[lug] Running a mixed Python environment

[Jeffrey S. Haemer]

Jed,

You guys are hard.

I remember Bill Waite's telling me what a bad idea terminals and editors
were; cards kept you from being less-than-careful. Didn't convince me,
since I knew how much less-than careful *I* was. I'd already spent years
picking up dropped decks of Hollerith cards and poring over green-striped
printouts of listings and core dumps generated by my own mis-punched
punctuation.

I burned my SIGPLAN notices decades ago, but I bet
the formatting/control-flow mismatches were more complex versions of this.

if foo {
    bar
}
    if mumble {
        frabitz
    }


Done that, too.

And passing tests is easy. I've written all sorts of wrong code that passed
wrong tests.

You'll now hasten to point out that I'm living proof you and Bill were
right. :-)

On Wed, Sep 16, 2020 at 9:03 AM Jed S. Baer <blug at jbaer.cotse.net> wrote:

> On Wed, 16 Sep 2020 07:58:19 -0600
> Jeffrey S. Haemer wrote:
>
> > The programmers (as we used to call ourselves) in the other ISC offices
> > all hated it because of that.
>
> Yeah, I was a "programmer" once. Back then, we worked in "data processing".
> I recall one Java developer who took offense to my use of that term, and
> said, quite pointedly, "I do NOT process data!"
>
> > I still think they were, arguably, wrong. Some time around 1982, I read a
> > study in SIGPLAN notices, a now-defunct pub of the ACM, by someone who'd
> > hunted through huge bodies of code (FORTRAN? PL/1? PASCAL? I forget.) for
> > cases where indentation didn't match control-flow.  In almost every case,
> > the indentation was right, the bracing/keywords/whatever, wrong.
> >
> > Humans are better at indenting than bracing.
>
> I could come at that the other way, and argue that it allows
> less-than-careful programmers to write working code, but then I'm biased,
> since I always made it a point to have indentation and block delimiters
> match up. This harkens back to the old days, when "online" code simply
> matched how code would come out on Hollerith cards. Eventually, we all got
> away from that, but in the mean time, editors and programmers handled tabs
> and spaces differently. I went through a phase where I used tabs
> exclusively, and had "set tabwidth=4" (this was a display setting) in my
> editor startup file. I think that was in EDT. In maintaining others' code,
> I
> encountered a mix of usage, even mixed tabs and spaces for indentation.
> Eventually, I just got in the habit of replacing all tabs with 4 spaces,
> and then fixing the whole thing. So, I ended up setting automatic tab
> expansion (i.e. substitution) in all my editing environments.
>
> The result of this is that when I first encountered Python, I wondered what
> level of indentation (and therefore logic) resulted from tabs vs. spaces,
> particularly if there is a mix in the leading whitespace.
>
> And I still consider the use of block delimiters to be more precise and
> robust.
>
> And I wonder how code with incorrect block delimiting got through compile
> and test?
>
> --
> All operating systems suck, but Linux just sucks less
>  - Linus Torvalds
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> Join us on IRC: irc.hackingsociety.org port=6667 channel=#hackingsociety



-- 
Jeffrey Haemer <jeffrey.haemer at gmail.com>   720-837-8908 [cell]
*פרייהייט? דאס איז יאַנג דינען וואָרט!*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lug.boulder.co.us/pipermail/lug/attachments/20200916/ec96c9c4/attachment.html>