Re: [PLUG] Challenges with "The Unix Way"

[Steve Litt <slitt@troubleshooters.com>]

On Fri, 27 Oct 2017 01:26:23 -0400
brent timothy saner <brent.saner@gmail.com> wrote:

> On 10/27/2017 01:15 AM, ac wrote:
> > On Thu, 26 Oct 2017 14:37:01 -0400
> > Rich Freeman <r-plug@thefreemanclan.net> wrote:  
> >> I thought that this article was a really interest contrast with
> >> "The Unix Way" and some of the ways it is mis-applied:
> >> http://www.codersnotes.com/notes/something-rotten-in-the-core/
> >> (The point isn't that the Unix Way is necessarily a bad thing, but
> >> just that not every problem is best solved with a maze of nested
> >> pipes and some shell scripting to glue it together.  And by Unix
> >> Way I'm referring specifically to pipes and such and not to the
> >> more general principle of doing one thing well.)
> >>  
> > bleh. ymmv but, one can also build in error reporting when using
> > ssh, etc. 
> > 
> > good design applies to everything in life. 
> > 
> > some systems that i have built as described in the linked post have
> > an engine running in the back (sometimes even an engine behind the
> > engine of the engine) and the 'public' stuff are all semi
> > 'fragile' (but so very easy to replicate / restore - as there
> > exists redundancy,error reporting and proper documentation :) )
> > 
> > when i started dev on ai/machine learning in 2008, my first algo
> > was a simple loop which created executable code loops, i note that
> > the same simplicity is what the trend of the past few years has
> > become - my point with that is that sometimes something very simple
> > is exactly that, simple. it is the further design (or lack) that
> > causes debugging stress :)
> > 
> > 2c  
> 
> if it makes y'all feel better, the "UNIX Way" is a myth:
> 
> http://web.mit.edu/~simsong/www/ugh.pdf

Puh-lease! That silly article was bad enough in the 1990's, but now,
with 22 years of pure Unix and another 26 very successful years of Linux
doing most things the Posix, if not the Linux way, you can see how
wrong they were. What next, trot out Andy Tannenbaum to decry the
monolithic kernel?

As far as the link to codernotes.com, as far as I can see the guy was
bitching about the debugger and somehow trying to smear that on to all
of Unix.

> 
> at least on the following points:
> -it is not always (or, perhaps, even often) the "best" way
> (fastest/most efficient, easiest to parse, etc.), 

Yeah, but a heck of a lot of the time it is the best, because grep, sed,
awk, tr, cut, cat and sort have been *highly* optimized and completely
debugged. Also, when those little executables can't efficiently solve
the problem, a lot of time throwing a couple custom programs in the
pipeline or shellscript can make things perfect. The Unix hierarchical
filesystem is a nice way to keep hierarchical config data, without the
need for parsers of XML and JSON and Yaml. And if you need GUI but you
don't need pretty, there are usually very simple ways to have GUI work
with shellscripts.

You're right: It's not always the best. But when it is, for gosh sakes
use it, because it's trivially easy and it was fully debugged decades
ago.


> especially in
> modern context -UNIX itself frequently strayed from the so-called
> "UNIX Way"

You wouldn't by any chance be talking about Nautilus, Gnome3, KDE,
and Firefox, would you? You can't blame the Unix way for those: They
simply abandoned every shred of the Unix way.

SteveT

Steve Litt 
October 2017 featured book: Rapid Learning for the 21st Century
http://www.troubleshooters.com/rl21
___________________________________________________________________________
Philadelphia Linux Users Group         --        http://www.phillylinux.org
Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug