From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.5-pre1 (2020-06-20) on
	ip-172-31-74-118.ec2.internal
X-Spam-Level: *
X-Spam-Status: No, score=1.8 required=3.0 tests=BAYES_50,TO_NO_BRKTS_PCNT
	autolearn=no autolearn_force=no version=3.4.5-pre1
Date: 20 Aug 93 03:46:46 GMT
From: world!srctran@uunet.uu.net  (Gregory Aharonian)
Subject: Re: Free Hawaii trip if you buy my Ada products
Message-ID: <SRCTRAN.93Aug19224646@world.std.com>
List-Id: <comp.lang.ada>

>     That's what the mandate is trying to do.  It's telling the military
>types who are in a 3-year tour of duty at the beginning of a 30-year project,
>"You can use another language if you can demonstrate a cost savings over all
>30 years.  Otherwise, we don't want to hear about the amount that can be
>saved during your watch."  If only they would can an Admiral or General or
>two for ignoring the mandate.  (I have a nominee.)
> Charlie

   Unfortunately, you also cannot demonstrate that use of Ada over a
30-year project is cost effective.  While many assume so, there exist
no validated economic model or data sets to prove so.

   In fact, the very few studies that do exist conflict with each other.
While Riefer's data shows Ada is about 30% more cost effective, Caper
Jones' data based on Function Points shows Ada at least 30% less cost
effective, thus canceling each other.
   Similarly, while the Air Force's Ada/C++ study of 1991 wimpily shows
that Ada might be 30% more cost effective (except for where the various
substudies contradict each other), an earlier SEI report undermines most
of the assumptions used by SEI and the others (TRW,NPS,CTA).  So like the
few data sets, the few economic studies also contradict each other.
   Then you have guys like Strassman as head of CIM saying Ada is they
way to go, and saying shortly out of a high office that Ada isn't the
way to go, it is less cost effective.

   So sure, it's hard to prove use of C++ or Smalltalk will be more cost
effective over the 30 year life cycle.  It's also hard to prove use of
Ada will also be more cost effective over the 30 year life cycle.

   And until the DoD shows some willingness to treat this issue with a
competent business/economic analysis, people will be justified in using
whatever language they want.  They can't prove their choice is right, and
you can't prove their choice is wrong.
-- 
**************************************************************************
 Greg Aharonian                                      srctran@world.std.com
 Source Translation & Optimization                            617-489-3727
 P.O. Box 404, Belmont, MA 02178