Re: [open-cobol-list] My getting started document

[David Essex]

I know that it is NOT what your article is saying, but human nature 
being what it is, users will likely misinterpret as Cygw*n being a 
requirement for OC.

There really is no such thing as a "official" way.
Just about any UN*X application will compile and run on Cygw*n with 
minimal changes.
So it is convenience not to support a Wind*ws native port.
This is NOT a criticism, as most native ports do require a dedicated 
developer.

As for the 'unofficial' way, perhaps you should try using MinGW and MSYS 
(Minimal SYStem providing a POSIX compatible Bourne shell environment).
I think you would find is not much different, than Cygw*n, and in some 
ways preferable. Certainly no more difficult for a "newbie".  I think 
perhaps less difficult.

Something you may wish to consider adding to your article.
In Wind*ws  there are console and winapi applications. The default for 
OC is console. If you add a '-mwindows' option to the command line, GCC 
will automatically create a winapi application.


Bill Klein wrote:
> David, (et al)
>  The latest version of my document (dated March 2, 2009) and not yet
> generally posted includes the following paragraph - that I *think* addresses
> your issue:
>
> "NOTE: Within the OpenCOBOL forum, there are a number of threads concerning
> installing OpenCOBOL as "native" in Windows.  They seem to report varying
> degrees of success (and ease of installation). This document does NOT
> address that approach.  Although it may (eventually) be the best way to go
> for existing Windows-only users, it seems to me that most of the help and
> information for OpenCOBOL assumes a Linux or UNIX environment.  This
> document gets you going in such an environment (Cygwin) and I recommend it
> for the "new" OpenCOBOL user.  However, as usual "your mileage may vary"! If
> you want to go with a native Windows OpenCOBOL environment, check out the
> OpenCOBOL forum for hints and tips on installation and use."
>
> Certainly, if someone has the "energy" (and endurance" to provide comparable
> information for the other approaches, I would be happy to reference such.
> However, given the number of problems I experienced trying to do the
> "official" way, I think that I would NOT suggest a "newbie" try an
> unofficial way.  As noted in the document, I also don't address any of the
> "real user" (after initial getting started) issues such as configuring for
> MF or IBM conversions or tuning for production work.
>
> I may get around to doing some of this (eventually).  These certainly ARE
> real topics of interest. However, they are simply beyond my initial
> document.