[Info-vax] VAX/VMS V1, V1.5 or anything older than V5

[Nomen Nescio]

glen herrmannsfeldt <gah at ugcs.caltech.edu> wrote:

> Nomen Nescio <nobody at dizum.com> wrote:
> 
> (snip on old VAX systems)
> 
> >> And you can even run TSO and see what IBM thinks of time-sharing.
> 
> > To be fair, TSO was no worse than what was available elsewhere 
> > *at the time* and in many cases, considerably better. 
> > Are you really going to tell me TSO edit is worse than 
> > flipping load switches on a PDP box (that wasn't built for 
> > another decade or so) or even using TECO (even though it 
> > hadn't been developed and wouldn't be for a decade?) 
> > Look at all the compilers that OS/360 had available, 
> > 2 versions of FORTRAN, PL/I, COBOL, RPG, and even things 
> > like ALGOL68, SNOBOL4, etc. There's a lot of fun stuff 
> > you can do. Show me another system from the late 1960s 
> > that's even half as capable or productive.
> 
> Hmm. My first time with DEC timesharing was TOPS-10 in 1976.
> 
> I was using OS/360 batch with WYLBUR from 1972.
> I didn't use TSO until 1979.

I liked SuperWylbur, it had a great macro language. I don't know if I ever
used Wylbur though. I punched cards through most of the early years and I
didn't use TSO until there was ISPF, and it was fine. I actually spent years
on CMS before I ever saw ISPF and I didn't like ISPF as much of course, but it
does what you need. The main thing missing was XEDIT and REXX. For decades now
REXX has run under TSO so that's not an issue anymore, but I still miss
XEDIT when I work in shops that don't run VM.

> I might have known which years different DEC systems became
> available, but mostly not. I believe TSO came late in the OS/360
> years, but again I am not sure when.
> 
> > What was the state of UNIX in 1970? VMS? Let's compare apples 
> > to apples if you want to make smartass remarks about OS/360 ;-)
> 
> It wasn't supposed to be a smartass remark. Most of the time, 
> I liked OS/360 batch. I never complained about JCL, which I know
> many people didn't like. 

I don't know of any mainframe people that didn't like JCL. Like everything
IBM does, it's perfect for what it is for. The only complaints seem to be
from UNIX fanatics like our buddy Johnny Sunkissed who never came close to
actually using a mainframe, but he does know people who might have used one
in the 80s.

> Most of the time when I was using TSO, it was for submitting
> batch jobs, where I think WYLBUR does a much better job.

Agreed, and so does CMS. The source for Wylbur is supposed to be available
somewhere, but I haven't looked into it. I think there was some discussion
about getting the code for SuperWylbur too but it is actually still being
marketed, believe it or not. There were some other third party systems too
but I don't recall using them. 

> In 1975, I was using Call/OS running on an IBM S/370 as a
> time-share system, which was also fine. The compilers aren't
> as good as the OS/360 compilers, though. I was mostly using PL/I,
> but only had the OS/360 PL/I (F) manuals. Many features were
> not supported by the Call/OS compiler.

That's true and not many people remember that. I don't know why they did
that. I didn't look into Call/370 or Call/OS internals but I do remember
they had subset compilers for some reason. I wonder if it was a pricing
issue. Luckily, the installation I was at that did run Call/OS also had a
couple other machines running MFT and MVT so we did have whatever we needed
when we needed it.

> >> Though VM/370 also runs on Hercules and isn't so bad 
> >> a time-sharing system.
> 
> > VM/370 with CMS and XEDIT and REXX is a combination that's 
> > hard to beat for productivity and stability even today.
> 
> Yes.
> 
> -- glen