Re: Why "Hello World" as a first exercise?

["Randy Brukardt" <randy@rrsoftware.com>]

"Dennis Lee Bieber" <wlfraed@ix.netcom.com> wrote in message 
news:6199gg1vsegpsqj6prdrqmmme4qo8avk71@4ax.com...
> On Fri, 30 Jul 2021 18:06:21 -0500, "Randy Brukardt" 
> <randy@rrsoftware.com>
> declaimed the following:
>
>
>>My first actual programming class spent the first two or three sessions on
>>the administrative things: where is the computer center? How do you use a
>>keypunch? (I admittedly am showing my age here; but at least we were the
>>second last semester to use the keypunches.) How to submit a card deck? 
>>What
>>magic incantations are needed to get the computer to accept a card desk?
>>Where to find your results afterwards (this being a batch system)? Etc. 
>>The
>>actual program was very secondary to all of that (I don't remember what it
>>was, but we had to key it and submit the results -- in order to prove that
>>we understood all of the admistrivia).
>>
> Sounds like my college... Here are the three 029 keypunches... Here's
> how to program a drum card to simplify entering code... Here's the minimum
> JCL to run FORTRAN(-IV) (Sigma CP/V had two FORTRAN compilers -- the
> traditional compiler outputting a relocatable object [ROM] file, to be
> followed by a linker outputting a load module [commonly called a LMN 
> file];
> the OS didn't use file extensions, so our practice was to name the source
> S:xxx, object O:xxx, executable L:xxx. The other compiler was FLAG --
> FORTRAN Load And Go -- compile/link/execute with one invocation). Turn in
> the card deck to the operators, here. Come back later to pick up your
> printed output.

Ah. We had the very advanced self-service card reader for simple jobs. You 
put your card deck in, pushed a large button, watched a very impressive 
swooshing of cards about, and then went and stood around a desk-sized 
printer with lots of other people waiting for a page with your user name in 
very large letters to head a printout, rip it off (preferably leaving anyone 
elses that was attached -- didn't always happen), and go read the output to 
see what you did. The original compile-execute-debug-repeat cycle (more like 
run-read-punch new cards-repeat cycle).

They had a few Decwriters, but only upper classmen got to use them (and they 
wasted tons of paper). Real terminals showed up the next year -- by the time 
of the compiler contruction class, most of the classes had moved to PDP/11s 
(way slower), but the compiler construction was still on the mainframe. But 
almost everything was done on the terminals (Janus/Ada never was on punched 
cards, thank goodness). We had to buy one of those huge computer tapes to 
rescue our source code and use another labs capability to transfer that to 
floppies in order to move our work to the CP/M computer on which RRS was 
born. A lot more engineering went into that sort of issue than today 
(probably a good thing).

                             Randy.


                                    Randy.