing this.

I'm not very fond of UML, as it seems you need a large number of
diagrams, each of which contributes a couple of bits to the total
description. I much prefer something like CAIM, in which a few diagrams
each contribute a great deal.

It also seems as if UML works best for business applications in C++, so
of course there would be no concept of tasking.

That said, however, I think a modeling notation should not represent
language-specific implementation features. The initial default
assumption should be that all objects are active, and that can then be
modified as you work towards an implementation.

-- 
Jeff Carter
"Your mother was a hamster and your father smelt of elderberries."
Monty Python & the Holy Grail

Re: UML & Ada

[Torben Hoffmann]

"Jeffrey Carter" <jrcarter@acm.org> wrote in message
news:3BF9EB02.D9946B9B@acm.org...
> Marin David Condic wrote:
> >
> > The last time I went informally cruising through anything to do with the
> > UML, I seem to recall that it lacked any means of representing Ada
tasks.
> > Through interaction diagrams, you kind of had a way of saying how one
thread
> > of control would interact with another, but when it came to representing
> > something that should be translated to a task, complete with rendesvous,
> > etc, there didn't seem to be any way of expressing this.
>
> I'm not very fond of UML, as it seems you need a large number of
> diagrams, each of which contributes a couple of bits to the total
> description. I much prefer something like CAIM, in which a few diagrams
> each contribute a great deal.
>
Where can I find some information on CAIM?

Cheers,
/Torben

Re: UML & Ada

[Jeffrey Carter]

Torben Hoffmann wrote:
> 
> Where can I find some information on CAIM?

and MDC had a similar post.

CAIM is the successor to MMAIM. After MM lost interest in MMAIM
(associated with not winning the project that MMAIM was associated
with), and especially after MM lost interest in me, I changed the MM to
C. CAIM has evolved some since those days.

MMAIM was documented in

"MMAIM: A Software Development Method for Ada", which appeared in 2
parts in Ada Letters, 1988 May/Jun and Sep/Oct

and in

"Reducing Software Development Costs with Ada", Proceedings of the
Seventh Annual National Conference on Ada Technology, ANCOST, Inc., 1989

and is referenced in other papers on Ada by J. Carter, whoever he is. I
doubt if you can find any of these on line. I only have most of them in
paper form.

CAIM seems to have some similarity to what I've been able to find out
about HOOD. I wish I could find more information about HOOD in a format
that I can read so I can tell if I really have anything to contribute.

CAIM uses a hierarchical approach to development, with a design at one
level serving as a specification at the next, and mechanical generation
of Ada specifications (and "empty bodies") at all levels and of
implementation code from the lowest levels. CAIM specifications are
fairly formal while still being fairly easy to understand and
non-mathematical.

-- 
Jeff Carter
"Sons of a silly person."
Monty Python & the Holy Grail

Re: UML & Ada

[Jean-Pierre Rosen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 616 bytes --]


"Jeffrey Carter" <jrcarter@acm.org> a �crit dans le message news: 3BFB163F.F0ECE879@acm.org...
> CAIM seems to have some similarity to what I've been able to find out
> about HOOD. I wish I could find more information about HOOD in a format
> that I can read so I can tell if I really have anything to contribute.
>
Did you have a look at www.hood-method.org ?
If you don't get enough information from there, you are welcome to get in touch directly with me.

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

Re: UML & Ada

[Preben Randhol]

On Tue, 20 Nov 2001 05:32:59 GMT, Jeffrey Carter wrote:
> I'm not very fond of UML, as it seems you need a large number of
> diagrams, each of which contributes a couple of bits to the total
> description. I much prefer something like CAIM, in which a few diagrams
> each contribute a great deal.

I have been looking a bit on OMT as it is used in an OOP text book I
bought. Does anybody know how good OMT is? I also found a page for
mapping from OMT associations to Ada:

http://www.clark.net/pub/dsmith/OO/overview.html

Thanks in advance

Preben
-- 
 ()   Join the worldwide campaign to protect fundamental human rights.
'||}
{||'                                           http://www.amnesty.org/

Re: UML & Ada

[Jean-Marc Bourguet]

Preben Randhol <randhol+abuse@pvv.org> writes:

> I have been looking a bit on OMT as it is used in an OOP text book I
> bought. Does anybody know how good OMT is? 

If I do remember correctly, OMT is a predecessor of UML.  If I'm
right, looking at OMT know is not something advisable.

-- 
Jean-Marc

Re: UML & Ada

[Martin Dowie]

"Preben Randhol" <randhol+abuse@pvv.org> wrote in message
news:slrn9vk9f4.3qs.randhol+abuse@kiuk0156.chembio.ntnu.no...
> On Tue, 20 Nov 2001 05:32:59 GMT, Jeffrey Carter wrote:
> > I'm not very fond of UML, as it seems you need a large number of
> > diagrams, each of which contributes a couple of bits to the total
> > description. I much prefer something like CAIM, in which a few diagrams
> > each contribute a great deal.
>
> I have been looking a bit on OMT as it is used in an OOP text book I
> bought. Does anybody know how good OMT is? I also found a page for
> mapping from OMT associations to Ada:
>
> http://www.clark.net/pub/dsmith/OO/overview.html

Well, Rumbaugh published his book in 1991 and has since joined up
with Rational together with Jacobson and Booch, publishing work
together in 1998/1999, so no matter how good it is, I suspect that
there will be little market left for it now...

I suspect that a lot of the OMT concepts are actually in the UML
anyway and so the mapping may be similar.

Re: UML & Ada

[Jeffrey Carter]

Preben Randhol wrote:
> 
> I have been looking a bit on OMT as it is used in an OOP text book I
> bought. Does anybody know how good OMT is? I also found a page for
> mapping from OMT associations to Ada:

OMT is one of the three OOA techniques that merged to form UML.

-- 
Jeff Carter
"Sons of a silly person."
Monty Python & the Holy Grail

Re: UML & Ada

[Adrian Hoe]

Jeffrey Carter <jrcarter@acm.org> wrote in message news:<3BFB1023.8E7151F3@acm.org>...
> Preben Randhol wrote:
> > 
> > I have been looking a bit on OMT as it is used in an OOP text book I
> > bought. Does anybody know how good OMT is? I also found a page for
> > mapping from OMT associations to Ada:
> 
> OMT is one of the three OOA techniques that merged to form UML.


UML := OMT + Booch + Jacobson;

I have been using OMT for some time and I just began to learn and use
UML. As personal taste, I prefer OMT to UML as someone posted in this
thread "UML takes a big diagram to represent design in bits....".

I still have not fully discovered why UML is a better OO methodology.
Anyone?

                                                    -- Adrian Hoe

Re: UML & Ada

[Martin Dowie]

> I still have not fully discovered why UML is a better OO methodology.
> Anyone?

The UML is not a method (methodology is the study of methods...) but
the Unified Software Development Process (from Jacobson, Booch,
Rumbaugh) is:

http://www.amazon.com/exec/obidos/ASIN/0201571692/qid=941652026/sr=1-8/103-9
272609-3732618

Re: UML & Ada

[Marin David Condic]

Thanks. I'll try to take a peek at CAIM - do you have any links?

I'd like to see a modeling language that did a couple of things: One would
be to have diagram elements that represent Ada constructs (packages, tasks,
subprograms, etc.) with varying levels of "visibility" (e.g. "show me this
package and its subprograms, but not its types and objects...") to which you
can attach matching code (either partially generated or all manual) so you
can punch a button and get the code from the diagrams. This would be useful
for describing the program starting at a fairly high level and iterating
down to a final product. The important thing being that you have a method of
representing anything that can be built in Ada. (Booch diagrams from several
years ago did a fairly nice job, but the last tool I used that did this was
Adagen - and while I found it to fit the niche reasonably well, they had
problems with really buggy software & poor support. I believe they went
casters-up - or maybe got bought up, but I've not heard anything about the
tool in years.)

The other thing would be to have some ability to describe a program at a
higher level of abstraction - I've found interaction diagrams to be useful
and I can see having something that supports object oriented design (that
breaks down into actual Ada components) or other methodologies. At a lower
level, I still find flowcharts to be occasionally useful - especially if you
can generate code from it.

I've found UML to be less than intuitively obvious to the casual observer.
Its been a long time since I've been out searching for a methodology tool,
so maybe there is something out there a little more to my liking. I may have
to once again dive in and look more seriously.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Jeffrey Carter" <jrcarter@acm.org> wrote in message
news:3BF9EB02.D9946B9B@acm.org...
> Marin David Condic wrote:
>
> I'm not very fond of UML, as it seems you need a large number of
> diagrams, each of which contributes a couple of bits to the total
> description. I much prefer something like CAIM, in which a few diagrams
> each contribute a great deal.
>
> It also seems as if UML works best for business applications in C++, so
> of course there would be no concept of tasking.
>
> That said, however, I think a modeling notation should not represent
> language-specific implementation features. The initial default
> assumption should be that all objects are active, and that can then be
> modified as you work towards an implementation.
>

Re: UML & Ada

[martin.m.dowie]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
message news:9tc0qk$cee$1@nh.pace.co.uk...
> The last time I went informally cruising through anything to do with the
> UML, I seem to recall that it lacked any means of representing Ada tasks.
> Through interaction diagrams, you kind of had a way of saying how one
thread
> of control would interact with another, but when it came to representing
> something that should be translated to a task, complete with rendesvous,
> etc, there didn't seem to be any way of expressing this.
>
> I hear rumors from Dr. Dobbs that there is some sort of revision going on
to
> come up with a v2.0 of UML. Has anybody heard if there is any intention to
> support tasks with some kind of diagram symbol? UML seems to be bent on
the
> assumption that all software will be built via OOD & I've had a hard time
> ibn the past trying to make it fit something that was not OOD, so I'm
> wondering if they are at all considering making it more general - or at
> least making it possible to describe Ada programs more thoroughly.

Artisan's Real-time Studio Pro certainly contain extensions to the UML which
they have been working to include in the new standard.

Pragsoft's UML Studio contains tasks as 'components'. I'm not sure either
use the real-time models in their code generation though ('Big Up' to Pat
Rogers for the UML=>Ada script in this tool! Just have to get a reverse
engineering tool to go with it now :-)

Re: UML & Ada

[Mark Fisher]

In sequence diagrams messages are shown as synchronous/asynchronous
using a half-stick arrowhead, which can be supplemented with extensions such
as balking, time-out, guarded etc.

The focus of control is also present to indicate the active nature of the
object(s)
so it is quite simple to represent tasks/concurrency.

Mark


"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
message news:9tc0qk$cee$1@nh.pace.co.uk...
> The last time I went informally cruising through anything to do with the
> UML, I seem to recall that it lacked any means of representing Ada tasks.
> Through interaction diagrams, you kind of had a way of saying how one
thread
> of control would interact with another, but when it came to representing
> something that should be translated to a task, complete with rendesvous,
> etc, there didn't seem to be any way of expressing this.
>
> I hear rumors from Dr. Dobbs that there is some sort of revision going on
to
> come up with a v2.0 of UML. Has anybody heard if there is any intention to
> support tasks with some kind of diagram symbol? UML seems to be bent on
the
> assumption that all software will be built via OOD & I've had a hard time
> ibn the past trying to make it fit something that was not OOD, so I'm
> wondering if they are at all considering making it more general - or at
> least making it possible to describe Ada programs more thoroughly.
>
> MDC
> --
> Marin David Condic
> Senior Software Engineer
> Pace Micro Technology Americas    www.pacemicro.com
> Enabling the digital revolution
> e-Mail:    marin.condic@pacemicro.com
> Web:      http://www.mcondic.com/
>
>
>

Re: UML & Ada

[Ted Dennison]

In article <9tc0qk$cee$1@nh.pace.co.uk>, Marin David Condic says...
>
>The last time I went informally cruising through anything to do with the
>UML, I seem to recall that it lacked any means of representing Ada tasks.
>Through interaction diagrams, you kind of had a way of saying how one thread
>of control would interact with another, but when it came to representing
>something that should be translated to a task, complete with rendesvous,
>etc, there didn't seem to be any way of expressing this.

Not directly, but there are ways. Before you give up on UML, check out
"Designing Concurrent, Distributed, and Real-Time Applications with UML" (
http://www.aw.com/product/0,2627,0201657937,00.html ). That helped us a bit.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.

Re: UML & Ada

[Marin David Condic]

Thanks. I'll see about getting a copy.

I'd still like a diagraming model that allows representation of anything
that can be built in Ada - with icons for all the Ada constructs. What I
know of UML is that it tends to dictate OOD - which makes it kind of lousy
if the system in question doesn't want to use OOD or is already built &
you're trying to reverse-engineer it. (I know this is A Bad Thing, but
sometimes, that's the mission...)

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Ted Dennison" <dennison@telepath.com> wrote in message
news:BrzK7.32049$xS6.54847@www.newsranger.com...
> In article <9tc0qk$cee$1@nh.pace.co.uk>, Marin David Condic says...
> >
> >The last time I went informally cruising through anything to do with the
> >UML, I seem to recall that it lacked any means of representing Ada tasks.
> >Through interaction diagrams, you kind of had a way of saying how one
thread
> >of control would interact with another, but when it came to representing
> >something that should be translated to a task, complete with rendesvous,
> >etc, there didn't seem to be any way of expressing this.
>
> Not directly, but there are ways. Before you give up on UML, check out
> "Designing Concurrent, Distributed, and Real-Time Applications with UML" (
> http://www.aw.com/product/0,2627,0201657937,00.html ). That helped us a
bit.
>
> ---
> T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html
>
> No trees were killed in the sending of this message.
> However a large number of electrons were terribly inconvenienced.

Re: UML & Ada

[Simon Wright]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> writes:

> I'd still like a diagraming model that allows representation of
> anything that can be built in Ada - with icons for all the Ada
> constructs. What I know of UML is that it tends to dictate OOD -
> which makes it kind of lousy if the system in question doesn't want
> to use OOD or is already built & you're trying to reverse-engineer
> it. (I know this is A Bad Thing, but sometimes, that's the
> mission...)

My view is that both UML and Ada do a lot more than you need, in
different ways.

UML has an enormous amount of notation (and 2.0 will have more), and
even where the semantics are precise I bet you normal engineers won't
use it consistently. A presenter at an Ada UK Technology Day (early
'01, I think) told us how neither the system designers creating
statecharts nor the programmers implementing them actually understood
the notation, with predictably unfortunate results.

If you are modelling a problem domain, you likely don't want to be
thinking about tasks. Indeed, the ideal is that your model should be
true (useful) in a number of implementations. Analyse Navigation, say,
and reuse in a cruise missile and a torpedo (I don't think that's
_very_ likely ..). Change your implementation language from C++ to Ada
but retain your model.

The point at which you need to be thinking about tasks (or threads) is
when you're deciding how to map the model onto code. You might want to
use UML, but what you're talking about now is the translation rules
which are the same for all the analysis domains of your application
(lower layers are going to have more handcrafting, of course, perhaps
with indicative-only UML models with lots of annotation).

Stuff on this at
http://www.projtech.com/
http://www.kc.com/html/products/products.html
http://www.kc.com/html/xuml.html
http://www.kc.com/as_site/
(these are quite old links, I haven't checked them recently).

The current translation rules for tasks in my ColdFrame
(http://www.pushface.org/coldframe/) are:

(1) if you mark the class "active" you get a task per instance

(2) the task is constrained by an access to its instance

(3) non-<<class>> procedure operations map to instance procedures
    which call the corresponding task entry.

Re: UML & Ada

[Marin David Condic]

Thanks for the links.

I agree that there is food value in being able to describe an overall design
in a language independent way, but I think I want something at more levels
than that.

I want a tool that lets me describe what a system has to be able to do -
that might be something UML might support in some manner (but my impression
is that it still sticks you pretty much with OOD - which rules out a lot of
other possible designs - especially ones that might already exist.) I want a
tool that describes the overall structure of the software I've built or
intend to build - a sort of overview of how the various Ada pieces hang
together. Then I might want something that allows me to describe lower-level
algorithms within subprograms/tasks.

UML's interaction diagrams - and possibly Use Cases - I find can be helpful,
but I don't see it supporting all the things I want in a design tool.
Perhaps I need to get a more thorough understanding of it to see if maybe I
can shoehorn the things I want to do into the modeling language. But it
would *still* be nice to have something that diagramatically represented all
possible Ada programs - and UML can't be made to do this any way I know of.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Simon Wright" <simon@pushface.org> wrote in message
news:x7v1yisk4d1.fsf@smaug.pushface.org...
>
> If you are modelling a problem domain, you likely don't want to be
> thinking about tasks. Indeed, the ideal is that your model should be
> true (useful) in a number of implementations. Analyse Navigation, say,
> and reuse in a cruise missile and a torpedo (I don't think that's
> _very_ likely ..). Change your implementation language from C++ to Ada
> but retain your model.
>

Re: UML & Ada

[Mark]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in message news:<9tgeqm$n1m$1@nh.pace.co.uk>...
> Thanks for the links.
> 
> I agree that there is food value in being able to describe an overall design
> in a language independent way, but I think I want something at more levels
> than that.

 May I suggest a book that describes some existing systems (with
hardware not just web junk) using the UML. I don't always agree with
their terminology but it is good book none the less, much better than
Unified Software Development book.

Applied Software Architecture 
By Christine Hoffmeister,Robert Nord,Dilip Soni

http://www1.fatbrain.com/asp/bookinfo/bookinfo.asp?theisbn=0201325713&vm=

The key point is that a single "view" of a system can not fully
describe it. You need at least a logical view (modularity) and an
execution/process/task view.

Re: UML & Ada

[Ted Dennison]

In article <c5b88987.0111220631.39445b76@posting.google.com>, Mark says...
>The key point is that a single "view" of a system can not fully
>describe it. You need at least a logical view (modularity) and an
>execution/process/task view.

That sound a bit like the old data flow vs. control flow dichotomy.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.

Re: UML & Ada

[Marc A. Criley]

Ted Dennison wrote:
> 
> In article <c5b88987.0111220631.39445b76@posting.google.com>, Mark says...
> >The key point is that a single "view" of a system can not fully
> >describe it. You need at least a logical view (modularity) and an
> >execution/process/task view.
> 
> That sound a bit like the old data flow vs. control flow dichotomy.

Though few are familiar with it, there is a unified control/data flow
notation that was developed in the late 80s that was amazingly
expressive and easy to use.  It was invented by William Bennett of Link
Flight Simulation and christened "Bennett notation".  Anyone could be
taught it in five minutes, and we found that it scaled extremely well,
and aided the generation of clear depictions of process and control
flow.  (We used Bennett in conjunction with Coad/Yourdon--CY to create
the object partitioning, and Bennett to capture the service
functionality.  Would've worked just as well with OMT's class diagrams.)

Even years after many of us left Link and the requirement to use it as
part of our jobs, we'll still employ it from time to time to help within
our current projects.

The book documenting Bennett notation is "Visualizing Software", by
William S. Bennett, and published by Marcel Decker, ISBN 0-8247-8714-5.

Marc A. Criley
Senior Staff Engineer
Quadrus Corporation
www.quadruscorp.com

Re: UML & Ada

[Martin Dowie]

"Marc A. Criley" <mcqada@earthlink.net> wrote in message
news:3C037EED.AB2A5AFD@earthlink.net...
> Ted Dennison wrote:
> >
> > In article <c5b88987.0111220631.39445b76@posting.google.com>, Mark
says...
> > >The key point is that a single "view" of a system can not fully
> > >describe it. You need at least a logical view (modularity) and an
> > >execution/process/task view.
> >
> > That sound a bit like the old data flow vs. control flow dichotomy.
>
> Though few are familiar with it, there is a unified control/data flow
> notation that was developed in the late 80s that was amazingly
> expressive and easy to use.  It was invented by William Bennett of Link
> Flight Simulation and christened "Bennett notation".  Anyone could be

Never heard of that one, but Ward/Mellor was _very_ popular (in the UK
anyway) and also combined control and data flow very easily (still my
favourite notation even after a while using UML). I understand they
were the first to combine STD into DFDs.

Their books are now amazingly cheap (compared to 10 years ago!)

"Structured Development for Real-Time Systems : Essential
 Modeling Techniques" and

"Structured Development for Real-Time Systems : Implementation
 Modeling Techniques"

I seem to remember a third book but can't find it on amazon...

Reading these should be mandatory for all s/w engineers even if
they never use the method or notations ever!!! :-)

Re: UML & Ada

[Mark]

"> 
> Though few are familiar with it, there is a unified control/data flow
> notation that was developed in the late 80s that was amazingly
> expressive and easy to use.  It was invented by William Bennett of Link
> Flight Simulation and christened "Bennett notation".  Anyone could be
> taught it in five minutes, and we found that it scaled extremely well,
> and aided the generation of clear depictions of process and control
> flow.  (We used Bennett in conjunction with Coad/Yourdon--CY to create
> the object partitioning, and Bennett to capture the service
> functionality.  Would've worked just as well with OMT's class diagrams.)
> 
> Even years after many of us left Link and the requirement to use it as
> part of our jobs, we'll still employ it from time to time to help within
> our current projects.
> 
> The book documenting Bennett notation is "Visualizing Software", by
> William S. Bennett, and published by Marcel Decker, ISBN 0-8247-8714-5.
> www.quadruscorp.com

Is there a Visio template or some other way to use this
notation?(haven't seen the book yet)

Re: UML & Ada

[Simon Wright]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> writes:

>                    But it would *still* be nice to have something
> that diagramatically represented all possible Ada programs - and UML
> can't be made to do this any way I know of.

I don't know if you ever saw Raymond Buhr's book System Design with
Ada (83)? He had a notation for every possible construct, it seemed, I
could never see the point, why not just write Ada!

Every description of a program has to leave something out; even the
code typically doesn't show the intent (well, not formally).

-- 
Simon Wright                       Work Email: simon.j.wright@amsjv.com
Alenia Marconi Systems                        Voice: +44(0)23-9270-1778
Integrated Systems Division                     FAX: +44(0)23-9270-1800

Re: UML & Ada

[Marin David Condic]

Yes, I recall Buhr's diagrams - I believe this was the basis for the Adagen
tool once being marketed by a company called Mark V systems. Probably others
as well - I think Software thru Pictures used this at one point.

Why not just write Ada? Easy! In fairly large systems, you often want to
understand things like how the whole system "hangs together" - the
heierarchy, what is using what, etc. Seeing a bunch of little icons
representing packages and having arrows between them can help with that sort
of overview. I like the concept of having a design tool that will filter out
levels of detail so you can increasinly push down into a design seeing more
and more as you need to. (Ultimately getting to the code.)

Maybe I want to start viewing a system as a set of "major" packages
(subsystems) to which certain responsibilities are assigned. These may break
down into sub-packages that flesh out more of the implementation details of
how those responsibilities are handled. From there, I may be interested in
seeing just the subprogram names of a package interface to see what is
callable within a given context. Pushing deeper, I'd want to see types and
objects made available in the spec. Pushing further into a package, I might
then want to see what the body provides, etc., etc., until I get down to the
code.

One of the problems with most of the tools I've encountered that let you
construct some flavor of a diagramatic representation of a program (in any
language) is that at some point, it is taking you more time to arrange icons
on the diagram than it would be to simply type the code. Providing a means
of typing the code & reverse engineering the diagrams has its own problems
in that the instant the code gets more complex than the "hello world"
examples, the diagrams that result get butt-ugly and you have to spend a
bunch of time rearranging them. And of course having rearranged them, you
can't generally re-generate the diagrams from modified code without messing
the whole works up again.

I don't know of any available tools out there that still provide Buhr
diagrams and the ones I used to know of never really did all the things I'd
like to have seen. Still, if something like the Visual Thought tool (see
http://www.confluent.com/) supported Buhr diagrams (or similar), at least
you could sketch-out an overview of what you intended to build (or already
found and need to document)

--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Simon Wright" <simon@pushface.org> wrote in message
news:x7v4rnkfwgp.fsf@smaug.pushface.org...
> "Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> writes:
>
> I don't know if you ever saw Raymond Buhr's book System Design with
> Ada (83)? He had a notation for every possible construct, it seemed, I
> could never see the point, why not just write Ada!
>
> Every description of a program has to leave something out; even the
> code typically doesn't show the intent (well, not formally).
>

Re: UML & Ada

[Mark Lundquist]

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
message news:9tc0qk$cee$1@nh.pace.co.uk...
> The last time I went informally cruising through anything to do with the
> UML, I seem to recall that it lacked any means of representing Ada tasks.
> Through interaction diagrams, you kind of had a way of saying how one
thread
> of control would interact with another, but when it came to representing
> something that should be translated to a task, complete with rendesvous,
> etc, there didn't seem to be any way of expressing this.
>
> I hear rumors from Dr. Dobbs that there is some sort of revision going on
to
> come up with a v2.0 of UML. Has anybody heard if there is any intention to
> support tasks with some kind of diagram symbol?
>

The real-time profile will be included in UML 2.0.  It comprises a tasking
model, timing, schedulability, quality of service, etc.

But I don't know enough about it to know how well it will map onto Ada's
tasking semantics.

-- mark
http://members.home.com/mlundquist2/consulting

Re: UML & Ada

[Bill Watkins]

Here's the UML Revision Task Force link (the UML 2.0 link is also on this
page):

http://www.celigent.com/omg/umlrtf/

Readers may also want to look here for more info on approaches to modeling
tasks and threads with UML:

http://www.therationaledge.com/content/may_01/t_activity2_bl.html

--
Bill Watkins
Applications Architect
Information Systems
Boeing Houston Operations


"Mark Lundquist" <up.yerz@nospam.com> wrote in message
news:BVzK7.51503$XJ4.30338539@news1.sttln1.wa.home.com...
>
> "Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
> message news:9tc0qk$cee$1@nh.pace.co.uk...
> > The last time I went informally cruising through anything to do with the
> > UML, I seem to recall that it lacked any means of representing Ada
tasks.
> > Through interaction diagrams, you kind of had a way of saying how one
> thread
> > of control would interact with another, but when it came to representing
> > something that should be translated to a task, complete with rendesvous,
> > etc, there didn't seem to be any way of expressing this.
> >
> > I hear rumors from Dr. Dobbs that there is some sort of revision going
on
> to
> > come up with a v2.0 of UML. Has anybody heard if there is any intention
to
> > support tasks with some kind of diagram symbol?
> >
>
> The real-time profile will be included in UML 2.0.  It comprises a tasking
> model, timing, schedulability, quality of service, etc.
>
> But I don't know enough about it to know how well it will map onto Ada's
> tasking semantics.
>
> -- mark
> http://members.home.com/mlundquist2/consulting
>
>
>
>

Re: UML & Ada

[Patrick Hohmeyer]

You may take a look at "Activity Diagrams"

It has a fork and a join notation.
Not as specific as Ada's rendezvous capability,
but with some comments it should work.

-- 
Patrick Hohmeyer

Re: UML & Ada

[David Phillips]

Marin,

I believe you will find that Ada tasks can be modeled as Active Classes in
the current UML standard.  An Active Class is a class that has it's own
thread of control.  The notation on the diagram is augmented by making the
border around the class on the diagram bold.  All places where the class is
referenced will have the bold border to signify that the class is active.

Hope this helps,

David R. Phillips
Member Senior Technical Staff
CohesionForce, Inc.

"Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org> wrote in
message news:9tc0qk$cee$1@nh.pace.co.uk...
> The last time I went informally cruising through anything to do with the
> UML, I seem to recall that it lacked any means of representing Ada tasks.
> Through interaction diagrams, you kind of had a way of saying how one
thread
> of control would interact with another, but when it came to representing
> something that should be translated to a task, complete with rendesvous,
> etc, there didn't seem to be any way of expressing this.
>
> I hear rumors from Dr. Dobbs that there is some sort of revision going on
to
> come up with a v2.0 of UML. Has anybody heard if there is any intention to
> support tasks with some kind of diagram symbol? UML seems to be bent on
the
> assumption that all software will be built via OOD & I've had a hard time
> ibn the past trying to make it fit something that was not OOD, so I'm
> wondering if they are at all considering making it more general - or at
> least making it possible to describe Ada programs more thoroughly.
>
> MDC
> --
> Marin David Condic
> Senior Software Engineer
> Pace Micro Technology Americas    www.pacemicro.com
> Enabling the digital revolution
> e-Mail:    marin.condic@pacemicro.com
> Web:      http://www.mcondic.com/
>
>
>

Re: UML & Ada

[Simon Wright]

"David Phillips" <dphillips@CohesionForce.com> writes:

> I believe you will find that Ada tasks can be modeled as Active
> Classes in the current UML standard.  An Active Class is a class
> that has it's own thread of control.  The notation on the diagram is
> augmented by making the border around the class on the diagram bold.
> All places where the class is referenced will have the bold border
> to signify that the class is active.

Rose (<=2000, anyway) doesn't bolden the borders of active classes.

You can probably make a good stab at mapping UML facilities onto
tasks, but there's a considerable mismatch. I guess it depends whether
you think the model should be constrained by the target
architecture. For example, do you ban operations that return values
(functions)? or do you provide some more complex mapping so that when
the analyst says "function" you know how to represent it?