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,FREEMAIL_FROM
	autolearn=ham autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,f948976d12c7ee33
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2003-06-24 11:01:17 PST
Path: archiver1.google.com!postnews1.google.com!not-for-mail
From: snarflemike@yahoo.com (Mike Silva)
Newsgroups: comp.lang.ada
Subject: Re: Boeing and Dreamliner
Date: 24 Jun 2003 11:01:16 -0700
Organization: http://groups.google.com/
Message-ID: <20619edc.0306241001.14dbc0e@posting.google.com>
References: <n%RJa.10829$Kg7.1489@nwrdny01.gnilink.net>
 <UlSJa.7517$e26.7165@rwcrnsc52.ops.asp.att.net>
 <i3YJa.40543$hI1.21397@nwrddc01.gnilink.net>
NNTP-Posting-Host: 154.6.152.68
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1056477677 31540 127.0.0.1 (24 Jun 2003 18:01:17
 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: 24 Jun 2003 18:01:17 GMT
Xref: archiver1.google.com comp.lang.ada:39686
Date: 2003-06-24T18:01:17+00:00
List-Id: <comp.lang.ada>

Hyman Rosen <hyrosen@mail.com> wrote in message news:<i3YJa.40543$hI1.21397@nwrddc01.gnilink.net>...
> tmoran@acm.org wrote:
> >   Whoa there!  A numerical value out of range is not a buffer overflow in
> > the languages used on c.l.a.  Surely as a programmer you appreciate the
> > value of precise expression.
> 
> In cause and effect, the two are precisely the same. The program with
> the potential buffer overflow works for inputs of a certain size, but
> fails catastrophically when input exceeds that size. That program is
> seldom documented with a description of this limit. The Ariane 4 SRI
> program worked fine for BH in a certain range, but failed
> catastrophically when input exceeded that range. That range was not
> documented.
> 
> Please tell me what you perceive as the meaningful difference between
> the two cases?

The difference is that the Ariane 4 software behaved *exactly* as
designed and intended for *all* ranges of inputs, for the Ariane 4. 
Raising an exception was not "failing catasrophically" but was the
correct behavior, triggering the correct shutdown of the unit, for the
data in question.  What data limitation is there to document, when the
software performs correctly for all data?

Then the Ariane 5 people came along, and they had an unstated, and
apparently even unrecognized, requirement for *different* behavior
over a certain range of data.  That certainly does not mean there was
a bug in the Ariane 4 software.

Mike