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=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,fedc2d05e82c9174 X-Google-Attributes: gid103376,public From: Richard D Riehle Subject: Re: Ada 83 - Sometimes still chosen Date: 1999/03/25 Message-ID: <7dc9ld$2u6@dfw-ixnews9.ix.netcom.com>#1/1 X-Deja-AN: 458770438 References: <87aex3pue4.fsf@mihalis.ix.netcom.com> <36F913E0.75F51763@lmco.com> <7dbcj3$e0l@dfw-ixnews7.ix.netcom.com> <7dbvl2$5bl$1@nnrp1.dejanews.com> Organization: Netcom X-NETCOM-Date: Wed Mar 24 9:13:49 PM CST 1999 Newsgroups: comp.lang.ada Date: 1999-03-24T21:13:49-06:00 List-Id: In article <7dbvl2$5bl$1@nnrp1.dejanews.com>, robert_dewar@my-dejanews.com wrote: >Sure, we know that some projects get stuck using obsolete >hardware, software, operating systems, compilers etc, and >that is understandable (and not necessarily an >inappropriate decision). I am uneasy with the selection of the word "obsolete." A technology is obsolete if it is not the best choice for the solution to a particular problem. Sometimes a newer CPU architecture might offer some improvement in performance, but the price/performance ratio might make the obsolete processor a better option. I return to the example of the 1750A. This processor, by all standards for earth-bound computing is obsolete. For a variety of technical reasons, it is still the optimal choice for many space applications. Sometimes the economics of an 8086 can make it the correct choice. In many circumstances we find some of our clients still choosing "obsolete eight-bit processors because they are excellent at small,single threaded tasks -- and one can populate a circuit board with them for incredible throughput. Ada 83 is the only Ada compiler available for many of these "obsolete" processors. But they are not obsolete. It is not likely anyone will bother to write a production quality Ada 95 compiler for an 8086, but some designers will continue to choose that processor because it offers better economics for the application being designed. >But in comp.lang.ada, I think we do not want to be >emphasizing this kind of stuck-in-the-dark-ages view. I know you like to tease us with hyperbole from time to time Robert. My own antiquity is such that I do long for those simpler times when I could repair my own car, map the entire memory of my program, and see the actual printing on a printed circuit board. Sigh. I suppose those were truly dark ages. >Ada 95 is a modern, recently designed language, and it >is important to always have the message that Ada is alive >and well (not just that ancient projects forced to use it >a long time ago are still alive and well). I agree with your observations about Ada 95. I am also delighted with the progress we have seen you and your colleagues at ACT make with it. I wonder if those of us, living in the dark ages, will see a time when Ada 95 will be available for some of those platforms you consider obsolete. I suspect not. It is a matter of economics. Consequently, when an older architecture is chosen, for perfectly good technical reasons, it would be appropriate to use a compiler that already exists. Often, this will be Ada 83. >Sure, if someone from one of these projects needs to know >something, we can probably provide the answer here to >questions about Ada 83, provided the question specifies >this (for that matter, this is probably the only place you >could still get Jovial questions answered :-) I realize you were not writing off those who have a genuine need for Ada 83 information. Also, there are few people who subscribe to this channel who are more able to answer such questions. I simply got a little over-exercised from your choice of the word "obsolete." Perhaps I took it to personally. :-) Richard Riehle richard@adaworks.com http://www.adaworks.com