Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!tut.cis.ohio-state.edu!rutgers!att!cbnewsd!popeye From: popeye@cbnewsd.ATT.COM (ken.a.irwin) Newsgroups: news.groups Subject: Re: CALL FOR DISCUSSION: comp.lang.cobol Summary: asdas Keywords: COBOL, Dead Language Message-ID: <3304@cbnewsd.ATT.COM> Date: 22 Nov 89 23:20:53 GMT References: <480@enea.se> <1989Nov21.013021.4170@eng.umd.edu> Distribution: na Organization: AT&T Bell Laboratories Lines: 99 This maybe an improper forum but this can not go unflamed, the squimish are encouraged to hit the "n" key. In article <1989Nov21.013021.4170@eng.umd.edu>, smaug@eng.umd.edu (Kurt Lidl) writes: > In article <480@enea.se> sommar@enea.se (Erland Sommarskog) writes: > >About all major programming languages have their own newsgroup, > >but Cobol has not. > > Yeah, ain't it great? Oh I don't know, it seems kinda stupid to me to have a group dedicated to a commercialy insignificant language like Pascal and not COBOL. And it seems we have a group dedicated to BAL, and I would say that about 99% of the machines running 370 assy have a majority of thier code written in COBOL. And we also have a group for the almost-as-antique FORTRAN. And since all the chip makers tell us how great RISC technology is, why is it that COBOL, the languge that most closly matches the RISC instruction set, is so frowned upon? > >Yet is the Cobol probably the most common > >language used in the software industry. The reason that Cobol > >doesn't have a newsgroup are quite obvious. Usenet has its > >roots in technical environments where Cobol is little used. > > USENET is also (at least for me) an instrument for INCREASING > the level of computer technology -- not reducing it. I would > hate to see the net being used as a means of saddling some > person with COBOL -- when the project could and probably should > be re-coded in something a little more modern. Why on earth would you recode a basic banking or insurance packages in a "more modern" languge, like what? Maybe C? tons of record reads and writes, huge amounts of report generation in fixed fields, shitloads of whole number arithmatic, maintainable, fast? Or maybe you are one who claims how slow COBOL is, slow to code: yes, slow to compile: depends on the compiler, slow to run: only if the company who wrote the compiler is a bunch of dweebs. COBOL is english language assembly and produces fast efficient object code with source code a 6 year old could follow, don't count on it going away any time soon, it will be around long after the revised edition of K&R E++ is printed by Prentice, Hall & Webster. 1/2 :-) > >But times are changing. Usenet are spreading into administrative > >environments. And technical and administrative are getting > >closer to each other. Myself, an electrical engineer, had to > >learn some Cobol for my current project. And, undoubtedly > >the number of Cobol articles in comp.lang.misc are increasing. > >OK, it's no deluge of Cobol articles, but what about it? > > Had to? Or was that simply the path of least resistance? I have > had to fight previous battles to have some things done correctly, > even if that means re-coding entire sections of code. Spend a > little time and consider the future... Gee, I'm sure Chase Manhattan would love to hire you to recode thier gigabytes of COBOL source, and I'm sure they wouldn't mind the measly few billion it would cost them. Why don't you spend a little time and consider the present. > >According the guidelines I calling for discussion now, and > >will call for votes (unless I'm severely discouraged) December > >3rd and close the vote on Christmas Eve. > > Please don't. [...listen to this guy] > > >Cobol is not dead, it just smells funny. > > Yes, an unpleasant fact of the industry. But please, don't help > propogate this foolishness. May your bank/mortgage company/insurance company/payroll service company be the first to completly rewrite in Pascal and pad your bill to recoup the costs. COBOL is alive and well and controlling your money, and your companys money, and your governments money. If I never write another COBOL program in my life it will be too soon, but to the thousands that do, they should be represented on the net. There may or may not be the readership justification for a COBOL group at present, but this is a case where a group should be created to fill an obvious hole in the name space. "comp" groups are the most widely carried on the net and to not represent such a commercialy important language, in my opinion, is a mistake. There are plenty of groups that have almost no daily traffic but are subscribed to by large numbers of people, a place to post questions and get answers when needed, where you don't have to wade through hundreds of articles and unsubscribe because you don't have time for it. This type of group should be created, it's why many companys bother to carry this stuff, so their people can get answers to their questions. > >Erland Sommarskog - ENEA Data, Stockholm - sommar@enea.se > /* Kurt J. Lidl (smaug@eng.umd.edu) | X Windows: Power Tools */ Ken A. Irwin AT&T Bell Laboratories Indian Hill 6G410 Naperville, Illinois (708) 979-4578 ...!ihlpa!kai