On Sat, 06 Aug 2022 17:39:10 +0100, Ibmekon <Ibmekon@google.com>
wrote:
>On Fri, 05 Aug 2022 13:13:34 -0400, Dan Espen <dan1espen@gmail.com>
>wrote:
>
>>Charlie Gibbs <cgibbs@kltpzyxm.invalid> writes:
>>
>>> On 2022-08-05, Scott Lurndal <scott@slp53.sl.home> wrote:
>>>
>>>> John Levine <johnl@taugh.com> writes:
>>>>
>>>>> According to 25B.Z969 <25B.Z969@noda.net>:
>>>>>
>>>>>>> There are too many things you need digital for, from anti-lock brakes to
>>>>>>> engine control.
>>>>>>
>>>>>> I drove cars for decades that didn't have those "luxury
>>>>>> features" ...... they got you from A to B just fine,
>>>>>> they WORKED, they could be MADE, SOLD and BOUGHT.
>>>>>
>>>>> So did I. They got 19 MPG and fell apart after 75,000 miles. No thanks.
>>>>
>>>> You were lucky to get 19MPG, 8 to 10mpg was common until
>>>> the late 70s.
>>>>
>>>> And you needed new tires every 10k miles (pre radial) and they
>>>> burned oil like locomotives. Frequent tuneups, poor cabin
>>>> acoustics, rough rides, thrown rods.
>>>
>>> You might be exaggerating a bit. My 1970 Suburban got 15 MPG
>>> with its 350-cubic-inch engine and 4-barrel carburetor. Mind
>>> you, those were Imperial gallons. And I figured you could get
>>> 30k miles from a set of tires - but I much prefer radials.
>>> (At one point I found myself with a mix of radial and bias-ply
>>> tires. It did _not_ handle well.)
>>>
>>>> I'll take a modern car anyday.
>>>
>>> So will I - they sure start more easily on a cold day.
>>> Or any day, for that matter. I might change my mind
>>> with the current genration, though - at least until
>>> someone hacks the onboard computers and disconnects
>>> the surveillance equipment.
>>
>>I don't have anything that modern but stopping the surveillance should
>>be as simple as failing to pay the monthly fee. (Leave your cell phone
>>at home if you really care.)
>>
>>My car is not new enough for the backup camera but that seems like a
>>worthwhile feature. The anti-lock brakes are really good and I really
>>like not needing a tune up every 10K miles.
>All new cars since 2019 sold in the EU have an emergency rescue
>transmitter.
>I read the technical spec which went out to tender.
>It is a computer and phone - and can be updated remotely without the
>owner having any control or knowledge.
>They are now bringing in legislation to automatically stop you
>exceeding speed limits.

If I were to see that in the US, I would expect a change where a
politicians' private vehicles were not subject to that.
--
Jim

On Sat, 06 Aug 2022 00:30:07 +0100, Ibmekon <Ibmekon@google.com>
wrote:
>On Fri, 05 Aug 2022 17:51:28 -0500, D.J. <chucktheouch@gmnol.com>
>wrote:
>
>>On Fri, 05 Aug 2022 23:35:43 +0100, Ibmekon <Ibmekon@google.com>
>>wrote:
>>>On Fri, 05 Aug 2022 21:40:43 GMT, Charlie Gibbs
>>><cgibbs@kltpzyxm.invalid> wrote:
>>>
>>>>On 2022-08-05, Dan Espen <dan1espen@gmail.com> wrote:
>>>>
>>>>> Charlie Gibbs <cgibbs@kltpzyxm.invalid> writes:
>>>>>
>>>>>> On 2022-08-05, Scott Lurndal <scott@slp53.sl.home> wrote:
>>>>>>
>>>>>>> I'll take a modern car anyday.
>>>>>>
>>>>>> So will I - they sure start more easily on a cold day.
>>>>>> Or any day, for that matter. I might change my mind
>>>>>> with the current genration, though - at least until
>>>>>> someone hacks the onboard computers and disconnects
>>>>>> the surveillance equipment.
>>>>>
>>>>> I don't have anything that modern but stopping the surveillance should
>>>>> be as simple as failing to pay the monthly fee. (Leave your cell phone
>>>>> at home if you really care.)
>>>>
>>>>Of course, if you stop paying the monthly fee, the car might refuse
>>>>to start. Maybe not this generation, but probably the next one.
>>>>
>>>>I'm not too worried about cell phones; I'll give up my flip phone
>>>>when they pry it from my cold dead fingers.
>>>>
>>>>About 10 years ago I read a science fiction story about someone driving
>>>>along when an ad came up on his radio^Winfotainment centre. The car
>>>>then pulled over to the side of the road, locked the doors, and refused
>>>>to let him out until he bought whatever they were selling. Fortunately
>>>>a friend came along and got him out - but it's coming soon, I'm sure.
>>>>
>>>>Recommended reading: _Unauthorized Bread_ by Cory Doctorow.
>>>
>>>I got a demented Nissan NV200 van.
>>>
>>>Early on in my driving it went crazy - six lights starting flashing on
>>>the display screen - including "dangerous engine fault, pull over
>>>immediately and get it towed."
>>>So, I pulled over , switched it off and on again.
>>>Maybe one light less - but the engine started, so I tried to drive.
>>>But it seemed limited in revs, speed to about 50 kph - it has a manual
>>>gearbox.
>>>So I guessed it was a computer fault - not mechanical.
>>>I put on the hazard lights and crawled home.
>>>Checking on the net I found that there is a push button at knee height
>>>that switches in/out ABS traction control.
>>>And if you dare to go manual - the van goes apeshit.
>>>
>>>The Nissan equivalent of CTR+ALT+DELETE is this -
>>>--------------------------------------------------------------------
>>>Open the driver's side door and get into the drivers seat.
>>>
>>>Insert the key into the ignition and turn it a quarter turn so the
>>>ignition turns on, but the engine doesn't start.
>>>
>>>Step on the accelerator and quickly release it.
>>>Do this 5 times in less than 5 seconds.
>>>When you step on the pedal, press it down as far as it can go.
>>>
>>>Wait 10 seconds.
>>>
>>>Press the accelerator completely down with your foot and hold it there
>>>for 10 seconds, or until the SES light blinks on and off.
>>>--------------------------------------------------------------------
>>>
>>>This is what you get with kids trained on Windoze.
>>
>>Yeah, that sounds dumb for the car makers to do that.
>>--
>>Jim
>
>Not quite the words I used.
>"dumb" I do not mind.
>But this was "accidentally on purpose" management sabotage to force my
>compliance illegally.

In this instance, 'dumb' and 'accidentally on purpose' could be used
interchangeably.

--
Jim

On 2022-08-06, 25B.Z969 <25B.Z969@noda.net> wrote:

> After some searching I couldn't really find much on
> the structural issues with "cone-shaped" buildings.
> Very annoying. I visualize something like an incense
> cone - with a 100-meter base, 1000 meters tall.
> I don't think it'd suffer much from sway or resonance
> issues. Glass dome for the penthouse suite. Oh the RENT
> you could charge :-)
>
> Might need a few
>
> \/
> --
> /\
>
> anti-buckling braces up the inner core ....

The Spire of Dubliln is a miniature version:
120 meters high, 3m diameter at the base.

--
/~\ Charlie Gibbs | Microsoft is a dictatorship.
\ / <cgibbs@kltpzyxm.invalid> | Apple is a cult.
X I'm really at ac.dekanfrus | Linux is anarchy.
/ \ if you read it the right way. | Pick your poison.

On 8/7/22 6:36 PM, Charlie Gibbs wrote:
> On 2022-08-06, 25B.Z969 <25B.Z969@noda.net> wrote:
>
>> After some searching I couldn't really find much on
>> the structural issues with "cone-shaped" buildings.
>> Very annoying. I visualize something like an incense
>> cone - with a 100-meter base, 1000 meters tall.
>> I don't think it'd suffer much from sway or resonance
>> issues. Glass dome for the penthouse suite. Oh the RENT
>> you could charge :-)
>>
>> Might need a few
>>
>> \/
>> --
>> /\
>>
>> anti-buckling braces up the inner core ....
>
> The Spire of Dubliln is a miniature version:
> 120 meters high, 3m diameter at the base.

Much greater height/base diameter ratio than I had in mind,
40:1 vs 10:1 ... I don't think that would be safe for a
very tall structure. But it IS the same idea.

Boxes/rectangles/tubes are easy to build but they all
suffer from a lot of sway issues if you build them tall.
There's also potential resonance issues in case of an
earthquake (or the rebirth of Tesla The Destroyer).

Roughly 10:1 cones would seem most immune to these
problems. Besides, the hollow interior could be just
spectacular - maybe better than an external view. Build
'em like they build big ships - modular sections you
just kinda slide into place and weld together.

Hey ... a smaller cone INSIDE the big cone ... shops,
eateries, auditoriums.

Of course everything is run on Debian, maybe BSD :-)

On 2022-08-07, Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
> On 2022-08-06, 25B.Z969 <25B.Z969@noda.net> wrote:
>
>> After some searching I couldn't really find much on
>> the structural issues with "cone-shaped" buildings.
>> Very annoying. I visualize something like an incense
>> cone - with a 100-meter base, 1000 meters tall.
>> I don't think it'd suffer much from sway or resonance
>> issues. Glass dome for the penthouse suite. Oh the RENT
>> you could charge :-)
>>
>> Might need a few
>>
>> \/
>> --
>> /\
>>
>> anti-buckling braces up the inner core ....
>
> The Spire of Dubliln is a miniature version:
> 120 meters high, 3m diameter at the base.
>

And utterly pointless. Lots of people walk along that Street and never
notice it.

--
greymausg@mail.org
Fi Fi Fo Fum, I smell the stench of an influencer
ten, twenty million tops

On Wed, 3 Aug 2022 01:05:13 -0400, "25B.Z969" <25B.Z969@noda.net>
wrote:

>On 8/2/22 1:05 PM, John Levine wrote:
>> According to Kerr-Mudd, John <admin@127.0.0.1>:
>>> On Mon, 1 Aug 2022 20:37:43 +0100
>>> Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>>>
>>>> On Mon, 1 Aug 2022 12:22:38 -0700
>>>> Peter Flass <peter_flass@yahoo.com> wrote:
>>>>
>>>>> If engineers built bridges the way programmers write programs, the first
>>>>> stiff breeze that came along would destroy civilization.
>>>>
>>>> Tacoma Narrows 1940 - nuff said ?
>>>>
>>> Beta-test first. And always keep a backup.
>>
>> Oddly, there was a backup for Galloping Gertie. The Bronx-Whitestone
>> Bridge in New York City which opened in 1939 used a similar design.
>> When the Tacoma Narrows bridge collapsed, and although it was shorter
>> and wider and so less likely to oscillate, they addded stiffening
>> trusses anyway after the war. It's been widened and refurblshed in
>> ways that let them remove the trusses in the mid 2000s. Now it's part
>> of I-678 and is used by over 100,000 vehicles per day.
>
> "Gertie" was a real (and expensive) textbook lesson
> in resonance. It was basically a big long guitar
> string whipping in the wind. Tesla, long before,
> fully understood structural resonance issues - even
> built a little "shaker" device he (claimed) could
> collapse large buildings and bridges. I have doubts,
> but the *principle* was sound.

I'm flashing on a rather small, rather slender female accountant of my
acquaintance who whenever she walked down the hall in the building in
which I was working shook the whole building. Seems her walking pace
came pretty close to hitting the resonant frequency of the flooring.

On Fri, 5 Aug 2022 21:45:42 -0400, "25B.Z969" <25B.Z969@noda.net>
wrote:

>On 8/5/22 10:05 AM, Scott Lurndal wrote:
>> John Levine <johnl@taugh.com> writes:
>>> According to 25B.Z969 <25B.Z969@noda.net>:
>>>>> There are too many things you need digital for, from anti-lock brakes to
>>>>> engine control.
>>>>
>>>> I drove cars for decades that didn't have those "luxury
>>>> features" ...... they got you from A to B just fine,
>>>> they WORKED, they could be MADE, SOLD and BOUGHT.
>>>
>>> So did I. They got 19 MPG and fell apart after 75,000 miles. No thanks.
>>
>> You were lucky to get 19MPG, 8 to 10mpg was common until
>> the late 70s.
>>
>> And you needed new tires every 10k miles (pre radial) and they
>> burned oil like locomotives. Frequent tuneups, poor cabin
>> acoustics, rough rides, thrown rods.
>>
>> I'll take a modern car anyday.
>
> But again, NONE of those issues had ANYTHING to do
> with whether they were 'digital' or not. It was just
> loosey-goosey manufacturing and Babbit bearings.

Actually it's all due to government regulation. The government has
applied safety, emission, and fuel economy standards. To comply with
all three requires a high degree of complexity, and the emission
standards include a durability requirement. Electronic engine
controls are part of that package.

On Fri, 5 Aug 2022 21:44:00 -0400, "25B.Z969" <25B.Z969@noda.net>
wrote:

>On 8/4/22 11:33 PM, John Levine wrote:
>> According to 25B.Z969 <25B.Z969@noda.net>:
>>>> There are too many things you need digital for, from anti-lock brakes to
>>>> engine control.
>>>
>>> I drove cars for decades that didn't have those "luxury
>>> features" ...... they got you from A to B just fine,
>>> they WORKED, they could be MADE, SOLD and BOUGHT.
>>
>> So did I. They got 19 MPG and fell apart after 75,000 miles. No thanks.
>
> Metallurgy & machining tech, and they weren't trying very
> hard anyway. The electronics had nothing to do with it.
> An all-digital '57 Chevy would still get 19mpg and have
> to be rebuilt after 70,000 miles.

You might be surprised at the mileage it could get with modern
electronic fuel injection and no emission controls.

On Sat, 13 Aug 2022 14:53:04 -0400, J. Clarke wrote:

> On Wed, 3 Aug 2022 01:05:13 -0400, "25B.Z969" <25B.Z969@noda.net> wrote:
>
>>On 8/2/22 1:05 PM, John Levine wrote:
>>> According to Kerr-Mudd, John <admin@127.0.0.1>:
>>>> On Mon, 1 Aug 2022 20:37:43 +0100 Ahem A Rivet's Shot
>>>> <steveo@eircom.net> wrote:
>>>>
>>>>> On Mon, 1 Aug 2022 12:22:38 -0700 Peter Flass
>>>>> <peter_flass@yahoo.com> wrote:
>>>>>
>>>>>> If engineers built bridges the way programmers write programs, the
>>>>>> first stiff breeze that came along would destroy civilization.
>>>>>
>>>>> Tacoma Narrows 1940 - nuff said ?
>>>>>
>>>> Beta-test first. And always keep a backup.
>>>
>>> Oddly, there was a backup for Galloping Gertie. The Bronx-Whitestone
>>> Bridge in New York City which opened in 1939 used a similar design.
>>> When the Tacoma Narrows bridge collapsed, and although it was shorter
>>> and wider and so less likely to oscillate, they addded stiffening
>>> trusses anyway after the war. It's been widened and refurblshed in
>>> ways that let them remove the trusses in the mid 2000s. Now it's part
>>> of I-678 and is used by over 100,000 vehicles per day.
>>
>> "Gertie" was a real (and expensive) textbook lesson in resonance. It
>> was basically a big long guitar string whipping in the wind. Tesla,
>> long before, fully understood structural resonance issues - even
>> built a little "shaker" device he (claimed) could collapse large
>> buildings and bridges. I have doubts,
>> but the *principle* was sound.
>
> I'm flashing on a rather small, rather slender female accountant of my
> acquaintance who whenever she walked down the hall in the building in
> which I was working shook the whole building. Seems her walking pace
> came pretty close to hitting the resonant frequency of the flooring.

I worked, for a while, in a 50+ floor office tower that had a gym near
the top floor. At lunch hour, you could feel the tower sway, even on
the near-ground-level floors, as the various aerobics and other gym
activity got underway. Not quite a resonant frequency, but close
enough.

--
Lew Pitcher
"In Skills, We Trust"

On Sat, 13 Aug 2022 15:44:49 -0400, Lew Pitcher <lew.pitcher@digitalfreehold.ca> wrote:
> I worked, for a while, in a 50+ floor office tower that had a gym near
> the top floor. At lunch hour, you could feel the tower sway, even on
> the near-ground-level floors, as the various aerobics and other gym
> activity got underway. Not quite a resonant frequency, but close
> enough.

Sounds like the ibm tower of the td center.

Regards, Dave Hodgins

On Sat, 13 Aug 2022 16:51:04 -0400, David W. Hodgins wrote:

> On Sat, 13 Aug 2022 15:44:49 -0400, Lew Pitcher
> <lew.pitcher@digitalfreehold.ca> wrote:
>> I worked, for a while, in a 50+ floor office tower that had a gym near
>> the top floor. At lunch hour, you could feel the tower sway, even on
>> the near-ground-level floors, as the various aerobics and other gym
>> activity got underway. Not quite a resonant frequency, but close
>> enough.
>
> Sounds like the ibm tower of the td center.
>
> Regards, Dave Hodgins

Spot on.

--
Lew Pitcher
"In Skills, We Trust"

On 2022-07-24 1:31 a.m., 25B.Z959 wrote:
>   Ah ... PL/I ... one of the great "and the kitchen sink too"
>   languages. Not awful - usually several ways to accomplish
>   the same thing ... ie "flexibility". A bit odd though, like
>   a mutant hybrid of Algol and BASIC

PL/I looked like a merge of Fortran, Cobol, and bit of Algol
that managed to produce the cross-product of the problems of
all three.

According to Alan Bowler <atbowler@thinkage.ca>:
>On 2022-07-24 1:31 a.m., 25B.Z959 wrote:
>>   Ah ... PL/I ... one of the great "and the kitchen sink too"
>>   languages. Not awful - usually several ways to accomplish
>>   the same thing ... ie "flexibility". A bit odd though, like
>>   a mutant hybrid of Algol and BASIC
>
>PL/I looked like a merge of Fortran, Cobol, and bit of Algol
>that managed to produce the cross-product of the problems of
>all three.

It didn't look like a merge of them, that's what it was, since its
goal was to be a language to replace Fortran and COBOL on IBM
mainframes.

The early PL/I compilers had their problems, such as terrible code
when you wrote stuff that wasn't what people typically did. I tried to
use an array of 12-bit strings to simulate a PDP-8's memory and
somehow the code to access it included conversion to decimal and back.

I believe that current PL/I compilers work quite well, and the language
has evolved somewhat to fix the rough edges, but outside of a few
old IBM shops, nobody cares.

--
Regards,
John Levine, johnl@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

John Levine <johnl@taugh.com> writes:
> It didn't look like a merge of them, that's what it was, since its
> goal was to be a language to replace Fortran and COBOL on IBM
> mainframes.

Some of the MIT/7094 CTSS people
https://en.wikipedia.org/wiki/Compatible_Time-Sharing_System
went to the 5th flr, project mac, and MULTICS.
https://en.wikipedia.org/wiki/Multics
.... folklore is some of the Bell Mutlics people then did UNIX as
simplified Multics

Others went to the 4th flr, IBM Cambridge Science Center, did virtual
machine CP40/CMS (on 360/40 with hardware mods for virtual memory,
morphs into CP67/CMS when 360/67 standard with virtual memory becomes
available, precursor to vm370), online and performance apps, CTSS RUNOFF
redid for CMS as SCRIPT, GML invented at science center in 1969 (and GML
tag processing added to SCRIPT, a decade later GML morphs into ISO SGML
and after another decade morphs into HTML at CERN), networking, etc.
https://en.wikipedia.org/wiki/CP/CMS
https://en.wikipedia.org/wiki/Conversational_Monitor_System
some more CP67/CMS and VM370/CMS history
http://www.leeandmelindavarian.com/Melinda#VMHist

5flr implemened Multics in PL/1
https://multicians.org/pl1.html
http://teampli.net/plisprg.html
https://multicians.org/pl1-raf.html

wikipedia
https://en.wikipedia.org/wiki/PL/I
multics pl/i and derivatives
https://en.wikipedia.org/wiki/PL/I#Multics_PL/I_and_derivatives

Compilers were implemented by several groups in the early 1960s. The
Multics project at MIT, one of the first to develop an operating system
in a high-level language, used Early PL/I (EPL), a subset dialect of
PL/I, as their implementation language in 1964. EPL was developed at
Bell Labs and MIT by Douglas McIlroy, Robert Morris, and others. The
influential Multics PL/I compiler[27] was the source of compiler
technology used by a number of manufacturers and software groups. EPL
was a system programming language and a dialect of PL/I that had some
capabilities absent in the original PL/I.

PL/I goals and principles
https://en.wikipedia.org/wiki/PL/I#Goals_and_principles

other tivia: Jean Sammet (FORMAC and COBOL)
https://en.wikipedia.org/wiki/Jean_E._Sammet
was in the IBM Boston Programming Center on the 3rd flr ... after the
decision to make all 370s virtual memory and do VM/370, some of the
people split off from the science center (on the 4th flr) and took over
the Boston Programming Center on the 3rd flr (when they outgrew the 3rd
floor, they moved out to the empty IBM SBC bldg in burlington mall).

--
virtualization experience starting Jan1968, online at home since Mar1970

Anne & Lynn Wheeler <lynn@garlic.com> writes:
> was in the IBM Boston Programming Center on the 3rd flr ... after the
> decision to make all 370s virtual memory and do VM/370, some of the
> people split off from the science center (on the 4th flr) and took over
> the Boston Programming Center on the 3rd flr (when they outgrew the 3rd
> floor, they moved out to the empty IBM SBC bldg in burlington mall).

re:
http://www.garlic.com/~lynn/2022f.html#3 COBOL and tricks
http://www.garlic.com/~lynn/2022f.html#13 COBOL and tricks
http://www.garlic.com/~lynn/2022f.html#80 COBOL and tricks

decade ago was asked to track down decision to make all 370s with
virtual memory ... basically MVT storage management was so bad that
region sizes had to be specified four times larger than used, a typical
1mbyte 370/165 only had enough memory for four regions ... insufficient
to keep 165 busy and justified. initial move MVT to "VS2 SVS" mapping to
a 16mbyte virtual address space (very similar to running MVT under CP67
in 16mbyte virtual address space) ... increasing regions by factor of
four times with little or no paging. archived AFC post from 11mar2011
http://www.garlic.com/~lynn/2011d.html#73

--
virtualization experience starting Jan1968, online at home since Mar1970

On 2022-07-25 7:42 p.m., Dan Espen wrote:
>
> I've sure seen lots of COBOL MOVE instructions generate a single
> hardware instruction.

And C compilers may generate in-line code for some of
the standard library functions. The C compiler for GCOS8
does for memcpy().

On Tuesday, July 19, 2022 at 6:27:37 PM UTC-6, Charlie Gibbs wrote:
> There wasn't a single GOTO, but remember
> that a subroutine call is just a GOTO paired with a "come from".

Subroutine calls don't create spaghetti code.

Unless the subroutines don't perform a specific function, that
can be thought of as part of a more fine-grained view of the
program, but instead are inextricably bound to the internal logic
of the routine that called them. In _that_ case, the spirit of the
rules of structured programming is being violated.

John Savard

Alan Bowler <atbowler@thinkage.ca> writes:
>On 2022-07-25 7:42 p.m., Dan Espen wrote:
>>
>> I've sure seen lots of COBOL MOVE instructions generate a single
>> hardware instruction.

Indeed, for Burroughs Medium Systems the instruction set was designed
from the start for COBOL (per Dave Dahm). Generally each cobol
verb generated a single instruction (there was even a table search
instruction and of course an edit instruction for PIC handling).

MVN - Move 1 to 100 unit numeric field to a 1 to 100 unit field, truncating,
padding(with zeros) or zone-filling/stripping as required.
MVA - Move a 1 to 100 unit alpha field to a 1 to 100 unit field, truncating,
padding(with spaces) or zone-filling/stripping (converting 4-bit to 8-bit
by adding a 0xF zone digit).
SEA - Search table for key (<, >, =)

On 2022-09-14, Quadibloc <jsavard@ecn.ab.ca> wrote:

> On Tuesday, July 19, 2022 at 6:27:37 PM UTC-6, Charlie Gibbs wrote:
>
>> There wasn't a single GOTO, but remember
>> that a subroutine call is just a GOTO paired with a "come from".
>
> Subroutine calls don't create spaghetti code.

The undisciplined use of subroutine calls can create spaghetti code
almost as bad as the undisciplined use of GOTOs (although in this
case the strands are doubled). In extreme cases, the overhead of
calling and returning from subroutines can take up a significant
amount of CPU time. There have been cases where such programs
cause virtual memory systems to thrash.

> Unless the subroutines don't perform a specific function, that
> can be thought of as part of a more fine-grained view of the
> program, but instead are inextricably bound to the internal logic
> of the routine that called them. In _that_ case, the spirit of the
> rules of structured programming is being violated.

You're probably thinking of the kind of cases I mentioned above,
which can be thought of as getting a little _too_ fine-grained.
Unfortunately, in the early days Structured Programming attracted
many zealots who saw the number of subroutine calls as a figure
of merit. They are the ones whose programs most often turned out
to be bulky and slow, with logic flow that was almost impossible
to follow. When I took over a program I could often shrink it
by 30% - in extreme cases 50% - with a significant improvement
in readability.

--
/~\ Charlie Gibbs | Microsoft is a dictatorship.
\ / <cgibbs@kltpzyxm.invalid> | Apple is a cult.
X I'm really at ac.dekanfrus | Linux is anarchy.
/ \ if you read it the right way. | Pick your poison.

On Wed, 14 Sep 2022 16:56:45 GMT
Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:

> The undisciplined use of subroutine calls can create spaghetti code
> almost as bad as the undisciplined use of GOTOs (although in this
> case the strands are doubled). In extreme cases, the overhead of
> calling and returning from subroutines can take up a significant
> amount of CPU time.

Brings to mind a little job under messy dos I got involved in many
many years ago - part of the job involved plotting points on the screen in
selected colours, and we had been given a utility library that we were
expected to use for nearly everything and sure enough it had a plot point
in a colour function - so we used it.

The resultant program worked fine except for a small detail of
abysmal performance - profiling ensued - it was spending all its time in
the plot function. Code inspection ensued - this function checked its
parameters (x and y on screen, colour valid) and called a lower level
library plot function which (you've guessed it) checked its parameters (x
and y on screen, colour valid) and called ... I forget how many levels it
went through (far too many) before it made a call to the DOS plot function
which (yes really) checks its parameters as before and calls the BIOS plot
function which checks its parameters and actually sets a pixel.

So I did the obvious and replaced all the plot function calls with
calls to the BIOS result the application went for abysmally slow to very
snappy.

--
Steve O'Hara-Smith
Odds and Ends at http://www.sohara.org/

On Wed, 14 Sep 2022 16:56:45 GMT
Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:

> You're probably thinking of the kind of cases I mentioned above,
> which can be thought of as getting a little _too_ fine-grained.
> Unfortunately, in the early days Structured Programming attracted
> many zealots who saw the number of subroutine calls as a figure
> of merit.

Almost any useful approach to program design can be abused to make
a mess - OTOH if you design for efficiency and code for clarity you can
produce nice[1] code in any paradigm.

[1] In it's precise meaning.

--
Steve O'Hara-Smith
Odds and Ends at http://www.sohara.org/

On 7/25/2022 5:52 PM, Charlie Gibbs wrote:
>
> A sufficiently determined programmer can write unreadable code
> in any language. Convoluted code with no comments is even worse.
> Convoluted code with outdated, incorrect comments is worse still.
>

At a PPoE, we received FORTRAN code under a tech-sharing agreement with
a competitor. A comment atop a convoluted subroutine read:

C
C IT TOOK ME A LONG TIME TO UNDERSTAND HOW THIS
c ROUTINE WORKS, AND IT WILL TAKE YOU A LONG TIME
C ALSO, CAUSE I AM NOT GOING TO TELL YOU HOW IT
C WORKS.
C

a very helpful comment indeed!
--

Charles Richmond

--
This email has been checked for viruses by Avast antivirus software.
www.avast.com

On 7/26/2022 2:38 PM, John Levine wrote:
> According to 25B.Z959 <25B.Z959@nada.net>:
>> Even my ASM ... every line has a comment, an ongoing
>> narrative of what's happening and why. Based on that
>> you could translate it into a dozen higher-level
>> languages.
>
> i += 2; // Add 1 to i.
>

Very common type... It is an "off by one" error. ;-)

--

Charles Richmond

--
This email has been checked for viruses by Avast antivirus software.
www.avast.com

On 7/27/2022 1:10 PM, Charlie Gibbs wrote:
> On 2022-07-27, 25B.Z959 <25B.Z959@nada.net> wrote:
>
>> On 7/26/22 4:53 PM, Vir Campestris wrote:
>>
>>> The useful comments don't say _what_ you're doing, they say _why_.
>>
>> 'Zactly ... it needs to be a 'narrative' - both WHAT and WHY.
>> Doesn't matter if it's ASM or COBOL or FORTRAN or Python. If
>> you or anybody would EVER need to go back and tweak the pgm
>> it really NEEDS serious commenting. Alas it's kinda RARE to
>> see this ..... the old "Well *I* know what it's all about NOW"
>> attitude. Wait a couple years until the boss decides your form
>> needs some new data fields or entry box #7 moved 10 spaces to
>> the right 'cause it would 'look nicer' .... :-)
>
> Don't laugh. Once upon a time, every shop in Canada that ran payroll
> had to modify their programs that generated T4 forms (Statement of
> Remuneration Paid) because someone in Ottawa decided to move one
> box one space to the right.
>
> The hairiest change I was involved in, though, was when the government
> decided to turn the form upside down, i.e. put the row of numbers
> across the top rather than the bottom. (Or maybe it was the other way
> around, I can't remember.) I was maintaining the program in a shop
> whose card-based machine's memory was so small that it didn't have
> room to store all of an employee's information. It would read the
> card containing the employee's name, print it out, then work its
> way through the cards with the employee's address, city, etc.
> Finally it would read the card with the actual numbers and
> print them. The card deck was sorted by employee number and
> a sequence code that ensured that name, address, and statistics
> were in the proper order. Not only did I have to overhaul the
> program's logic to handle the different sequence, the entire
> card deck had to be re-punched (and re-sorted) with the modified
> sequence code.
>
> There has to be a better way to fight unemployment...
>

Charlie Gibbs: Reminds me of your story about an existing program that
read in *all* the large number of data lines, then processed the lines,
and only then sent all of it to the line printer. You re-wrote the code
to read a card, process that one item and immediately send the result to
the printer. This took less code and less memory for processing.

As soon as your program started, the printer began printing results.
And your cow-orkers thought the program must be broken... cause it never
acted like that before!!!

--

Charles Richmond

--
This email has been checked for viruses by Avast antivirus software.
www.avast.com

On 7/26/2022 3:57 PM, Peter Flass wrote:
> John Levine <johnl@taugh.com> wrote:
>> According to 25B.Z959 <25B.Z959@nada.net>:
>>> Even my ASM ... every line has a comment, an ongoing
>>> narrative of what's happening and why. Based on that
>>> you could translate it into a dozen higher-level
>>> languages.
>>
>> i += 2; // Add 1 to i.
>>
>
> For some values of one ;-)
>

No... for smaller values of two. Or perhaps the programmer thought:
"I'm gonna have to increment 'i' again, so I might as well add 2 now
now and cut down on later coding..." ;-(

--

Charles Richmond

--
This email has been checked for viruses by Avast antivirus software.
www.avast.com