[Info-vax] Whither VMS?

[P. Sture]

In article <rrM5wGEDzZv8 at eisner.encompasserve.org>,
 koehler at eisner.nospam.encompasserve.org (Bob Koehler) wrote:

> In article <paul.nospam-5F565E.16454311102009 at pbook.sture.ch>, "P. Sture" 
> <paul.nospam at sture.ch> writes:
> > 
> > FORTRAN was the first HLL compiler offered on the VAX and in early days 
> > the only one.  We were in the market for a new system at the time and 
> > that ruled the VAX out for us; we wanted a COBOL capable system.
> > 
> > What VAX FORTRAN did offer was the header files for calling system 
> > services, although these were an option (IIRC still are?) on 
> > installation (curses all round if someone had said NO to that option).
> 
>    Those header files didn't ship with the compiler ever, that I recall,
>    and didn't ship with VMS until VMS 3.0, I recall dealing with home-built 
>    header files under 2.5.  IIRC, COBOL was available during 2.x, but not 
>    native, and native COBOL later during the 2.x timeframe.

I've just dug into the FORTRAN 6.6 (VAX) KITINSTAL.COM:

    The Compaq Fortran 77 kit contains three components:

        Compiler (includes help text, command definition, IVP)
        Enhanced Run-Time Libraries (required if Compiler selected)
        FORSYSDEF.TLB (OpenVMS system symbol definition library)

FORSYSDEF.TLB is what I'm thinking of. Things like $IODEF are in there, 
and the installation procedure builds it:

   When a new OpenVMS version is installed on this system, be sure to
    reinstall Compaq Fortran 77 to update FORSYSDEF.TLB.

No version of VMS COBOL I have ever seen includes an equivalent, 
although you can (since the native version was shipped or later?) define 
external variables which are resolved by the linker.  You still have to
define your own structures.

-- 
Paul Sture