From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Thread: 103376,d2fdf39976bd1585
X-Google-Attributes: gid103376,public
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news2.google.com!news3.google.com!border1.nntp.dca.giganews.com!nntp.giganews.com!opentransit.net!newsfeed.icl.net!peer-uk.news.demon.net!kibo.news.demon.net!news.demon.co.uk!demon!not-for-mail
From: Simon Wright <simon@pushface.org>
Newsgroups: comp.lang.ada
Subject: Re: ANN: Ada source code decorator
Date: Sat, 27 May 2006 13:06:24 +0100
Organization: Pushface
Message-ID: <m23bev65a7.fsf@grendel.local>
References: <447306ee$0$11066$9b4e6d93@newsread4.arcor-online.net>
 <m24pzg78w4.fsf@grendel.local> <ur72j0wym.fsf@acm.org>
 <eMadg.1943$No1.1153@attbi_s71> <ufyiy9pwi.fsf@acm.org>
 <4475d870@news.upm.es> <1148622876.114738.62910@i40g2000cwc.googlegroups.com>
 <m2ejyg5xth.fsf@grendel.local> <4392612.ATc5MT7D57@linux1.krischik.com>
NNTP-Posting-Host: pogner.demon.co.uk
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Trace: news.demon.co.uk 1148731585 9166 62.49.19.209 (27 May 2006 12:06:25
 GMT)
X-Complaints-To: abuse@demon.net
NNTP-Posting-Date: Sat, 27 May 2006 12:06:25 +0000 (UTC)
Cancel-Lock: sha1:WJnfGrG/3PJDzMZpcWF5h+9HDts=
User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.50 (darwin)
Xref: g2news2.google.com comp.lang.ada:4533
Date: 2006-05-27T13:06:24+01:00
List-Id: <comp.lang.ada>

Martin Krischik <krischik@users.sourceforge.net> writes:

> I think that this a full natural way of mapping Ada <-> UML - After
> all 'First and 'Last are called Attributes on both sides. And if you
> continue the way of stereotypes then you can map all Ada feature to
> UML.

Hmm, yes -- I see your point, though it's certainly not the route I
took with http://coldframe.sf.net/. The difference is because I was
coming from a Shlaer/Mellor OOA viewpoint, and also I'd much rather
have a model that accurately reflects the problem domain
(platform-independent model, PIM) and that tools can translate into
code (possibly via a platform-specific model, PSM).

I certainly haven't achieved that with ColdFrame, which is a framework
generator, and has a lot of vaguely Ada-like stereotypes to support
code generation; for example, <<atomic>>. Well, really that could have
a meaning in a translator to C, but it's hardly part of the
application's subject matter[1], and doesn't make it any easier to
discuss with a client whether you are solving the right problem!

So I'm uncomfortable with the idea of having many stereotypes to allow
you to specify every detail of Ada in a model. Why not just write the
Ada? The model needs to be an abstraction. People may remember Raymond
Buhr's book "System design with Ada", which had a diagrammatic
convention for every aspect of Ada 83 and which I can't imagine helped
anyone develop real software.

XMI which could accurately reflect full Ada would have to be extended;
I know UML can be extended using profiles (ie, stereotypes, tagged
values and the associated semantics), one would suppose that XMI
reflects this.

[1] Of course, if you _have_ to say <<atomic>> it's possible that the
subject matter in hand may be one where it _is_ appropriate to discuss
this sort of thing. It seems to tend to the 'implementation detail',
though.