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-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,c6e9700a33963193 X-Google-Attributes: gid103376,public From: Richard D Riehle Subject: Re: The future of Ada Date: 1999/03/12 Message-ID: <7cc7v3$gre@dfw-ixnews5.ix.netcom.com> X-Deja-AN: 454372210 References: <36E690FA.4B9C@sandia.gov> <7c7coa$nvt$4@plug.news.pipex.net> <1999Mar11.080820.1@eisner> <7c92hb$r8n@dfw-ixnews3.ix.netcom.com> <87g17axtv2.fsf@mihalis.ix.netcom.com> Organization: Netcom X-NETCOM-Date: Fri Mar 12 5:28:35 PM CST 1999 Newsgroups: comp.lang.ada Date: 1999-03-12T17:28:35-06:00 List-Id: In article <87g17axtv2.fsf@mihalis.ix.netcom.com>, Chris Morgan wrote: >Richard D Riehle writes: > >> There is another argument in favor of Ada that is beginning to >> manifest itself in organizations converting to C++: employee >> turnover. > >I'm sorry Richard but this reasoning absolutely disgusts me. Well, Chris, it has upset others besides you. ... >We are not cannon fodder to be >kept away from information that might help us, we deserve as green a >pasture as the next man. Did not mean suggest than anyone is "cannon fodder." On the other hand, DoD contractors are not known for their generosity. For many, when you are no longer "billable" you are superfluous. Sometimes you are, indeed, regarded as "fodder." >You seem to be saying "keep hold of staff by keeping them from having >a marketable skill". I am saying that, training people in skills for some other marketplace is a good way to encourage them to seek opportunities in that other marketplace. There is rarely a good technical reason to abandon Ada in favor of C++. Doing so does open new career opportunities for the programmers. The resulting employee turnover is inevitable. That might not be a bad thing. It will happen. An employer must understand this. If you stay with Ada, the probability of such turnover, at least from this cause, is diminished. >People can see through such tactics. Even if they >work in 100% Ada at the office, how can you stop them developing e.g. >killer Perl skills at home and going to be a webmaster for a bank? You >can't. Absolutely true. There are some self-starters out there. They are "inner directed" and will develop new skills on their own. Some will take these new skills into the marketplace in search of new jobs. Others will write some "killer app" and become entrepreneurs. These kinds of people will never be stopped - I hope. >You have to keep hold of staff by making them want to work at >your company (money, equity participation, technology, management >attitude, any number of factors). These are certainly factors in personnel retention. If you keep your people happy, most will want to stay with you. But many people can "resist anything except temptation." And the temptation to see whether it is possible to improve your lot with newly acquired programming language skills is an on-going temptation. Employee turnover in software is not rare. > In fact you should be able to hire C++ victims and convert them to Ada. I would not characterize C++ programmers as "victims." I do understand what you mean by the sentence. It is not clear how this relates to the issue of employee turnover. >I worked on a huge Ada project where some of the people got to do C++ >and some didn't. It caused a lot of resentment because they failed to >make any effort to improve the lot of the Ada programmers once it >became clear they were paying less than the going rate for >programmers. Exactly. This returns to the employee turnover issue. The going rate for programmers varies all over the place. At present, some DoD contractors and sub-contractors are below the high end of the compensation rates. Here in Silicon Valley, programmers are sometimes given a salary and stock (at minimum stock options). This is rarely the case for large DoD contractors. >In fact the management showed some contempt for the mass >of us Ada programmers which were their prime asset. The benefits were >average, the hours long, the technology mostly backwards, yet I liked >it (I got to use GNAT for money). If they had simply paid anything >close to the market rate they could have kept hold of a lot of us, but >they didn't. I would have preferred to continue to work in Ada even at >a small salary disadvantage, but not for a 25-50% one, defence >projects are hard enough work as it is. OK. Now, take those dissatisfied programmers and teach them a set of skills that makes them even more attractive in the marketplace. Do you seriously expect them to ignore those outside opportunities. At present, DoD budgets are tight, competitive bidding demands cutting salaries to the lowest margins, and software is still not given the respect we give the hardware engineering. It is very difficult for a DoD contractor to match market rates. This is one reason why there is so much emphasis on COTS software, a trend that will eventually come back to haunt us. >I'm not interested in working for a company that pays a lot less than >the going rate for good programmers whatever the language. I would be >even less likely to work for a company that had the attitude you are >recommending. I want them to like using Ada for better reasons than >that. As it turns out, the company in question has had a revelation >and is now writing to all its ex-employees trying to tempt them back. Glad to hear the attempt at re-hire from the "company in question." I too would like them to select Ada for "better reasons." But management will rarely make decisions on the basis of better technology. Such decisions are made for largely economic reasons. The economics will be manifested in many forms. The reasons are often limited to the cost of development tools, cost of hiring, and the cost of retention. The DoD gives no incentive for software productivity, no incentive for software reuse, no incentive for future maintainbility. Without these incentives, stated in economic language, there is every incentive to take a short-range view of language selection. Choosing Ada requires a long-range view of the software process. It requires enlightened managers. It requires program managers who understand how Ada will benefit their mission. There are some of these. I personally know some of the enlightened managers. I wish there more. So, Chris, I understand your concern with my contention that employee retention is a factor in language choice. Unfortunately, it is. It would be nice if it were otherwise. Richard Riehle richard@adaworks.com http://www.adaworks.com >Sincerely, > >Chris >-- >Chris Morgan "We're going to start selling Linux to single-party users very > soon. Q: It's going to be on the menu? A: Yes. You'll go to Dell, > pull down "operating system," and click "Linux." - Michael Dell