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=1.0 required=3.0 tests=BAYES_40,TO_NO_BRKTS_PCNT autolearn=no autolearn_force=no version=3.4.5-pre1 Date: 21 Jul 93 13:53:23 GMT From: world!srctran@uunet.uu.net (Gregory Aharonian) Subject: Re: COTS 2 percent in Ada Message-ID: List-Id: >>Well most COTS stuff is not done in Ada >>(less than two percent of all COTS stuff has Ada inside), so Perry's >>policy, if successful, will do as much to undermine Ada as the current >>hypocrisy that is tolerated. >Two percent. Is that by weight or volume (contents may have settled during >shipping :-)) or by amount of CPU cycles consumed on a yearly basis or >dollar volume in the "software market" or what? What's the margin of error >on the survey (whose source, by the way, you have conveniently failed to >mention)? What are the percentages for other languages of interest? At the recent debate at the WadaS symposium, I presented a lot of data on programming language use, mostly with regards to C/C++ and Ada. As I have said many times before here, my data and analysis vis-a-vis the 2% figure are neither comprehensive nor methodical. They are however, in the Ada world, the only data available, and hopefully my year long campaign to interest someone in the DoD to fund a study to find out the truth and do a comprehensive survey and census of programming language use will eventually succeed :-) However, I can give you an rough idea of how I calculate the 2% figure, which is probably very generous, for reasons I will explain below. At the debate, I presented data on indicators of programming language use. These indicators included new jobs in help wanted ads, numbers of companies and products for each language, surveys of embedded use and R&D use, use by top fifty independent software vendors, thesis use, and number of product bindings available in each language. I also was subjectively biased by knowledge of the contents of my database of information on all of the reusable computer programs in the country (15,000+ and growing) coming out of defense, government, university and private facilities. I mention this not so much because it influences the analysis, but more so to plug my software reuse consluting services :-) The results were, discounting any government (DoD/NASA) use of Ada, that use of C/C++ is forty times more prevalent than use of Ada outside the Mandated world, which when you include data for other newer languages like Smalltalk, translates into about a 2% COTS share for Ada. Given that the non-Cobol/non-Fortran world is only about 40% of all programming, you can probably conclude that Ada's COTS share is about 1%. (Also, without detailed investigation, it is hard to subtract out sales of non-Ada products by Ada companies, which has some affect on these figures). Admittedly, not the most accurate or comprehensive analysis, but it is the only analysis. This is very sad and very dangerous for Ada, since my data was collected over many years, and the trend shows little changes over the next few years. The DoD, including the Ada9X office, the STARS project, and Perry's COTS initiative, are developing strategies and policies in the blind that based on past results, will only serve to weaken Ada's market share outside the Mandated world. What's the truth? Who knows? And frankly, who cares? I have yet to encounter anyone inside the DoD who gets alarmed by either these figures, or the lack of more reliable figures. The DoD does not fight wars without military intelligence on the enemy and its terrain - why is it doing so in its Ada battle? -- ************************************************************************** Greg Aharonian Source Translation & Optimization P.O. Box 404, Belmont, MA 02178