Re: [Q]: How many lines of Cobol running worldwide???

From: Francis ANDRE (francis.andre_at_easynet.fr)
Date: 01/08/04


Date: Thu, 8 Jan 2004 00:54:34 +0100

It does not matter so much to count included lines by copy....compare to the
figures I got from the net....My question is more to understand if the
Gartner's evaluation is more or less close to the "reality"...200 billions
of Cobol lines!!!

75% of business data is processed in COBOL (Source: Gartner).
There are 180 billion to 200 billion lines of COBOL in use worldwide
(Gartner).
15% of new applications are written in COBOL (Gartner).
 Replacement costs for COBOL systems estimated at $25 per line are in the
hundreds of billions of dollars (Tactical Strategy Group).

FA

"Thane Hubbell" <thaneh@softwaresimple.com> a écrit dans le message de news:
bfdfc3e8.0401071224.423c4e0b@posting.google.com...
> You can also get into the whole "what's a line of code". Pre or post
> copybook expansion? Comments - do they count? Etc etc...
>
> A relatively small system I maintain has 140,000 lines of code
> precopybook expansion and including comments.
>
>
> Arnold Trembley <arnold.trembley@worldnet.att.net> wrote in message
news:<0zNKb.2083$Ub6.63921@bgtnsc04-news.ops.worldnet.att.net>...
> > LX-i wrote:
> > > Francis ANDRE wrote:
> > > (snip)
> > >> In my understanding from the various sources, it is counting in
> > >> billions of
> > >> lines....which is quite large in term of evaluation....Does anybody
could
> > >> give me more precise numbers, specially on past, present and future
> > >> estimations??
> > >
> > >
> > > Billions sounds right to me. Our one little Air Force aircraft
> > > maintenance system has about 1.2M, and the supply system we interface
> > > with has close to 5M.
> > >
> > >
> >
> > In 1996, as part of a Y2K project, we counted up how many COBOL
> > programs and lines of code we had in production in order to estimate
> > the scope of our Y2K work. It worked to something like 4,500
> > production programs and about 7 million lines of code. This was a
> > shop with maybe 100 to 150 actual programmers. Of course, the number
> > is a moving target as new programs are written, existing programs are
> > enhanced (usually resulting in a slight increase in LOC), and even
> > occasionally deleting an old program.
> >
> > I can easily believe the number would be in the billions just for the
> > USA, but there's no way to actually count it. We missed some programs
> > that weren't stored in the proper libraries.



Relevant Pages

  • Re: Gartner on Assessing the Age of Software Languages and Tools
    ... MicroFocus has a courtesy copy of a Gartner Report, ... MicroFocus are promoting it because it shows their implementation of COBOL ...
    (comp.lang.cobol)
  • Re: OT: Newsgroup Name Change?
    ... less than 200 billion lines of COBOL. ... equivalent COBOL programmers, there are about ... Ask Gartner what they mean by "legacy code". ... supports another notion of "the declining popularity of Cobol". ...
    (comp.lang.cobol)
  • Re: OT: Newsgroup Name Change?
    ... less than 200 billion lines of COBOL. ... equivalent COBOL programmers, there are about ... The intial statement mentioned "the declining popularity ... the information from Gartner does not ...
    (comp.lang.cobol)
  • Re: OT: Newsgroup Name Change?
    ... less than 200 billion lines of COBOL. ... 210,000 COBOL programmers in the US. ... estimated some 10,000,000 worldwide for 2000. ... Do these statistics from Gartner Group [Internet posting in ...
    (comp.lang.cobol)
  • Re: END-IF
    ... >> It is really a matter of opinion. ... COBOL ... They caused endless heartbreak, wailing, and gnashing ...
    (comp.lang.cobol)