The Unix Heritage Society mailing list
 help / color / mirror / Atom feed
From: Tomasz Rola <rtomek@ceti.pl>
To: tuhs <tuhs@tuhs.org>
Subject: Re: [TUHS] What was your "Aha, Unix!" moment?
Date: Fri, 11 Oct 2019 23:56:20 +0200	[thread overview]
Message-ID: <20191011215620.GB12901@tau1.ceti.pl> (raw)
In-Reply-To: <20191010205546.GA29154@minnie.tuhs.org>

On Fri, Oct 11, 2019 at 06:55:46AM +1000, Warren Toomey wrote:
> All, we had another dozen TUHS suscribers to the list overnight. Welcome.
> A reminder that we're here to discuss Unix Heritage, so I'll nudge you
> if the conversation goes a bit off-topic.

I am from this previous dozen. The current one, I guess, might have
come here with help of outlets like Lobsters and HN.

> So I'll kick off another thread. What was your "ahah" moment when you
> first saw that Unix was special, especially compared to the systems you'd
> previously used?

Oh, that is a good question.

I think I had a few aha's which are now faded, but the one I remember
well was when I wrote a program which was to be described in a
document, and it generated lots of data (20mb, which for me, in
mid-1990-ies, was a lot - today, one of my org files is bigger and I
keep adding to it every day). There was no way I would want to process
this data by hand, and reading it into Excel was a no-no either,
because typical Windows machine had 4-8 megs of ram and I did not
think they would be able to handle this task. The files compressed
well (like, 20:1). So I wrote some shell scripts and (n)awk scripts,
find-ed compressed files, feeded uncompressed data via pipe and awk
made nice tables for LaTeX and data for gnuplot. On a machine that had
4mb of ram itself.

So, that was a moment - I could use Unix and the tools to make things
requiring much bigger machines (if they were not running Unix).

Later on, knowing some nuances of Unix helped me a lot, even if those
were really small tricks. Like, using 'cp -l' to copy huge source tree
to user with inadequate quota and then compile it. Later on, I bought
me a crappy modem without hardware error control, and it was unusable
when connection was made from Windows/DOS (time counted in seconds,
then hangup because of line errors). Again, Unix (Linux, actually) to
the rescue - after I learned to use ppp, hangups became extremely
rare, even if many times connection was despetately slow. Still, slow
was better than none.

> Mine was: Oh, I can:
>   + write a simple script
>   + to edit a file on the fly
>   + with no temporary files (a la pipes)
>   + AND I can change the file suffix and the system won't stop me!
> 
> I was using TOPS-20 beforehand.

Before I started to use university SunOS, I was using university's VAX
for a year and Amiga at home. On Amiga, I sometimes played and
sometimes played with a system. So I already had some expectations
about what a computer should be like (at least, the "my
computer"). Before that, I spent few years playing on and off with
other people's 8-bit computers (with their knowledge, of course) -
Atari (800?), Polish-made Meritum-1 (docs said it was CP/M capable,
but at that time, I had hard time getting more info on this and
programing simple stuff in Basic was low hanging fruit).

While Amiga gave me many early pleasures of multitasking (playing
Centurion and doing low-priority fractal in the background, just for
the kicks), and I kept her up to 1994, SunOS quickly dwarfed
her. Around 1994 I got introduced to the Linux concept by a bud.

Anyway, I installed Slackware on 486 and was able to configure olvwm
on it, thus having almost same look as uni-SunOS. Which helped with
adoption. After that, fvwm. After some time with KDE and Gnome, back
to fvwm, because it just works.

There is still plenty to learn. I feel like I barely licked the
surface. My recent memorable aha was when I wrote a script in Elisp
rather than in bash, as usually. Elisp is hardly a scripting language,
it just copes in some cases, but for some reason complicated flow
control looked much better than I expected.

Another aha was after reading that running stuff in pipes may
outperform some well know parallel computing frameworks. Yes indeed,
each part of multipipe runs on different cpu and OS takes care of
making the flows as fluent as it can. And the concept of pipe is,
what, more than fourty years old? Almost fifty? This quite an aha, I
think.

-- 
Regards,
Tomasz Rola

--
** A C programmer asked whether computer had Buddha's nature.      **
** As the answer, master did "rm -rif" on the programmer's home    **
** directory. And then the C programmer became enlightened...      **
**                                                                 **
** Tomasz Rola          mailto:tomasz_rola@bigfoot.com             **

  parent reply	other threads:[~2019-10-11 21:56 UTC|newest]

Thread overview: 122+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-10 20:55 Warren Toomey
2019-10-10 21:11 ` SPC
2019-10-10 21:25   ` Jim Capp
2019-10-10 21:33     ` greg travis
2019-10-10 21:13 ` Michael Parson
2019-10-10 21:15 ` Matt Rudge
2019-10-10 21:31 ` Steffen Nurpmeso
2019-10-10 21:34 ` Larry McVoy
2019-10-10 23:10 ` Seth J. Morabito
2019-10-10 23:23 ` Brian L. Stuart
2019-10-10 23:28   ` George Michaelson
2019-10-10 23:35 ` Bakul Shah
2019-10-10 23:49   ` David
2019-10-11  1:44 ` Gregg Levine
2019-10-11  1:45   ` Gregg Levine
2019-10-11 10:55 ` Leah Neukirchen
2019-10-11 12:04   ` Tyler Adams
2019-10-13 19:46     ` Peter Jeremy
2019-10-14  2:13       ` Lawrence Stewart
2019-10-14  2:32         ` Rico Pajarola
2019-10-14  9:49           ` Michael Kjörling
2019-10-14 18:36             ` Nemo Nusquam
2019-10-14 21:10               ` Jim Geist
2019-10-14 22:22                 ` Warren Toomey
2019-10-14 22:56                 ` Arthur Krewat
2019-10-14 23:44                   ` Jim Geist
2019-10-14 23:47                     ` Adam Thornton
2019-10-14 23:54                       ` Arthur Krewat
2019-10-15  0:03                         ` Henry Bent
2019-10-14 23:54                       ` Ronald Natalie
2019-10-15  0:04                         ` Arthur Krewat
2019-10-15  0:06                           ` Ronald Natalie
2019-10-15  0:27                             ` Arthur Krewat
2019-10-15  0:07                         ` George Michaelson
2019-10-15  0:10                           ` Larry McVoy
2019-10-15  0:27                           ` [TUHS] Supercomputer UNIX (was Aga moments) Ronald Natalie
2019-10-15  0:33                             ` [TUHS] Supercomputer UNIX (was Aha moments) Ronald Natalie
2019-10-14 23:52                     ` [TUHS] What was your "Aha, Unix!" moment? Arthur Krewat
2019-10-11 12:53 ` KatolaZ
2019-10-11 16:44 ` Pete Wright
2019-10-11 17:13 ` Jim Geist
2019-10-11 17:20   ` Larry McVoy
2019-10-11 17:40   ` Jim Capp
2019-10-11 17:48 ` Larry W. Cashdollar via TUHS
2019-10-11 21:56 ` Tomasz Rola [this message]
2019-10-12  2:41 ` ricercar
2019-10-12  3:01   ` Larry McVoy
2019-10-12 14:37     ` Theodore Y. Ts'o
2019-10-12 16:12     ` David
2019-10-12 17:59       ` Adam Thornton
2019-10-12 19:10         ` Jon Forrest
2019-10-13 21:45           ` Steve Johnson
2019-10-14  0:36             ` Jon Forrest
2019-10-14  2:08               ` Lawrence Stewart
2019-10-16 13:29             ` Pierre DAVID
2019-10-12  3:23   ` Richard Salz
2019-10-12  3:34     ` Larry McVoy
2019-10-12 21:32     ` Steffen Nurpmeso
2019-10-12  8:55 ` Wesley Parish
2019-10-12 10:17 ` Naveen Nathan
2019-10-12 11:54 ` markus schnalke
2019-10-12 19:33 ` Michael Kjörling
2019-10-12 22:38   ` Gregg Levine
2019-10-13  1:37 ` Dan Cross
2019-10-13 15:00 ` Robert Brockway
2019-10-13 15:33 ` arnold
2019-10-13 15:41   ` Larry McVoy
2019-10-13 15:47     ` David Potesta
2019-10-18  1:49 ` Dave Horsfall
2019-10-18 12:07   ` Ronald Natalie
2019-10-21 16:16 ` Dario Niedermann
2019-10-23  5:13 ` Gilles Gravier
2019-10-23  6:19   ` Adam Thornton
2019-10-23 15:08   ` Arthur Krewat
2019-10-23 15:11     ` Larry McVoy
2019-10-23 15:26       ` Arrigo Triulzi
2019-10-23 15:33         ` Dan Cross
2019-10-23 16:19           ` Arthur Krewat
2019-10-23 15:17     ` Richard Salz
2019-10-23 15:22       ` Arrigo Triulzi
2019-10-23 16:45     ` Will Senn
2019-10-23 22:19       ` Adam Thornton
2019-10-24 17:32       ` Tom Ivar Helbekkmo via TUHS
2019-10-26  0:33     ` Dave Horsfall
2019-10-25 20:58 ` John S Quarterman
2019-10-25 22:11   ` Will Senn
2019-10-26  0:57 ` William Corcoran
2019-10-12  4:42 Doug McIlroy
2019-10-12  6:12 ` Adam Thornton
2019-10-12 10:05 ` Steve Nickolas
2019-10-12 12:49   ` Doug McIlroy
2019-10-12 14:26     ` Steve Nickolas
2019-10-12 13:55 Noel Chiappa
2019-10-12 14:51 ` Ronald Natalie
2019-10-13  1:57 Jon Steinhart
2019-10-13  3:45 Rudi Blom
2019-10-13 16:07 Noel Chiappa
2019-10-13 16:25 ` Richard Salz
2019-10-14 15:32 Doug McIlroy
2019-10-15  8:42 ` Thomas Paulsen
2019-10-14 16:45 Noel Chiappa
2019-10-15  0:28 Pat Barron
2019-10-15  1:19 ` Andrew Warkentin
2019-10-15  4:06   ` Warner Losh
2019-10-16 14:39 Doug McIlroy
2019-10-20 19:25 Norman Wilson
2019-10-20 20:12 ` Sean Dwyer
2019-10-21  2:31   ` Ken Thompson via TUHS
2019-10-21  2:37     ` Warren Toomey
2019-10-22  5:25       ` Peter Jeremy
2019-10-22  6:29       ` Sean Dwyer
2019-10-21  2:40     ` Bakul Shah
2019-10-21  2:45     ` Larry McVoy
2019-10-21 10:45       ` jason-tuhs
2019-10-21 11:55         ` William Corcoran
2019-10-22  5:19         ` Adam Thornton
2019-10-21 12:10 Noel Chiappa
2019-10-21 16:50 ` Angelo Papenhoff
2019-10-21 12:34 Noel Chiappa
2019-10-22 13:36 Noel Chiappa
2019-10-22 14:22 ` Abhinav Rajagopalan
2019-10-22 20:08   ` Warren Toomey

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20191011215620.GB12901@tau1.ceti.pl \
    --to=rtomek@ceti.pl \
    --cc=tuhs@tuhs.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).