Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.lang.lisp
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!ix.netcom.com!netcom.com!vsocci
From: vsocci@netcom.com (Vance Socci)
Subject: Re: Retro-Computing!
Message-ID: <vsocciD6KsxB.Fyz@netcom.com>
Sender: vsocci@netcom17.netcom.com
Organization: Vcc Technical Services
X-Newsreader: News Xpress Version 1.0 Beta #2.1
References: <D5yxwn.5BG@sdf.saomai.org> <aldersonD67wtL.63n@netcom.com>
	<1995Mar29.231246.13715@ivax>
	<Pine.NXT.3.92.950330161611.2916C-100000@Tomobiki-Cho.CAC.Washington.EDU>
	<MURPHY.95Apr4220807@world.std.com>
	<Pine.NXT.3.92.950404215056.1339G-100000@Tomobiki-Cho.CAC.Washington.EDU> <DLM.95Apr5115216@outpost.osf.org>
Date: Wed, 5 Apr 1995 16:06:37 GMT
Lines: 91

dlm@outpost.osf.org (Dan Murphy) wrote:
>In article <Pine.NXT.3.92.950404215056.1339G-100000@Tomobiki-Cho.CAC.Washington.EDU> Mark Crispin <MRC@Panda.COM> writes:
>
>>    Dan, as I remember, everyone in the "macho programmer" discussion was a
>>    TOPS-20 hacker (with the possible exception of C. Dunn)...  ;-)
>
>    Indeed.  I remember well all those you mentioned.  I thought it
>    was amusing that a bunch of TOPS20 folks would be carrying on
>    (facetiously of course) about the macho programming ideal, when
>    TOPS20 took exactly the opposite view -- using *or* programming
>    the system should be as easy as possible, and programmers
>    shouldn't have to remember lots of obscure sh*t to prove their
>    manhood.
>
>    Many was the time I heard complaints that TOPS20 *wasted* precious
>    cpu cycles on frills like command recognition, and squandered
>    precious disk and memory space with symbol tables for debugging.
>    And why give the users multiple forks when they'll just fill up
>    even more memory!  And...  well, you get the idea.  The macho
>    attitude was very real, even if not to the extremes parodied in
>    the "macho programmer" discussion.
>

I've been on both sides of this . . . as a TOPS10 developer, and later at Tymshare
when we created over time a third PDP-10 operating system very similar in
its basic architecture to TOPS-20, I always felt the pressure from the users/customers
to bum instructions and write drum-tight code, conserve system resources, etc.

I think the macho attitude comes from having to keep systems running with what in retrospect
look ridiculously under-equipped. What about those guys who only have 48K of core
and still want the next release of their monitor to work without buying any more?

I pissed more than one dreamer off at Tymshare by stopping what would now be
considered modern advances in user interfacing for the sake of the business. We were
constantly getting razzed from the systems planning guys about not making them
spend any more money for memory or extra systems. They were constantly loading
systems up to the max, trying to make maximum profit. Even a change like adding
a non-resident command parser could tip the balance in a situation like that and
make them buy one or more new systems.

I found myself becoming very, very conservative and not very imaginative about adding
new things to the monitor. In fact, I was downright paranoid. That's what I got
for working in the cut-throat commercial side of things.

I think things were a little better in the scientific/academia world than with the
commercial users. They seemed to be willing to accept new features and pay the price
in terms of extra hardware (mostly memory). Their systems were probably not loaded
with users as much as the commercial guys loaded theirs. It was a much better
environment to advance the state of the art.

I think TENEX grew up in a friendlier environment than its gnarlier ancestor, TOPS-10.
Just as people's personalities are shaped in a large part by their environment, I think
operating systems are also shaped in that way. A lot of people refused to switch
over to TOPS-20 from TOPS-10 because of issues like the ones I described above.
In many ways, TENEX and TOPS-20 were ahead of their time - they represented
significant advances over TOPS-10. The extra overhead in providing a friendlier
environment would have eventually become insignificant had DEC let the product
line continue into the future and produced more modern hardware.

With the advent of the TD-1 at XKL, you'll finally get to see TOPS-20 in an environment
where people won't have to worry about price/performance degredations due to
the extra resources. There is so much memory, and so much disk space, that most people
probably won't understand what all the fuss was about back in 1977.

And now, with all the major workstation companies dropping the old standby 68000
and x86 series micros in favor of RISC, is the PDP-10 really that much out of place anymore?
TOPS-20 has enough goodies built into it to look very much like Unix . . . who
knows what will happen now?

Back to my story:

After I left the PDP-10 world, the hardware explosion began to happen, and the huge
advances in price/performance ratios changed everything. As I began to use several
different kinds of systems, sometimes at once, I began to hate the plethora of command
switches and editors . . . having my PDP-10 culture torn away, I realized just how
much of my comfort had come from locking into one particular system environment.

Nowadays, I've dropped my old attitude and I go for convenience and technical
advancement.




- Vance

/=======================================\
|    Vance Socci   vsocci@netcom.com	|
| "The worst secrets are those we keep	|
|   from ourselves . . ."		|
| "I am not a number; I am a free man!	|
\=======================================/
