Megalextoria
Retro computing and gaming, sci-fi books, tv and movies and other geeky stuff.

Home » Digital Archaeology » Computer Arcana » Computer Folklore » Holy wars of the past - how did they turn out?
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Switch to threaded view of this topic Create a new topic Submit Reply
Re: Holy wars of the past - how did they turn out? [message #405689 is a reply to message #405682] Sun, 14 February 2021 20:27 Go to previous messageGo to next message
Anonymous
Karma:
Originally posted by: J. Clarke

On Sun, 14 Feb 2021 15:31:51 -0700, Peter Flass
<peter_flass@yahoo.com> wrote:

> J. Clarke <jclarke.873638@gmail.com> wrote:
>> On Sun, 14 Feb 2021 00:28:44 -0800 (PST), Quadibloc
>> <jsavard@ecn.ab.ca> wrote:
>>
>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> > On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>
>>>> >> We had a couple hundred programmers
>>>> >> hitting enter all day long waiting for their compile to finish.
>>>
>>>> > Somebody should have taught them how to submit a batch job.
>>>
>>>> Huh?
>>>
>>>> They submitted a batch job, they were waiting for it to finish.
>>>> A 3270 won't show you that unless you hit some key that goes to the
>>>> host.
>>>
>>> And here I thought that, after submitting a batch job for their compile,
>>> they would be free to do other work. Of course, this would make sense
>>> if the compile was on the critical path and so on...
>>
>> "Doing other work" takes one's mind out of the problem-space. Then
>> time is wasted getting back into problem-space for that problem.
>>>
>>> John Savard
>>
>
> On the other hand, often the best way to solve a problem is to work on
> something else, and let your subconscious work on it in the background. I
> have had solutions to problems and come to me in a dream.

My employer had a big kick on "mindfulness" shortly after I started
working there. Paid for an 8 week training including a weekend
retreat. I decided to go for it. To my surprise it is quite
beneficial. Sometimes I get stuck on a problem, pull up the
meditation timer on my phone, meditate for 10 minutes, and the
solution is there. Occasionally I get multiples.
Re: Holy wars of the past - how did they turn out? [message #405691 is a reply to message #405688] Sun, 14 February 2021 20:33 Go to previous messageGo to next message
Dan Espen is currently offline  Dan Espen
Messages: 3867
Registered: January 2012
Karma: 0
Senior Member
Peter Flass <peter_flass@yahoo.com> writes:

> Dan Espen <dan1espen@gmail.com> wrote:
>> Peter Flass <peter_flass@yahoo.com> writes:

>> Hmm, is the Iron Spring PL/I compiler compatible with the IBM debug
>> tool?
>
> I don’t think so, unless the debug tool runs on Linux.

Nope. My mistake, I was thinking Iron Spring ran on mainframes.
I should have thought a little more.

--
Dan Espen
Re: Holy wars of the past - how did they turn out? [message #405692 is a reply to message #405682] Sun, 14 February 2021 21:48 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Monday, February 15, 2021 at 9:31:53 AM UTC+11, Peter Flass wrote:
> J. Clarke <jclarke...@gmail.com> wrote:
>> On Sun, 14 Feb 2021 00:28:44 -0800 (PST), Quadibloc
>> <jsa...@ecn.ab.ca> wrote:
>>
>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> > On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>
>>>> >> We had a couple hundred programmers
>>>> >> hitting enter all day long waiting for their compile to finish.
>>>
>>>> > Somebody should have taught them how to submit a batch job.
>>>
>>>> Huh?
>>>
>>>> They submitted a batch job, they were waiting for it to finish.
>>>> A 3270 won't show you that unless you hit some key that goes to the
>>>> host.
>>>
>>> And here I thought that, after submitting a batch job for their compile,
>>> they would be free to do other work. Of course, this would make sense
>>> if the compile was on the critical path and so on...
>>
>> "Doing other work" takes one's mind out of the problem-space. Then
>> time is wasted getting back into problem-space for that problem.
>>
> On the other hand, often the best way to solve a problem is to work on
> something else, and let your subconscious work on it in the background. I
> have had solutions to problems and come to me in a dream.
..
For me it was under the shower. or playing the piano when some such
inspirations came.
Re: Holy wars of the past - how did they turn out? [message #405693 is a reply to message #405680] Sun, 14 February 2021 21:52 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Monday, February 15, 2021 at 9:31:51 AM UTC+11, Peter Flass wrote:
> Thomas Koenig <tko...@netcologne.de> wrote:
>> Quadibloc <jsa...@ecn.ab.ca> schrieb:
>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> > On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>
>>>> >> We had a couple hundred programmers
>>>> >> hitting enter all day long waiting for their compile to finish.
>>>
>>>> > Somebody should have taught them how to submit a batch job.
>>>
>>>> Huh?
>>>
>>>> They submitted a batch job, they were waiting for it to finish.
>>>> A 3270 won't show you that unless you hit some key that goes to the
>>>> host.
>>>
>>> And here I thought that, after submitting a batch job for their compile,
>>> they would be free to do other work.
>>
>> That would have been the case if they handed their deck to the
>> operator and waited for the printout.
>>
>> Less so if they were just debugging, when short turnaround is
>> _extremely_ important.
>>
>> Makes me wonder a bit - how much programming have you done?
>>
> I didn’t do online debugging until, ate in life. Throw in a few displays,
> ABEND with a core dump, what more do you need? I still view gdb as a last
> option. VM/370 had much more robust facilities, and gdb doesn’t seem to (or
> I haven’t figured out how to get it to) do what I want.
>
> I’m happy with the debug facilities in Iron Spring PL/I. The standard “ON
> condition(xxx) SNAP SYSTEM” is invaluable (often using a condition DEBUG to
> answer the question “how the heck did I get here...?”) I recently also
> added a trace table for more convoluted logic paths.
..
and statements like ON ERROR ... ON SUBSCRIPTRANGE ... etc ?
The SNAP option was the original way IBM produced a traceback,
which was one of the most important debugging tools of the time.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405694 is a reply to message #405668] Sun, 14 February 2021 21:58 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
> On 14 Feb 2021, Robin Vowels wrote
> (in article<dde00e31-f307-4ac2...@googlegroups.com>):
>> On Sunday, February 14, 2021 at 1:36:26 AM UTC+11, Bill Findlay wrote:
>>> On 13 Feb 2021, Robin Vowels wrote
>>> (in article<fe0371e1-cb0f-4ac6...@googlegroups.com>):
>>>> There was an error in the floating-point multiply instruction in the KDF9.
>>> What was that error, and where is it documented?
>> .
>> Bill, more specifically, the error was discovered at Sydney University
>> shortly after the KDF9 was installed. That was about 1963.
>> The program being run was one that had been running on the
>> University's SILLIAC.
>> The program used SILLIAC's hardware multiplier, but also
>> did a software emulation for each multiplication. (The
>> reliability of SILLIAC - a valve machine - necessitated
>> the duplication.) When run on the KDF9, the program's
>> result for the hardware multiplication differed from the
>> software emulation, for certain operands.
>> English Electric decided not to make changes to the hardware.
..
> Thanks for that.
> It sounds like it was a rounding issue, where the 'correct' result
> would be somewhat a matter of opinion and philosophy.
..
Do you think that the problem would have hit the headlines if
it had been a "rounding issue"?
..
> If it had
> significantly affected accuracy there certainly would have been a change
> (which would have been relatively easy to do, as KDF9 was microcoded).
Re: Holy wars of the past - how did they turn out? [message #405699 is a reply to message #405687] Mon, 15 February 2021 03:25 Go to previous messageGo to next message
Quadibloc is currently offline  Quadibloc
Messages: 4399
Registered: June 2012
Karma: 0
Senior Member
On Sunday, February 14, 2021 at 5:30:15 PM UTC-7, Dan Espen wrote:

> "Teach them how to submit a batch job"?

Until you explained that they were pressing Enter in order to see a
notification of their batch job's completion, I thought they were
compiling it interactively, and for some reason they had to keep pressing
Enter to keep from being logged out.

And I imagined they were doing this for hours on end based on your
description. So I had formed an incorrect mental picture of the circumstance.

John Savard
Re: Holy wars of the past - how did they turn out? [message #405700 is a reply to message #405691] Mon, 15 February 2021 03:26 Go to previous messageGo to next message
Quadibloc is currently offline  Quadibloc
Messages: 4399
Registered: June 2012
Karma: 0
Senior Member
On Sunday, February 14, 2021 at 6:33:55 PM UTC-7, Dan Espen wrote:

> Nope. My mistake, I was thinking Iron Spring ran on mainframes.
> I should have thought a little more.

Tiny quibble: what you mean is that you were thinking that Iron Spring
ran on mainframe operating systems. Linux runs on mainframes, but
that isn't particularly useful when what you have is a z/OS installation.

John Savard
Re: Holy wars of the past - how did they turn out? [message #405701 is a reply to message #405682] Mon, 15 February 2021 03:55 Go to previous messageGo to next message
Anonymous
Karma:
Originally posted by: maus

On 2021-02-14, Peter Flass <peter_flass@yahoo.com> wrote:
> J. Clarke <jclarke.873638@gmail.com> wrote:
>> On Sun, 14 Feb 2021 00:28:44 -0800 (PST), Quadibloc
>> <jsavard@ecn.ab.ca> wrote:
>>
>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> > On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>
>>>> >> We had a couple hundred programmers
>>>> >> hitting enter all day long waiting for their compile to finish.
>>>
>>>> > Somebody should have taught them how to submit a batch job.
>>>
>>>> Huh?
>>>
>>>> They submitted a batch job, they were waiting for it to finish.
>>>> A 3270 won't show you that unless you hit some key that goes to the
>>>> host.
>>>
>>> And here I thought that, after submitting a batch job for their compile,
>>> they would be free to do other work. Of course, this would make sense
>>> if the compile was on the critical path and so on...
>>
>> "Doing other work" takes one's mind out of the problem-space. Then
>> time is wasted getting back into problem-space for that problem.
>>>
>>> John Savard
>>
>
> On the other hand, often the best way to solve a problem is to work on
> something else, and let your subconscious work on it in the background. I
> have had solutions to problems and come to me in a dream.
>
+!

--
greymausg@mail.com
Re: Holy wars of the past - how did they turn out? [message #405705 is a reply to message #405699] Mon, 15 February 2021 07:10 Go to previous messageGo to next message
Dan Espen is currently offline  Dan Espen
Messages: 3867
Registered: January 2012
Karma: 0
Senior Member
Quadibloc <jsavard@ecn.ab.ca> writes:

> On Sunday, February 14, 2021 at 5:30:15 PM UTC-7, Dan Espen wrote:
>
>> "Teach them how to submit a batch job"?
>
> Until you explained that they were pressing Enter in order to see a
> notification of their batch job's completion, I thought they were
> compiling it interactively, and for some reason they had to keep pressing
> Enter to keep from being logged out.
>
> And I imagined they were doing this for hours on end based on your
> description. So I had formed an incorrect mental picture of the circumstance.

Your comment now makes some sense.

Elsewhere in the thread I'd mentioned that foreground compiles were
ruled out for silly reasons.

Thanks for clarifying.

Same site also had a mechanism to log out idle users but that was a
matter of 4-6 hours on the mainframe. This was the work of the security
group.

I think it was less on the Unix machines, maybe 2-3 hours.
It's a royal pain to set up a bunch of terminal windows and suddenly
have them start to disappear. SSH has a keepalive option that was
widely in use.

With Emacs, you can just turn on the date/time display.


--
Dan Espen
Re: Holy wars of the past - how did they turn out? [message #405706 is a reply to message #405701] Mon, 15 February 2021 07:15 Go to previous messageGo to next message
Dan Espen is currently offline  Dan Espen
Messages: 3867
Registered: January 2012
Karma: 0
Senior Member
maus <maus@dmaus.org> writes:

> On 2021-02-14, Peter Flass <peter_flass@yahoo.com> wrote:
>> J. Clarke <jclarke.873638@gmail.com> wrote:
>>> On Sun, 14 Feb 2021 00:28:44 -0800 (PST), Quadibloc
>>> <jsavard@ecn.ab.ca> wrote:
>>>
>>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> > Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> >> On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>>
>>>> >>> We had a couple hundred programmers
>>>> >>> hitting enter all day long waiting for their compile to finish.
>>>>
>>>> >> Somebody should have taught them how to submit a batch job.
>>>>
>>>> > Huh?
>>>>
>>>> > They submitted a batch job, they were waiting for it to finish.
>>>> > A 3270 won't show you that unless you hit some key that goes to the
>>>> > host.
>>>>
>>>> And here I thought that, after submitting a batch job for their compile,
>>>> they would be free to do other work. Of course, this would make sense
>>>> if the compile was on the critical path and so on...
>>>
>>> "Doing other work" takes one's mind out of the problem-space. Then
>>> time is wasted getting back into problem-space for that problem.
>>>>
>>>> John Savard
>>>
>>
>> On the other hand, often the best way to solve a problem is to work on
>> something else, and let your subconscious work on it in the background. I
>> have had solutions to problems and come to me in a dream.
>>
> +!

All this is true, lots of times I've figured a problem out while driving
home. But the subject was the interval between problem diagnosis,
submitting a compile and waiting for it to finish. While you're waiting
for the compile you're mentally going over the changes you made and
thinking about the test and it's results. That's not a good time to
be thinking about something else.

--
Dan Espen
Re: Holy wars of the past - how did they turn out? [message #405709 is a reply to message #405700] Mon, 15 February 2021 11:57 Go to previous messageGo to next message
Anonymous
Karma:
Originally posted by: J. Clarke

On Mon, 15 Feb 2021 00:26:56 -0800 (PST), Quadibloc
<jsavard@ecn.ab.ca> wrote:

> On Sunday, February 14, 2021 at 6:33:55 PM UTC-7, Dan Espen wrote:
>
>> Nope. My mistake, I was thinking Iron Spring ran on mainframes.
>> I should have thought a little more.
>
> Tiny quibble: what you mean is that you were thinking that Iron Spring
> ran on mainframe operating systems. Linux runs on mainframes, but
> that isn't particularly useful when what you have is a z/OS installation.

You make the mistake that many others do of assuming "runs on Linux"
implies "runs on all platforms supported by Linux". If I understand
correctly Iron Spring PL/I is distributed as a 32-bit executable for
Intel, which means that it will not run on an IBM mainframe.
Re: Holy wars of the past - how did they turn out? [message #405710 is a reply to message #405709] Mon, 15 February 2021 12:10 Go to previous messageGo to next message
Dan Espen is currently offline  Dan Espen
Messages: 3867
Registered: January 2012
Karma: 0
Senior Member
J. Clarke <jclarke.873638@gmail.com> writes:

> On Mon, 15 Feb 2021 00:26:56 -0800 (PST), Quadibloc
> <jsavard@ecn.ab.ca> wrote:
>
>> On Sunday, February 14, 2021 at 6:33:55 PM UTC-7, Dan Espen wrote:
>>
>>> Nope. My mistake, I was thinking Iron Spring ran on mainframes.
>>> I should have thought a little more.
>>
>> Tiny quibble: what you mean is that you were thinking that Iron Spring
>> ran on mainframe operating systems. Linux runs on mainframes, but
>> that isn't particularly useful when what you have is a z/OS installation.
>
> You make the mistake that many others do of assuming "runs on Linux"
> implies "runs on all platforms supported by Linux". If I understand
> correctly Iron Spring PL/I is distributed as a 32-bit executable for
> Intel, which means that it will not run on an IBM mainframe.

Gee, and less and less Intel platforms.

Oh, just checked, you can run 32 bit apps but there are things you need
to install.

--
Dan Espen
Re: Holy wars of the past - how did they turn out? [message #405711 is a reply to message #405710] Mon, 15 February 2021 12:48 Go to previous messageGo to next message
Anonymous
Karma:
Originally posted by: J. Clarke

On Mon, 15 Feb 2021 12:10:10 -0500, Dan Espen <dan1espen@gmail.com>
wrote:

> J. Clarke <jclarke.873638@gmail.com> writes:
>
>> On Mon, 15 Feb 2021 00:26:56 -0800 (PST), Quadibloc
>> <jsavard@ecn.ab.ca> wrote:
>>
>>> On Sunday, February 14, 2021 at 6:33:55 PM UTC-7, Dan Espen wrote:
>>>
>>>> Nope. My mistake, I was thinking Iron Spring ran on mainframes.
>>>> I should have thought a little more.
>>>
>>> Tiny quibble: what you mean is that you were thinking that Iron Spring
>>> ran on mainframe operating systems. Linux runs on mainframes, but
>>> that isn't particularly useful when what you have is a z/OS installation.
>>
>> You make the mistake that many others do of assuming "runs on Linux"
>> implies "runs on all platforms supported by Linux". If I understand
>> correctly Iron Spring PL/I is distributed as a 32-bit executable for
>> Intel, which means that it will not run on an IBM mainframe.
>
> Gee, and less and less Intel platforms.
>
> Oh, just checked, you can run 32 bit apps but there are things you need
> to install.

Generally speaking 32 bit Intel code runs fine on 64-bit Intel
systems. Where things get sticky is if the 32-bit code needs to be
linked to 64-bit code.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405714 is a reply to message #405694] Mon, 15 February 2021 13:30 Go to previous messageGo to next message
Bill Findlay is currently offline  Bill Findlay
Messages: 286
Registered: January 2012
Karma: 0
Senior Member
On 15 Feb 2021, Robin Vowels wrote
(in article<dac7db79-5ba6-4473-a9fc-1e238fbbcaa5n@googlegroups.com>):

> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>
>> Thanks for that.
>> It sounds like it was a rounding issue, where the 'correct' result
>> would be somewhat a matter of opinion and philosophy.
> .
> Do you think that the problem would have hit the headlines if
> it had been a "rounding issue"?

What headlines did it hit?
I've been a KDF9 user for 54 years and this is the first time I heard of it.
I repeat:

>> If it had
>> significantly affected accuracy there certainly would have been a change
>> (which would have been relatively easy to do, as KDF9 was microcoded).

--
Bill Findlay
Re: Holy wars of the past - how did they turn out? [message #405715 is a reply to message #405711] Mon, 15 February 2021 13:45 Go to previous messageGo to next message
scott is currently offline  scott
Messages: 4237
Registered: February 2012
Karma: 0
Senior Member
J. Clarke <jclarke.873638@gmail.com> writes:
> On Mon, 15 Feb 2021 12:10:10 -0500, Dan Espen <dan1espen@gmail.com>
> wrote:
>
>> J. Clarke <jclarke.873638@gmail.com> writes:
>>
>>> On Mon, 15 Feb 2021 00:26:56 -0800 (PST), Quadibloc
>>> <jsavard@ecn.ab.ca> wrote:
>>>
>>>> On Sunday, February 14, 2021 at 6:33:55 PM UTC-7, Dan Espen wrote:
>>>>
>>>> > Nope. My mistake, I was thinking Iron Spring ran on mainframes.
>>>> > I should have thought a little more.
>>>>
>>>> Tiny quibble: what you mean is that you were thinking that Iron Spring
>>>> ran on mainframe operating systems. Linux runs on mainframes, but
>>>> that isn't particularly useful when what you have is a z/OS installation.
>>>
>>> You make the mistake that many others do of assuming "runs on Linux"
>>> implies "runs on all platforms supported by Linux". If I understand
>>> correctly Iron Spring PL/I is distributed as a 32-bit executable for
>>> Intel, which means that it will not run on an IBM mainframe.
>>
>> Gee, and less and less Intel platforms.
>>
>> Oh, just checked, you can run 32 bit apps but there are things you need
>> to install.
>
> Generally speaking 32 bit Intel code runs fine on 64-bit Intel
> systems.

Assuming that the 32-bit shared objects (for a real OS) or
32-bit DLL's (for windows) have been installed.

> Where things get sticky is if the 32-bit code needs to be
> linked to 64-bit code.
>

I do that regularly on ARMv8 CPUs (when both AArch32
and AArch64 are supported by the CPU).

Granted, it is in the context of the processor validation suite which needs
to test switching between AArch32 and AArch64 :-).
Re: Holy wars of the past - how did they turn out? [message #405717 is a reply to message #405711] Mon, 15 February 2021 15:21 Go to previous messageGo to next message
Dan Espen is currently offline  Dan Espen
Messages: 3867
Registered: January 2012
Karma: 0
Senior Member
J. Clarke <jclarke.873638@gmail.com> writes:

> On Mon, 15 Feb 2021 12:10:10 -0500, Dan Espen <dan1espen@gmail.com>
> wrote:
>
>> J. Clarke <jclarke.873638@gmail.com> writes:
>>
>>> On Mon, 15 Feb 2021 00:26:56 -0800 (PST), Quadibloc
>>> <jsavard@ecn.ab.ca> wrote:
>>>
>>>> On Sunday, February 14, 2021 at 6:33:55 PM UTC-7, Dan Espen wrote:
>>>>
>>>> > Nope. My mistake, I was thinking Iron Spring ran on mainframes.
>>>> > I should have thought a little more.
>>>>
>>>> Tiny quibble: what you mean is that you were thinking that Iron Spring
>>>> ran on mainframe operating systems. Linux runs on mainframes, but
>>>> that isn't particularly useful when what you have is a z/OS installation.
>>>
>>> You make the mistake that many others do of assuming "runs on Linux"
>>> implies "runs on all platforms supported by Linux". If I understand
>>> correctly Iron Spring PL/I is distributed as a 32-bit executable for
>>> Intel, which means that it will not run on an IBM mainframe.
>>
>> Gee, and less and less Intel platforms.
>>
>> Oh, just checked, you can run 32 bit apps but there are things you need
>> to install.
>
> Generally speaking 32 bit Intel code runs fine on 64-bit Intel
> systems. Where things get sticky is if the 32-bit code needs to be
> linked to 64-bit code.

I looked I don't seem to have libc6 installed.
I can see where a compiler that's supposed to be mainframe compatible
might have some issues with 64 bit. More than the average application
anyway.

So, if there are linking issues, I guess it's more important to be able
to produce 64 bit code than having the compiler run that way.

Where I used to work they had a ton of PL/I code that they'd like
to be able to port but they also had tons of HLASM mixed in.
I helped them do the conversion to Enterprise PL/I. Even that
was a nightmare.

--
Dan Espen
Re: Holy wars of the past - how did they turn out? [message #405718 is a reply to message #405693] Mon, 15 February 2021 15:48 Go to previous messageGo to next message
Peter Flass is currently offline  Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
Senior Member
Robin Vowels <robin.vowels@gmail.com> wrote:
> On Monday, February 15, 2021 at 9:31:51 AM UTC+11, Peter Flass wrote:
>> Thomas Koenig <tko...@netcologne.de> wrote:
>>> Quadibloc <jsa...@ecn.ab.ca> schrieb:
>>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> > Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> >> On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>>
>>>> >>> We had a couple hundred programmers
>>>> >>> hitting enter all day long waiting for their compile to finish.
>>>>
>>>> >> Somebody should have taught them how to submit a batch job.
>>>>
>>>> > Huh?
>>>>
>>>> > They submitted a batch job, they were waiting for it to finish.
>>>> > A 3270 won't show you that unless you hit some key that goes to the
>>>> > host.
>>>>
>>>> And here I thought that, after submitting a batch job for their compile,
>>>> they would be free to do other work.
>>>
>>> That would have been the case if they handed their deck to the
>>> operator and waited for the printout.
>>>
>>> Less so if they were just debugging, when short turnaround is
>>> _extremely_ important.
>>>
>>> Makes me wonder a bit - how much programming have you done?
>>>
>> I didn’t do online debugging until, ate in life. Throw in a few displays,
>> ABEND with a core dump, what more do you need? I still view gdb as a last
>> option. VM/370 had much more robust facilities, and gdb doesn’t seem to (or
>> I haven’t figured out how to get it to) do what I want.
>>
>> I’m happy with the debug facilities in Iron Spring PL/I. The standard “ON
>> condition(xxx) SNAP SYSTEM” is invaluable (often using a condition DEBUG to
>> answer the question “how the heck did I get here...?”) I recently also
>> added a trace table for more convoluted logic paths.
> .
> and statements like ON ERROR ... ON SUBSCRIPTRANGE ... etc ?
> The SNAP option was the original way IBM produced a traceback,
> which was one of the most important debugging tools of the time.
>

That’s all supported. I love the traceback, I use it a lot.

--
Pete
Re: Holy wars of the past - how did they turn out? [message #405719 is a reply to message #405705] Mon, 15 February 2021 15:48 Go to previous messageGo to next message
Peter Flass is currently offline  Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
Senior Member
Dan Espen <dan1espen@gmail.com> wrote:
> Quadibloc <jsavard@ecn.ab.ca> writes:
>
>> On Sunday, February 14, 2021 at 5:30:15 PM UTC-7, Dan Espen wrote:
>>
>>> "Teach them how to submit a batch job"?
>>
>> Until you explained that they were pressing Enter in order to see a
>> notification of their batch job's completion, I thought they were
>> compiling it interactively, and for some reason they had to keep pressing
>> Enter to keep from being logged out.
>>
>> And I imagined they were doing this for hours on end based on your
>> description. So I had formed an incorrect mental picture of the circumstance.
>
> Your comment now makes some sense.
>
> Elsewhere in the thread I'd mentioned that foreground compiles were
> ruled out for silly reasons.
>
> Thanks for clarifying.
>
> Same site also had a mechanism to log out idle users but that was a
> matter of 4-6 hours on the mainframe. This was the work of the security
> group.
>
> I think it was less on the Unix machines, maybe 2-3 hours.
> It's a royal pain to set up a bunch of terminal windows and suddenly
> have them start to disappear. SSH has a keepalive option that was
> widely in use.
>
> With Emacs, you can just turn on the date/time display.
>
>

We used to do, I think, half an hour. At any rate, not long enough to keep
you logged on over lunch. I thought most systems did this by default. My
home Ubuntu is systems do this. Something like what you describe doesn’t
work.

--
Pete
Re: Holy wars of the past - how did they turn out? [message #405720 is a reply to message #405719] Mon, 15 February 2021 16:27 Go to previous messageGo to next message
Dan Espen is currently offline  Dan Espen
Messages: 3867
Registered: January 2012
Karma: 0
Senior Member
Peter Flass <peter_flass@yahoo.com> writes:

> Dan Espen <dan1espen@gmail.com> wrote:
>> Quadibloc <jsavard@ecn.ab.ca> writes:
>>
>>> On Sunday, February 14, 2021 at 5:30:15 PM UTC-7, Dan Espen wrote:
>>>
>>>> "Teach them how to submit a batch job"?
>>>
>>> Until you explained that they were pressing Enter in order to see a
>>> notification of their batch job's completion, I thought they were
>>> compiling it interactively, and for some reason they had to keep pressing
>>> Enter to keep from being logged out.
>>>
>>> And I imagined they were doing this for hours on end based on your
>>> description. So I had formed an incorrect mental picture of the circumstance.
>>
>> Your comment now makes some sense.
>>
>> Elsewhere in the thread I'd mentioned that foreground compiles were
>> ruled out for silly reasons.
>>
>> Thanks for clarifying.
>>
>> Same site also had a mechanism to log out idle users but that was a
>> matter of 4-6 hours on the mainframe. This was the work of the security
>> group.
>>
>> I think it was less on the Unix machines, maybe 2-3 hours.
>> It's a royal pain to set up a bunch of terminal windows and suddenly
>> have them start to disappear. SSH has a keepalive option that was
>> widely in use.
>>
>> With Emacs, you can just turn on the date/time display.
>
> We used to do, I think, half an hour. At any rate, not long enough to keep
> you logged on over lunch. I thought most systems did this by default. My
> home Ubuntu is systems do this. Something like what you describe doesn’t
> work.

Hmm, I thought we had working work-arounds but maybe not.
I remember at the end I couldn't keep my remote Emacs session overnight
any more. It used to be up for weeks at a time.

--
Dan Espen
Re: Holy wars of the past - how did they turn out? [message #405724 is a reply to message #405718] Mon, 15 February 2021 19:15 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Tuesday, February 16, 2021 at 7:48:52 AM UTC+11, Peter Flass wrote:
> Robin Vowels <robin....@gmail.com> wrote:
>> On Monday, February 15, 2021 at 9:31:51 AM UTC+11, Peter Flass wrote:
>>> Thomas Koenig <tko...@netcologne.de> wrote:
>>>> Quadibloc <jsa...@ecn.ab.ca> schrieb:
>>>> > On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> >> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> >>> On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>> >
>>>> >>>> We had a couple hundred programmers
>>>> >>>> hitting enter all day long waiting for their compile to finish.
>>>> >
>>>> >>> Somebody should have taught them how to submit a batch job.
>>>> >
>>>> >> Huh?
>>>> >
>>>> >> They submitted a batch job, they were waiting for it to finish.
>>>> >> A 3270 won't show you that unless you hit some key that goes to the
>>>> >> host.
>>>> >
>>>> > And here I thought that, after submitting a batch job for their compile,
>>>> > they would be free to do other work.
>>>>
>>>> That would have been the case if they handed their deck to the
>>>> operator and waited for the printout.
>>>>
>>>> Less so if they were just debugging, when short turnaround is
>>>> _extremely_ important.
>>>>
>>>> Makes me wonder a bit - how much programming have you done?
>>>>
>>> I didn’t do online debugging until, ate in life. Throw in a few displays,
>>> ABEND with a core dump, what more do you need? I still view gdb as a last
>>> option. VM/370 had much more robust facilities, and gdb doesn’t seem to (or
>>> I haven’t figured out how to get it to) do what I want.
>>>
>>> I’m happy with the debug facilities in Iron Spring PL/I. The standard “ON
>>> condition(xxx) SNAP SYSTEM” is invaluable (often using a condition DEBUG to
>>> answer the question “how the heck did I get here...?”) I recently also
>>> added a trace table for more convoluted logic paths.
>> .
>> and statements like ON ERROR ... ON SUBSCRIPTRANGE ... etc ?
>> The SNAP option was the original way IBM produced a traceback,
>> which was one of the most important debugging tools of the time.
>
> That’s all supported. I love the traceback, I use it a lot.
..
Excellent !! Does the traceback come with statement numbers?

I also routinely specify traceback (with SNAP).
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405725 is a reply to message #405714] Mon, 15 February 2021 19:18 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Tuesday, February 16, 2021 at 5:30:38 AM UTC+11, Bill Findlay wrote:
> On 15 Feb 2021, Robin Vowels wrote
> (in article<dac7db79-5ba6-4473...@googlegroups.com>):
>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>
>>> Thanks for that.
>>> It sounds like it was a rounding issue, where the 'correct' result
>>> would be somewhat a matter of opinion and philosophy.
>> .
>> Do you think that the problem would have hit the headlines if
>> it had been a "rounding issue"?
..
> What headlines did it hit?
> I've been a KDF9 user for 54 years and this is the first time I heard of it.
..
The event occurred in Sydney, Australia, years before you were associated with KDF9.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405726 is a reply to message #405714] Mon, 15 February 2021 19:30 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Tuesday, February 16, 2021 at 5:30:38 AM UTC+11, Bill Findlay wrote:
> On 15 Feb 2021, Robin Vowels wrote
> (in article<dac7db79-5ba6-4473...@googlegroups.com>):
>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>
>>> Thanks for that.
>>> It sounds like it was a rounding issue, where the 'correct' result
>>> would be somewhat a matter of opinion and philosophy.
>> .
>> Do you think that the problem would have hit the headlines if
>> it had been a "rounding issue"?
> What headlines did it hit?

As to the general problems with the KDF9:

Canberra Times 8 March 1965 p. 2


Computers on television. It is
inevitable that there should
from time to time be controver-
sies about television pro-
grammes. Public protest against
some programmes — Govern-
ment protest against others. But
we ran into a very curious type
of controversy this week when
B.B.C.-2 proposed to show, and
in spite of government expres-
sions of anxiety did in the end
show, a programme called The
Great Computers Scandal.
The point of the programme
was to assert that a large num-
ber of British computers that
have recently been put on the
market have been most unsatis-
factory and purchasers have
scrapped them and bought
American computers instead. I
have no opinion of my own on
the merits of these computers.
But in the programme Dr. Bene-
dict Nixon, one of the experts
operating the Ferranti Atlas
machine which had been in-
stalled for London University's
computer unit at the cost of
£3.4 million, complained that it
broke down on an average every
fifteen minutes and that its cal-
culations were frequently inac-
curate. The English Electric
Marconi KDF-9, which had
been installed in seven universi-
ties, was alleged not to be up
to its job. It work at Birming-
ham University was taken over
at a reduced rate by the Ameri-
can International Business
Machines. But the Atlas and the
English Electric people have
their answers. The Atlas people
say that they do not make these
models any more. The English
Electric say that their first
models had "teething troubles"
but that these troubles have been
largely eliminated in subsequent
models.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405728 is a reply to message #405714] Mon, 15 February 2021 19:47 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Tuesday, February 16, 2021 at 5:30:38 AM UTC+11, Bill Findlay wrote:
> On 15 Feb 2021, Robin Vowels wrote
> (in article<dac7db79-5ba6-4473...@googlegroups.com>):
>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>
>>> Thanks for that.
>>> It sounds like it was a rounding issue, where the 'correct' result
>>> would be somewhat a matter of opinion and philosophy.
>> .
>> Do you think that the problem would have hit the headlines if
>> it had been a "rounding issue"?
..
> What headlines did it hit?
> I've been a KDF9 user for 54 years and this is the first time I heard of it.
..
More on the KDF9 (From "A Vision Splendid", p. 43):
..
SILLIAC ended its life as a peripheral device to the KDF9. Despite being
a much newer machine, the KDF’s tape drives were temperamental
and SILLIAC had four reliable tape drives, although they could not be
adapted for the KDF9 in any way other than building a data communications
link between the two.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405730 is a reply to message #405714] Mon, 15 February 2021 19:56 Go to previous messageGo to next message
Anonymous
Karma:
Originally posted by: Bob Eager

On Mon, 15 Feb 2021 18:30:35 +0000, Bill Findlay wrote:

> On 15 Feb 2021, Robin Vowels wrote (in
> article<dac7db79-5ba6-4473-a9fc-1e238fbbcaa5n@googlegroups.com>):
>
>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>
>>> Thanks for that.
>>> It sounds like it was a rounding issue, where the 'correct' result
>>> would be somewhat a matter of opinion and philosophy.
>> .
>> Do you think that the problem would have hit the headlines if it had
>> been a "rounding issue"?
>
> What headlines did it hit?
> I've been a KDF9 user for 54 years and this is the first time I heard of
> it.

Strangely, I was scanning a KDF8 manual and a KDP10 manual a couple of
weeks ago...

--
Using UNIX since v6 (1975)...

Use the BIG mirror service in the UK:
http://www.mirrorservice.org
Re: Holy wars of the past - how did they turn out? [message #405744 is a reply to message #405724] Tue, 16 February 2021 12:57 Go to previous messageGo to next message
Peter Flass is currently offline  Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
Senior Member
Robin Vowels <robin.vowels@gmail.com> wrote:
> On Tuesday, February 16, 2021 at 7:48:52 AM UTC+11, Peter Flass wrote:
>> Robin Vowels <robin....@gmail.com> wrote:
>>> On Monday, February 15, 2021 at 9:31:51 AM UTC+11, Peter Flass wrote:
>>>> Thomas Koenig <tko...@netcologne.de> wrote:
>>>> > Quadibloc <jsa...@ecn.ab.ca> schrieb:
>>>> >> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> >>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> >>>> On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>> >>
>>>> >>>>> We had a couple hundred programmers
>>>> >>>>> hitting enter all day long waiting for their compile to finish.
>>>> >>
>>>> >>>> Somebody should have taught them how to submit a batch job.
>>>> >>
>>>> >>> Huh?
>>>> >>
>>>> >>> They submitted a batch job, they were waiting for it to finish.
>>>> >>> A 3270 won't show you that unless you hit some key that goes to the
>>>> >>> host.
>>>> >>
>>>> >> And here I thought that, after submitting a batch job for their compile,
>>>> >> they would be free to do other work.
>>>> >
>>>> > That would have been the case if they handed their deck to the
>>>> > operator and waited for the printout.
>>>> >
>>>> > Less so if they were just debugging, when short turnaround is
>>>> > _extremely_ important.
>>>> >
>>>> > Makes me wonder a bit - how much programming have you done?
>>>> >
>>>> I didn’t do online debugging until, ate in life. Throw in a few displays,
>>>> ABEND with a core dump, what more do you need? I still view gdb as a last
>>>> option. VM/370 had much more robust facilities, and gdb doesn’t seem to (or
>>>> I haven’t figured out how to get it to) do what I want.
>>>>
>>>> I’m happy with the debug facilities in Iron Spring PL/I. The standard “ON
>>>> condition(xxx) SNAP SYSTEM” is invaluable (often using a condition DEBUG to
>>>> answer the question “how the heck did I get here...?”) I recently also
>>>> added a trace table for more convoluted logic paths.
>>> .
>>> and statements like ON ERROR ... ON SUBSCRIPTRANGE ... etc ?
>>> The SNAP option was the original way IBM produced a traceback,
>>> which was one of the most important debugging tools of the time.
>>
>> That’s all supported. I love the traceback, I use it a lot.
> .
> Excellent !! Does the traceback come with statement numbers?
>
> I also routinely specify traceback (with SNAP).
>

No. You know, it could. Modules optionally have a table of correspondence
between offsets and line numbers. I’ll think about adding this. You
currently only get the line number where the condition is raised.

--
Pete
Re: Holy wars of the past - how did they turn out? [message #405746 is a reply to message #405744] Tue, 16 February 2021 13:35 Go to previous messageGo to next message
Dan Espen is currently offline  Dan Espen
Messages: 3867
Registered: January 2012
Karma: 0
Senior Member
Peter Flass <peter_flass@yahoo.com> writes:

> Robin Vowels <robin.vowels@gmail.com> wrote:
>> On Tuesday, February 16, 2021 at 7:48:52 AM UTC+11, Peter Flass wrote:
>>> Robin Vowels <robin....@gmail.com> wrote:
>>>> On Monday, February 15, 2021 at 9:31:51 AM UTC+11, Peter Flass wrote:
>>>> > Thomas Koenig <tko...@netcologne.de> wrote:
>>>> >> Quadibloc <jsa...@ecn.ab.ca> schrieb:
>>>> >>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> >>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> >>>>> On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>> >>>
>>>> >>>>>> We had a couple hundred programmers
>>>> >>>>>> hitting enter all day long waiting for their compile to finish.
>>>> >>>
>>>> >>>>> Somebody should have taught them how to submit a batch job.
>>>> >>>
>>>> >>>> Huh?
>>>> >>>
>>>> >>>> They submitted a batch job, they were waiting for it to finish.
>>>> >>>> A 3270 won't show you that unless you hit some key that goes to the
>>>> >>>> host.
>>>> >>>
>>>> >>> And here I thought that, after submitting a batch job for their compile,
>>>> >>> they would be free to do other work.
>>>> >>
>>>> >> That would have been the case if they handed their deck to the
>>>> >> operator and waited for the printout.
>>>> >>
>>>> >> Less so if they were just debugging, when short turnaround is
>>>> >> _extremely_ important.
>>>> >>
>>>> >> Makes me wonder a bit - how much programming have you done?
>>>> >>
>>>> > I didn’t do online debugging until, ate in life. Throw in a few displays,
>>>> > ABEND with a core dump, what more do you need? I still view gdb as a last
>>>> > option. VM/370 had much more robust facilities, and gdb doesn’t seem to (or
>>>> > I haven’t figured out how to get it to) do what I want.
>>>> >
>>>> > I’m happy with the debug facilities in Iron Spring PL/I. The standard “ON
>>>> > condition(xxx) SNAP SYSTEM” is invaluable (often using a condition DEBUG to
>>>> > answer the question “how the heck did I get here...?”) I recently also
>>>> > added a trace table for more convoluted logic paths.
>>>> .
>>>> and statements like ON ERROR ... ON SUBSCRIPTRANGE ... etc ?
>>>> The SNAP option was the original way IBM produced a traceback,
>>>> which was one of the most important debugging tools of the time.
>>>
>>> That’s all supported. I love the traceback, I use it a lot.
>> .
>> Excellent !! Does the traceback come with statement numbers?
>>
>> I also routinely specify traceback (with SNAP).
>
> No. You know, it could. Modules optionally have a table of correspondence
> between offsets and line numbers. I’ll think about adding this. You
> currently only get the line number where the condition is raised.

I can describe how the table of line numbers is done in mainframe PL/I
if you are interested. Of course this will be from memory but I can get
close. Don't forget line numbers have to point at lines in macros too.

--
Dan Espen
Re: Holy wars of the past - how did they turn out? [message #405749 is a reply to message #405746] Tue, 16 February 2021 17:14 Go to previous messageGo to next message
Peter Flass is currently offline  Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
Senior Member
Dan Espen <dan1espen@gmail.com> wrote:
> Peter Flass <peter_flass@yahoo.com> writes:
>
>> Robin Vowels <robin.vowels@gmail.com> wrote:
>>> On Tuesday, February 16, 2021 at 7:48:52 AM UTC+11, Peter Flass wrote:
>>>> Robin Vowels <robin....@gmail.com> wrote:
>>>> > On Monday, February 15, 2021 at 9:31:51 AM UTC+11, Peter Flass wrote:
>>>> >> Thomas Koenig <tko...@netcologne.de> wrote:
>>>> >>> Quadibloc <jsa...@ecn.ab.ca> schrieb:
>>>> >>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> >>>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> >>>>>> On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>> >>>>
>>>> >>>>>>> We had a couple hundred programmers
>>>> >>>>>>> hitting enter all day long waiting for their compile to finish.
>>>> >>>>
>>>> >>>>>> Somebody should have taught them how to submit a batch job.
>>>> >>>>
>>>> >>>>> Huh?
>>>> >>>>
>>>> >>>>> They submitted a batch job, they were waiting for it to finish.
>>>> >>>>> A 3270 won't show you that unless you hit some key that goes to the
>>>> >>>>> host.
>>>> >>>>
>>>> >>>> And here I thought that, after submitting a batch job for their compile,
>>>> >>>> they would be free to do other work.
>>>> >>>
>>>> >>> That would have been the case if they handed their deck to the
>>>> >>> operator and waited for the printout.
>>>> >>>
>>>> >>> Less so if they were just debugging, when short turnaround is
>>>> >>> _extremely_ important.
>>>> >>>
>>>> >>> Makes me wonder a bit - how much programming have you done?
>>>> >>>
>>>> >> I didn’t do online debugging until, ate in life. Throw in a few displays,
>>>> >> ABEND with a core dump, what more do you need? I still view gdb as a last
>>>> >> option. VM/370 had much more robust facilities, and gdb doesn’t seem to (or
>>>> >> I haven’t figured out how to get it to) do what I want.
>>>> >>
>>>> >> I’m happy with the debug facilities in Iron Spring PL/I. The standard “ON
>>>> >> condition(xxx) SNAP SYSTEM” is invaluable (often using a condition DEBUG to
>>>> >> answer the question “how the heck did I get here...?”) I recently also
>>>> >> added a trace table for more convoluted logic paths.
>>>> > .
>>>> > and statements like ON ERROR ... ON SUBSCRIPTRANGE ... etc ?
>>>> > The SNAP option was the original way IBM produced a traceback,
>>>> > which was one of the most important debugging tools of the time.
>>>>
>>>> That’s all supported. I love the traceback, I use it a lot.
>>> .
>>> Excellent !! Does the traceback come with statement numbers?
>>>
>>> I also routinely specify traceback (with SNAP).
>>
>> No. You know, it could. Modules optionally have a table of correspondence
>> between offsets and line numbers. I’ll think about adding this. You
>> currently only get the line number where the condition is raised.
>
> I can describe how the table of line numbers is done in mainframe PL/I
> if you are interested. Of course this will be from memory but I can get
> close. Don't forget line numbers have to point at lines in macros too.
>

Thanks, I have that part. As I said, the line number will appear in the
message for the ON-condition. It just never occurred to me to put them in
the traceback.

--
Pete
Re: Holy wars of the past - how did they turn out? [message #405752 is a reply to message #405746] Tue, 16 February 2021 19:15 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Wednesday, February 17, 2021 at 5:35:28 AM UTC+11, Dan Espen wrote:
> Peter Flass <peter...@yahoo.com> writes:
>
>> Robin Vowels <robin....@gmail.com> wrote:
>>> On Tuesday, February 16, 2021 at 7:48:52 AM UTC+11, Peter Flass wrote:
>>>> Robin Vowels <robin....@gmail.com> wrote:
>>>> > On Monday, February 15, 2021 at 9:31:51 AM UTC+11, Peter Flass wrote:
>>>> >> Thomas Koenig <tko...@netcologne.de> wrote:
>>>> >>> Quadibloc <jsa...@ecn.ab.ca> schrieb:
>>>> >>>> On Saturday, February 13, 2021 at 4:06:11 AM UTC-7, Dan Espen wrote:
>>>> >>>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>> >>>>>> On Sunday, February 7, 2021 at 6:33:51 PM UTC-7, Dan Espen wrote:
>>>> >>>>
>>>> >>>>>>> We had a couple hundred programmers
>>>> >>>>>>> hitting enter all day long waiting for their compile to finish..
>>>> >>>>
>>>> >>>>>> Somebody should have taught them how to submit a batch job.
>>>> >>>>
>>>> >>>>> Huh?
>>>> >>>>
>>>> >>>>> They submitted a batch job, they were waiting for it to finish.
>>>> >>>>> A 3270 won't show you that unless you hit some key that goes to the
>>>> >>>>> host.
>>>> >>>>
>>>> >>>> And here I thought that, after submitting a batch job for their compile,
>>>> >>>> they would be free to do other work.
>>>> >>>
>>>> >>> That would have been the case if they handed their deck to the
>>>> >>> operator and waited for the printout.
>>>> >>>
>>>> >>> Less so if they were just debugging, when short turnaround is
>>>> >>> _extremely_ important.
>>>> >>>
>>>> >>> Makes me wonder a bit - how much programming have you done?
>>>> >>>
>>>> >> I didn’t do online debugging until, ate in life. Throw in a few displays,
>>>> >> ABEND with a core dump, what more do you need? I still view gdb as a last
>>>> >> option. VM/370 had much more robust facilities, and gdb doesn’t seem to (or
>>>> >> I haven’t figured out how to get it to) do what I want.
>>>> >>
>>>> >> I’m happy with the debug facilities in Iron Spring PL/I. The standard “ON
>>>> >> condition(xxx) SNAP SYSTEM” is invaluable (often using a condition DEBUG to
>>>> >> answer the question “how the heck did I get here...?”) I recently also
>>>> >> added a trace table for more convoluted logic paths.
>>>> > .
>>>> > and statements like ON ERROR ... ON SUBSCRIPTRANGE ... etc ?
>>>> > The SNAP option was the original way IBM produced a traceback,
>>>> > which was one of the most important debugging tools of the time.
>>>>
>>>> That’s all supported. I love the traceback, I use it a lot.
>>> .
>>> Excellent !! Does the traceback come with statement numbers?
>>>
>>> I also routinely specify traceback (with SNAP).
>>
>> No. You know, it could. Modules optionally have a table of correspondence
>> between offsets and line numbers. I’ll think about adding this. You
>> currently only get the line number where the condition is raised.
..
That's at least better than VA PL/I, which gives none.
..
> I can describe how the table of line numbers is done in mainframe PL/I
> if you are interested. Of course this will be from memory but I can get
> close. Don't forget line numbers have to point at lines in macros too.
Re: Holy wars of the past - how did they turn out? [message #405779 is a reply to message #405516] Wed, 17 February 2021 12:04 Go to previous messageGo to next message
Charlie Gibbs is currently offline  Charlie Gibbs
Messages: 5313
Registered: January 2012
Karma: 0
Senior Member
On 2021-02-11, Kerr-Mudd,John <notsaying@127.0.0.1> wrote:

> On Thu, 11 Feb 2021 18:59:31 GMT, Charlie Gibbs
> <cgibbs@kltpzyxm.invalid> wrote:
>
>> I used a couple of mainframe editors in that brief period between
>> card-based systems and personal computers, but don't remember too
>> much of them now. MS-DOS's ed was enough like CP/M's ed that you
>> could pick it up quickly, while being different enough that you
>> could shoot yourself in the foot. One shop used KEDIT, which I
>> hear is somewhat like IBM's XEDIT. It's a nice enough editor that
>> I still keep a copy around and use it when Notepad doesn't turn
>> my crank.
>
> I also was a Kedit user! I still occasionally run it to delete line
> numbers from an output listing. (column delete)
>
> I wish I'd kept a copy of 3.53; it was smaller and ran my ked programs
> fine before they changed to kex (a rexx subset) as the key macro language
> in version 5.
>
> I don't suppose you have a copy hanging around anywhere that needs
> (cough) an off-site backup?

Let me know in e-mail where the site is.
I'm the guy standing on a board with no clothes on.

--
/~\ Charlie Gibbs | "Some of you may die,
\ / <cgibbs@kltpzyxm.invalid> | but it's a sacrifice
X I'm really at ac.dekanfrus | I'm willing to make."
/ \ if you read it the right way. | -- Lord Farquaad (Shrek)
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405830 is a reply to message #405725] Fri, 19 February 2021 16:21 Go to previous messageGo to next message
Bill Findlay is currently offline  Bill Findlay
Messages: 286
Registered: January 2012
Karma: 0
Senior Member
On 16 Feb 2021, Robin Vowels wrote
(in article<94b86c54-1864-40e1-be10-1d67cb51521dn@googlegroups.com>):

> On Tuesday, February 16, 2021 at 5:30:38 AM UTC+11, Bill Findlay wrote:
>> On 15 Feb 2021, Robin Vowels wrote
>> (in article<dac7db79-5ba6-4473...@googlegroups.com>):
>>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>>
>>>> Thanks for that.
>>>> It sounds like it was a rounding issue, where the 'correct' result
>>>> would be somewhat a matter of opinion and philosophy.
>>> .
>>> Do you think that the problem would have hit the headlines if
>>> it had been a "rounding issue"?
> .
>> What headlines did it hit?
>> I've been a KDF9 user for 54 years and this is the first time I heard of it.
> .
> The event occurred in Sydney, Australia, years before you were associated
> with KDF9.

where years <= 3

All the more reason for thinking the problem was quickly resolved.
KDF9 was a 'scientific' computer used in Universities and in engineering
companies, on projects such as nuclear power plants and supersonic aircraft.
A significant inaccuracy could not have been tolerated.

--
Bill Findlay
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405831 is a reply to message #405728] Fri, 19 February 2021 16:24 Go to previous messageGo to next message
Bill Findlay is currently offline  Bill Findlay
Messages: 286
Registered: January 2012
Karma: 0
Senior Member
On 16 Feb 2021, Robin Vowels wrote
(in article<1073e8b1-b258-4f1a-92bc-487fa0b584e6n@googlegroups.com>):

> On Tuesday, February 16, 2021 at 5:30:38 AM UTC+11, Bill Findlay wrote:
>> On 15 Feb 2021, Robin Vowels wrote
>> (in article<dac7db79-5ba6-4473...@googlegroups.com>):
>>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>>
>>>> Thanks for that.
>>>> It sounds like it was a rounding issue, where the 'correct' result
>>>> would be somewhat a matter of opinion and philosophy.
>>> .
>>> Do you think that the problem would have hit the headlines if
>>> it had been a "rounding issue"?
> .
>> What headlines did it hit?
>> I've been a KDF9 user for 54 years and this is the first time I heard of it.
> .
> More on the KDF9 (From "A Vision Splendid", p. 43):
> .
> SILLIAC ended its life as a peripheral device to the KDF9. Despite being
> a much newer machine, the KDF´s tape drives were temperamental

Not in my experience. Perhaps they were not well maintained in Sydney.
(They were made by RCA, of course.)

> and SILLIAC had four reliable tape drives

A bit of boosterism for the home team?
--
Bill Findlay
Re: Holy wars of the past - how did they turn out? [message #405850 is a reply to message #405639] Sat, 20 February 2021 17:47 Go to previous messageGo to next message
Jorgen Grahn is currently offline  Jorgen Grahn
Messages: 606
Registered: March 2012
Karma: 0
Senior Member
On Sat, 2021-02-13, Anssi Saari wrote:
> Jorgen Grahn <grahn+nntp@snipabacken.se> writes:
>
>> On my own Linux systems, I uninstall vim and vim-tiny and whatever,
>> and install nvi instead -- that should be pretty close to what you
>> want.
>
> I like nvi but in this day and age I want Unicode support in an
> editor. So vim it is for me, when I'm not using Emacs.

Ah. I must admit I have been naughty and stuck with iso8859-1 for most
of my files, especially those I'm likely to use with vi.

/Jorgen

--
// Jorgen Grahn <grahn@ Oo o. . .
\X/ snipabacken.se> O o .
Re: Holy wars of the past - how did they turn out? [message #405851 is a reply to message #405850] Sat, 20 February 2021 18:21 Go to previous messageGo to next message
Peter Flass is currently offline  Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
Senior Member
Jorgen Grahn <grahn+nntp@snipabacken.se> wrote:
> On Sat, 2021-02-13, Anssi Saari wrote:
>> Jorgen Grahn <grahn+nntp@snipabacken.se> writes:
>>
>>> On my own Linux systems, I uninstall vim and vim-tiny and whatever,
>>> and install nvi instead -- that should be pretty close to what you
>>> want.
>>
>> I like nvi but in this day and age I want Unicode support in an
>> editor. So vim it is for me, when I'm not using Emacs.
>
> Ah. I must admit I have been naughty and stuck with iso8859-1 for most
> of my files, especially those I'm likely to use with vi.
>
> /Jorgen
>

Me too. I used this translating from EBCDIC - IIRC IBM CP1050 to whatever
codepage 8859-1 is.

--
Pete
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405854 is a reply to message #405830] Sat, 20 February 2021 20:26 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Saturday, February 20, 2021 at 8:21:48 AM UTC+11, Bill Findlay wrote:
> On 16 Feb 2021, Robin Vowels wrote
> (in article<94b86c54-1864-40e1...@googlegroups.com>):
>> On Tuesday, February 16, 2021 at 5:30:38 AM UTC+11, Bill Findlay wrote:
>>> On 15 Feb 2021, Robin Vowels wrote
>>> (in article<dac7db79-5ba6-4473...@googlegroups.com>):
>>>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>> >
>>>> > Thanks for that.
>>>> > It sounds like it was a rounding issue, where the 'correct' result
>>>> > would be somewhat a matter of opinion and philosophy.
>>>> .
>>>> Do you think that the problem would have hit the headlines if
>>>> it had been a "rounding issue"?
>> .
>>> What headlines did it hit?
>>> I've been a KDF9 user for 54 years and this is the first time I heard of it.
>> .
>> The event occurred in Sydney, Australia, years before you were associated
>> with KDF9.
..
> where years <= 3
>
> All the more reason for thinking the problem was quickly resolved.
..
You're clutching at straws.
..
> KDF9 was a 'scientific' computer used in Universities and in engineering
> companies, on projects such as nuclear power plants and supersonic aircraft.
> A significant inaccuracy could not have been tolerated.

You've already ignored "The Great Computers Scandal"
in 1965 involving SEVEN repeat SEVEN KDF9 installations at
SEVEN repeat SEVEN universities, and that Manchester University
had already replaced its KDF9 with an IBM machine.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405855 is a reply to message #405831] Sat, 20 February 2021 20:34 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Saturday, February 20, 2021 at 8:24:12 AM UTC+11, Bill Findlay wrote:
> On 16 Feb 2021, Robin Vowels wrote
> (in article<1073e8b1-b258-4f1a...@googlegroups.com>):
>> On Tuesday, February 16, 2021 at 5:30:38 AM UTC+11, Bill Findlay wrote:
>>> On 15 Feb 2021, Robin Vowels wrote
>>> (in article<dac7db79-5ba6-4473...@googlegroups.com>):
>>>> On Monday, February 15, 2021 at 2:30:36 AM UTC+11, Bill Findlay wrote:
>>>> >
>>>> > Thanks for that.
>>>> > It sounds like it was a rounding issue, where the 'correct' result
>>>> > would be somewhat a matter of opinion and philosophy.
>>>> .
>>>> Do you think that the problem would have hit the headlines if
>>>> it had been a "rounding issue"?
>> .
>>> What headlines did it hit?
>>> I've been a KDF9 user for 54 years and this is the first time I heard of it.
>> .
>> More on the KDF9 (From "A Vision Splendid", p. 43):
>> .
>> SILLIAC ended its life as a peripheral device to the KDF9. Despite being
>> a much newer machine, the KDF9´s tape drives were temperamental
..
> Not in my experience.
..
Your experience started some three years after the events.
..
> Perhaps they were not well maintained in Sydney.
..
Sydney University had been maintaining their computers since 1956 --
and that included their magnetic tape drives.
You think that they did not know how to maintain magnetic tape drives?
..
> (They were made by RCA, of course.)
..
>> and SILLIAC had four reliable tape drives
..
> A bit of boosterism for the home team?
..
I merely quoted from the book.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405864 is a reply to message #405854] Sun, 21 February 2021 07:12 Go to previous messageGo to next message
Anonymous
Karma:
Originally posted by: Andy Walker

On 21/02/2021 01:26, Robin Vowels wrote [to Bill Findlay]:
> You've already ignored "The Great Computers Scandal"
> in 1965 involving SEVEN repeat SEVEN KDF9 installations at
> SEVEN repeat SEVEN universities, and that Manchester University
> had already replaced its KDF9 with an IBM machine.
Well, there's something wrong there. Manchester University
in 1965 was running Atlas, preceded by Mercury; they never had a
KDF9. Of the universities running KDF9s delivered in 1965 or earlier
[Birmingham, Glasgow, Leeds, Liverpool, Newcastle, Nottingham, Oxford,
Salford and Sydney], all were still running until at least December
1971. I don't know whether any were replaced by IBM machines; the
three I know about were replaced by ICL 1900-series computers, there
being a strong steer in that direction from government.

Personally, I preferred Atlas [who wouldn't!], then by some
measures the most powerful computer in the world, but KDF9 was more
reliable, and served well.

What was this "scandal"?

--
Andy Walker, Nottingham.
Andy's music pages: www.cuboid.me.uk/andy/Music
Composer of the day: www.cuboid.me.uk/andy/Music/Composers/Liszt
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405873 is a reply to message #405864] Sun, 21 February 2021 12:06 Go to previous messageGo to next message
Quadibloc is currently offline  Quadibloc
Messages: 4399
Registered: June 2012
Karma: 0
Senior Member
On Sunday, February 21, 2021 at 5:12:56 AM UTC-7, Andy Walker wrote:

> What was this "scandal"?

A Google search shows that BBC 2 aired, on March 3, 1965, an episode
of Horizon titled "The Great Computer Scandal".

John Savard
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405883 is a reply to message #405864] Sun, 21 February 2021 21:10 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Sunday, February 21, 2021 at 11:12:56 PM UTC+11, Andy Walker wrote:
> On 21/02/2021 01:26, Robin Vowels wrote [to Bill Findlay]:
>> You've already ignored "The Great Computers Scandal"
>> in 1965 involving SEVEN repeat SEVEN KDF9 installations at
>> SEVEN repeat SEVEN universities, and that Manchester University
>> had already replaced its KDF9 with an IBM machine.
> Well, there's something wrong there. Manchester University
> in 1965 was running Atlas, preceded by Mercury; they never had a
> KDF9. Of the universities running KDF9s delivered in 1965 or earlier
> [Birmingham, Glasgow, Leeds, Liverpool, Newcastle, Nottingham, Oxford,
> Salford and Sydney], all were still running until at least December
> 1971. I don't know whether any were replaced by IBM machines; the
> three I know about were replaced by ICL 1900-series computers, there
> being a strong steer in that direction from government.
>
> Personally, I preferred Atlas [who wouldn't!], then by some
> measures the most powerful computer in the world, but KDF9 was more
> reliable, and served well.
>
> What was this "scandal"?
..
It was posted upthread. Here is an extract re Atlas:
..
But in the programme Dr. Bene-
dict Nixon, one of the experts
operating the Ferranti Atlas
machine which had been in-
stalled for London University's
computer unit at the cost of
£3.4 million, complained that it
broke down on an average every
fifteen minutes and that its cal-
culations were frequently inaccurate.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405884 is a reply to message #405864] Sun, 21 February 2021 21:26 Go to previous messageGo to next message
Robin Vowels is currently offline  Robin Vowels
Messages: 426
Registered: July 2012
Karma: 0
Senior Member
On Sunday, February 21, 2021 at 11:12:56 PM UTC+11, Andy Walker wrote:
> On 21/02/2021 01:26, Robin Vowels wrote [to Bill Findlay]:
>> You've already ignored "The Great Computers Scandal"
>> in 1965 involving SEVEN repeat SEVEN KDF9 installations at
>> SEVEN repeat SEVEN universities, and that Manchester University
>> had already replaced its KDF9 with an IBM machine.
..
> Well, there's something wrong there. Manchester University
..
A copying error. Should have read "Birmingham University".
Birmingham Uni replaced its KDF9 with an IBM machine.
See upthread for the full original text.
..
> in 1965 was running Atlas, preceded by Mercury; they never had a
> KDF9. Of the universities running KDF9s delivered in 1965 or earlier
> [Birmingham, Glasgow, Leeds, Liverpool, Newcastle, Nottingham, Oxford,
> Salford and Sydney], all were still running until at least December
> 1971. I don't know whether any were replaced by IBM machines;
..
Birmingham Uni's KDF9 was replaced by an IBM computer.
..
> the three I know about were replaced by ICL 1900-series computers, there
> being a strong steer in that direction from government.
Re: where did RISC come from, Holy wars of the past - how did they turn out? [message #405892 is a reply to message #405873] Mon, 22 February 2021 07:06 Go to previous messageGo to previous message
Quadibloc is currently offline  Quadibloc
Messages: 4399
Registered: June 2012
Karma: 0
Senior Member
On Sunday, February 21, 2021 at 10:06:53 AM UTC-7, Quadibloc wrote:
> On Sunday, February 21, 2021 at 5:12:56 AM UTC-7, Andy Walker wrote:

>> What was this "scandal"?

> A Google search shows that BBC 2 aired, on March 3, 1965, an episode
> of Horizon titled "The Great Computer Scandal".

I couldn't find more about what this program contained, so I'm glad that
Robin Vowels has helped to clarify this.

One thing I did find, though, was that the Flowers report was released the
next year, in 1966. This report noted that Britain was very much behind the
U.S. in computing; it proposed more funding to Universities, and, to help
ensure efficient use of the funding, part of it would go to setting up computer
centers shared by several universities, allowing for the purchase of larger
computers so that the efficiencies of scale (Grosch's Law) could be taken
advantage of.

John Savard
Pages (8): [ «    1  2  3  4  5  6  7  8    »]  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: F11 was a DNS side job
Next Topic: Come to: http : / / mybeautifulworld . loveslife . biz . I am waiting ..........
Goto Forum:
  

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Thu Mar 28 22:11:07 EDT 2024

Total time taken to generate the page: 0.09043 seconds