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: 111d6b,328622178ec8b832 X-Google-Attributes: gid111d6b,public X-Google-Thread: 10d15b,328622178ec8b832 X-Google-Attributes: gid10d15b,public X-Google-Thread: 109fba,a03ae7f4e53958e1 X-Google-Attributes: gid109fba,public X-Google-Thread: 1014db,a03ae7f4e53958e1 X-Google-Attributes: gid1014db,public X-Google-Thread: 103376,8775b19e3c68a5dc X-Google-Attributes: gid103376,public X-Google-Thread: 114809,a03ae7f4e53958e1 X-Google-Attributes: gid114809,public X-Google-Thread: 1094ba,a03ae7f4e53958e1 X-Google-Attributes: gid1094ba,public From: gwinn@res.ray.com (Joe Gwinn) Subject: Re: Which language pays most -- C++ vs. Java? Date: 1998/01/29 Message-ID: #1/1 X-Deja-AN: 320286423 Content-Transfer-Encoding: 7bit References: <67et6o$dql@bgtnsc03.worldnet.att.net> <67ktrg$ibk@bgtnsc03.worldnet.att.net> <883319809snz@genesis.demon.co.uk> <68bt2p$d48@lotho.delphi.com> <34a991f0.2379476@news.diac.com> <68dm0i$brv1@news.fiberlink.net> <01bd198f$4050d960$68c8b5cc@dhite.unicomp.net> <34B71B71.1EFDCAD8@ix.netcom.com> <34B8DC0F.BA0554DB@acm.org> <01bd1ebd$8580b9a0$b2684bc2@xzSys> <34BA520B.534F@mail.state.wi.us> <6alu5l$onm$1@owl.slip.net> <01bd2c2a$69b107a0$9f684bc2@xzSys> <01bd2c45$f7ff6f40$7261b693@HP5079Q> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Organization: Raytheon Electronic Systems Newsgroups: comp.lang.java.misc,comp.lang.c,comp.lang.c++,comp.lang.fortran,comp.lang.cobol,comp.lang.smalltalk,comp.lang.ada Date: 1998-01-29T00:00:00+00:00 List-Id: In article <01bd2c45$f7ff6f40$7261b693@HP5079Q>, "dogmat" wrote: > > > > It's really very simple. Anyone smart enough and focused enough to > > earn a > > > > PhD in *any* hard-core technical subject can learn programming and > > > > computer science on the job, and is therefore a good bet. > > > > > > Some of the worst code and UI design I've ever seen came from a guy > > > with *3* PhDs. I can only guess that people with a tolerance for the > > > BS of academia also have a tolerance for BS/ugly code and designs. > > Both points probably true and not inconsistent. Programming is a special > subject and a PhD doth not a programmer make. Agree. See previous posting as well. > > I knew another guy, PhD in mathematics. It seemed strange to me that he > > ended as a programmer. But it really struck me when I saw some of his > progs > > : you ever unlocked a record that doesn't exist? Yes, I have, many times, and it didn't require any special schooling. > > By the way, my thoughts, based on experience (and I have no PhD) : Having > 3 > > PhD's warrants that one can say out loud (and probably proudly) that he > has > > 3 PhD's. Furthermore, directives who care more for appearance than for > > substance tend to preffer guys with, at least, a PhD, since they fit > > fantastically in their corporative picture. Fortunately there's always > > still enough no-PhDed-people around to get the work done. > > Well, since we're taking off our gloves, I'll throw back the insult (I have > a PhD): > > In engineering, if the programmer writes the application, it'll look great > and be absolutely useless. > If the engineer writes the application, it'll look like hell, both > internally and externally, but it will actually be useful. My experience has been that it's lots easier to teach an engineer programming than to teach a programmer engineering, and the reeducated engineer will write perfectly fine code. Engineers, reeducated or not, as a group tend not to be too impressed with all the current good-programming theories and fads, which is generally a good thing, although it does drive the software process folk to drink, which is also a good thing. > Aside: Its comes down to requirements analysis, you might say. Sorry, but > get more than 3 engineers > in a room, and they'll argue forever. That's why we appoint a Chief Engineer. Somebody must decide. > ... You'll never get a engineering > software requirements package that is sufficiently precise for a > programmer's needs. Blame the engineers for their lack-of-rigor? Yes, but > that's the nature of the domain beast. > Its easier for an engineer to write the program himself or train the > programmer to be an engineer than it > is to write out the specifications "in full". Yes, up to a point. My experience is that engineers untrained in software can generally get a 20,000-DSI program to work, but will most often fail at 100,000 DSIs and above. As the scale increases, software architecture and design issues become more and more important, and cannot be overwhelmed by pure engineering knowledge. Thus, the engineer turned programmer is necessary. > Perhaps it comes down to training. I doubt many programmers would be > willing to go back and spend 4+ years > learning the intricacies of engineering principles (most haven't even a > clue about numerical methods). > This cannot be picked up on the job. But anyone with an engineering > background can learn to program > (and does, e.g., one semester of Fortran). The problem is that many of them > think they're good at it. > > Or maybe its just caring to read a new book every now and then. I'm an > engineer self-taught in OO, and I can run rings around most programmers. > (Although, not of course someone like yourself who follows these > newsgroups. If you are reading this, you are different from the majority of > dead-fish.) P.S. This is too depressing to be a boast. I don't expect programmers to learn engineering anytime soon, but there is one big problem I do have with most programmers -- they are ignorant of assembly code and the internals of generated code, runtime systems, operating systems, etc. This ignorance causes serious problems; it simply isn't enough to know only the surface of . Some learn these things on the job, but most don't.