On 20/06/2013 11:22, polygonum wrote:
My only exposure to VME/K things was when documentation referred to it
and, later, when K was rolled up into VME2900 or whatever actually
happened.

It was canned. Killed completely and not kept at all. Which was a pity,
because it was going from the time Bob knew it wehre it fell into a heap
all the time (and should have been killed) to a half-decent O/S with an
order book bigger than the installed base.

BTW Bob, do you recall which version you used?

Rumours reached my ears that everyone got shoved over the B, with some
free memory to ease the pain. The story given to mgmt was I heard that
you couldn't convert some of the B users to K, because they had things
like CAFS and DAP. But you could convert all the K users with a megabyte
of memory and a disc.

Rumour was it turned out to be double the memory (which wasn't always a
meg) and 10% on the CPU. Which wasn't always possible.

My first job was K development. I suppose it's like that thing of you
never forget your first girlfriend!

Andy

Vir Campestris wrote:

On 20/06/2013 16:37, Tim Watts wrote:
Perl has a well defined and elegant syntax. "Compact" != "Bad"

Did you ever see APL?

Yes; I spent summer 1981 (before my last Uni year) working for IBM in
Edinburgh, doing some programming. I can't remember what they wanted me to
do in APL, but as well as that - to make it all much easier to do - I wrote
a basic full-screen editor (like a very simple Xedit) for APL function
definitions.

I also converted a FORTRAN program from one dialect to another so that a
potential customer could see how fast (or not) their program would run on a
particular IBM machine - I think maybe a 4341.

Harking back to an earlier topic here, I also had the good fortune to meet
Peter Brown, but only once, when he gave a guest lecture in the Comp Sci
dept at St Andrews Uni. I always liked his 'Writing Interactive Compilers
and Interpreters' book.

--
Jeremy C B Nicoll - my opinions are my own.

Email sent to my from-address will be deleted. Instead, please reply
to replacing "aaa" by "284".

On Thu, 20 Jun 2013 21:09:57 +0100, Vir Campestris wrote:

On 20/06/2013 11:22, polygonum wrote:
My only exposure to VME/K things was when documentation referred to it
and, later, when K was rolled up into VME2900 or whatever actually
happened.

It was canned. Killed completely and not kept at all. Which was a pity,
because it was going from the time Bob knew it wehre it fell into a heap
all the time (and should have been killed) to a half-decent O/S with an
order book bigger than the installed base.

BTW Bob, do you recall which version you used?

SV12 was the last. The problem we had was that SV13 was going to have
major changes, which removed a lot of management features that we really
needed in a university environment. And it still didn't have a proper
backup system. We had after all been waiting over three years for a
reliable, fault tolerant system.

That's when we found an alternative. Which also had a proper backup
system! We managed to write a program to transfer the contents of the
filestore disk-to-disk, so the changeover was relatively painless.

Rumours reached my ears that everyone got shoved over the B, with some
free memory to ease the pain.

That's what I heard.

The story given to mgmt was I heard that
you couldn't convert some of the B users to K, because they had things
like CAFS and DAP. But you could convert all the K users with a megabyte
of memory and a disc.

Indeed. In fact K performed so badly for us that we were given free
upgrades of eight EDS100s to eight EDS200s, and 1MB to 2MB of memory. And
(for some reason) a second GPC.

It also didn't have any real networking support apart from a terminal
server. Not that we could afford anyway.

My first job was K development. I suppose it's like that thing of you
never forget your first girlfriend!

I'll just...never forget it! I was in charge of testing K when it
arrived, and sometimes I'd submit 200 fault reports in a week.

I also got to mastermind the changeover to the replacement operating
system eventually.

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Thu, 20 Jun 2013 21:25:00 +0100, Jeremy Nicoll - news posts wrote:

Harking back to an earlier topic here, I also had the good fortune to
meet Peter Brown, but only once, when he gave a guest lecture in the
Comp Sci dept at St Andrews Uni. I always liked his 'Writing
Interactive Compilers and Interpreters' book.

I think I got a credit at the start of that one. He also managed to get a
dig at Heather into that one...he thanked her for prooof reading it, and
said that the errors had been burned away by the acid dripping from her
tongue...
--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

In article ,
Bob Eager wrote:
Interactive Compilers and Interpreters' book.

I think I got a credit at the start of that one. He also managed to get a
dig at Heather into that one...he thanked her for prooof reading it, and
said that the errors had been burned away by the acid dripping from her
tongue...


Hmmm... pretty sure he had that in the Unix book. Along with the dedication
"To H, half hard, half soft" or similar.

Still got a copy somewhere in the office IIRC, will hunt tomorrow

I still remember the day we had two lectures following one another. First
was Heather lecturing on SGML (I think) and that was followed by Peter
about Guide. Heather had almost lost her voice and gave up half way through.

We all arrived early at Peters lecture and when we told him why we were
early he decided to head home early to get the most of the peace and quiet.

Darren

On Thu, 20 Jun 2013 20:56:12 +0000, D.M.Chapman wrote:

In article ,
Bob Eager wrote:
Interactive Compilers and Interpreters' book.

I think I got a credit at the start of that one. He also managed to get
a dig at Heather into that one...he thanked her for prooof reading it,
and said that the errors had been burned away by the acid dripping from
her tongue...

Hmmm... pretty sure he had that in the Unix book. Along with the
dedication "To H, half hard, half soft" or similar.

Still got a copy somewhere in the office IIRC, will hunt tomorrow

Yes, an insult in every book! Although the macro processors one was
dedicated to "Heather of the West" (I have it here on the shelf).

I still remember the day we had two lectures following one another.
First was Heather lecturing on SGML (I think) and that was followed by
Peter about Guide. Heather had almost lost her voice and gave up half
way through.

We all arrived early at Peters lecture and when we told him why we were
early he decided to head home early to get the most of the peace and
quiet.

LOL!


--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On 20/06/2013 21:49, Bob Eager wrote:
SV12 was the last. The problem we had was that SV13 was going to have
major changes, which removed a lot of management features that we really
needed in a university environment. And it still didn't have a proper
backup system. We had after all been waiting over three years for a
reliable, fault tolerant system.

That's what I suspected. I started just before the release of SV16, and
IIRC we were up to 18.50 by the time it was killed.

Backup? Ah, I owned the backup! I have two major memories of it. The
first was standing in front of a tape deck smiling as I watched the tape
running through at nearly device speed, as I had just got the ring
buffering working. One of the operators said to me "Looks like you've
got a runaway there". I said nothing, and just pointed at the write
light

The other wasn't long afterwards. We had a ... ah... better not ID
them... customer with a large ISAM file of critical information on it.
They'd copied it to tape with my new code, cleaned up the discs, and
when copied back it was broken. At first sight the file was complete
rubbish - no-one could find any real data, it was full of overflow
blocks. It turned out I'd made a mistake in the end-of-tape handling,
and got the buffers in the ring mixed up. On restore, not backup,
luckily. And the overflow blocks? Yes, they were real. We advised them
to do a record-level copy of the file to clean them up.

Andy

On Thu, 20 Jun 2013 21:25:00 +0100, Jeremy Nicoll - news posts
wrote:

.... I always liked his 'Writing Interactive Compilers
and Interpreters' book.

Still have my copy. £5.95 from the UKC bookshop. Still in almost as new condition (I have
read it!)



--
Regards, Paul Herber, Sandrila Ltd.
http://www.sandrila.co.uk/ twitter: @sandrilaLtd

On 20 Jun 2013 20:51:41 GMT, Bob Eager wrote:

On Thu, 20 Jun 2013 21:25:00 +0100, Jeremy Nicoll - news posts wrote:

Harking back to an earlier topic here, I also had the good fortune to
meet Peter Brown, but only once, when he gave a guest lecture in the
Comp Sci dept at St Andrews Uni. I always liked his 'Writing
Interactive Compilers and Interpreters' book.

I think I got a credit at the start of that one. He also managed to get a
dig at Heather into that one...he thanked her for prooof reading it, and
said that the errors had been burned away by the acid dripping from her
tongue...

yes, first in the list of acknowledgements.
"Above all, Heather Brown deserves appreciation. She has devoted hundreds of hours to
reading and checking this book, and has criticized every draft with acid severity."



--
Regards, Paul Herber, Sandrila Ltd.
http://www.sandrila.co.uk/ twitter: @sandrilaLtd

On 20/06/2013 21:07, Bob Eager wrote:
On Thu, 20 Jun 2013 20:54:04 +0100, John Rumm wrote:

On 20/06/2013 17:10, Bob Eager wrote:
On Thu, 20 Jun 2013 14:39:05 +0100, John Rumm wrote:

On 20/06/2013 10:47, Tim Streater wrote:
In article ,
Bob Eager wrote:

On Thu, 20 Jun 2013 09:58:28 +0100, polygonum wrote:

On 20/06/2013 09:47, Huge wrote:
On 2013-06-20, Gordon Henderson wrote:

And while I do enjoy a bit of "retro" computing, I have no
yearning whatsoever to go back and write another BCPL program.

That's how I feel about the PDP-11 programming job that started
this thread.

Mind you, I was getting enquiries for MUMPS contracts until quite
recently. (I installed the Linux MUMPS port recently, out of idle
curiousity, and now wonder why I liked it...)

BTW, I emailed the guy to (politely) express my incredulity, but I
haven't had an answer.

I enjoyed my time using the VME/B SCL language. Would be very happy
to write some more things in it! Limited in some ways but that made
the challenges interesting. Plus a few bits and pieces in Message
Text Modules... And the odd extra functionality crafted in Cobol.
:-)

I never got to use SCL because we had VME/K (one of the few...). But
we dumped VME/K in favour of a third party operating system, after
three years.

Ever tried REXX? IBM's rough equivalent. I use it a lot.

Never really liked it. These days I use PHP and JavaScript. It's far
easier with those - no types, really, and some amount of string
handling.

Never liked weakly typed language myself... ;-)

BCPL was untyped!

Not too surprising when you look at how the original K&R style C turned
out... did not like that much either until it at least got as far as
ANSI and had typed parameters ;-)

Surely it had typed parameters before that....just not in the actual
parameter list.

Well they were typed when you passed em in, and typed when you got them
out, only no one was checking they actually matched ;-)

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On Thu, 20 Jun 2013 22:52:39 +0100, Vir Campestris wrote:

On 20/06/2013 21:49, Bob Eager wrote:
SV12 was the last. The problem we had was that SV13 was going to have
major changes, which removed a lot of management features that we
really needed in a university environment. And it still didn't have a
proper backup system. We had after all been waiting over three years
for a reliable, fault tolerant system.

That's what I suspected. I started just before the release of SV16, and
IIRC we were up to 18.50 by the time it was killed.

Backup? Ah, I owned the backup!

We had to do a LD() to get a list of files, then edit it with a script to
separate out the filenames. Then feed to to a disk-to-tape copy program.
Gruesome.

The backup system on the replacement was great. The tape cycle was
automated, and if you had to restore...it worked out which bits of which
tapes had to be restored (there were incrementals in there) and just told
the operators which tapes to load on any available deck.

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Thu, 20 Jun 2013 22:57:40 +0100, Paul Herber wrote:

On Thu, 20 Jun 2013 21:25:00 +0100, Jeremy Nicoll - news posts
wrote:

.... I always liked his 'Writing Interactive Compilers
and Interpreters' book.

Still have my copy. £5.95 from the UKC bookshop. Still in almost as new
condition (I have read it!)

Mine is at work. Together with a more advanced (but still very practical)
book by Richard Bornat.
--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Thu, 20 Jun 2013 23:18:00 +0100, Paul Herber wrote:

On 20 Jun 2013 20:51:41 GMT, Bob Eager wrote:

On Thu, 20 Jun 2013 21:25:00 +0100, Jeremy Nicoll - news posts wrote:

Harking back to an earlier topic here, I also had the good fortune to
meet Peter Brown, but only once, when he gave a guest lecture in the
Comp Sci dept at St Andrews Uni. I always liked his 'Writing
Interactive Compilers and Interpreters' book.

I think I got a credit at the start of that one. He also managed to get
a dig at Heather into that one...he thanked her for prooof reading it,
and said that the errors had been burned away by the acid dripping from
her tongue...

yes, first in the list of acknowledgements.
"Above all, Heather Brown deserves appreciation. She has devoted
hundreds of hours to reading and checking this book, and has criticized
every draft with acid severity."

Aha. As I said, mine was at work...!

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Thu, 20 Jun 2013 23:21:36 +0100, John Rumm wrote:

On 20/06/2013 21:07, Bob Eager wrote:
On Thu, 20 Jun 2013 20:54:04 +0100, John Rumm wrote:

On 20/06/2013 17:10, Bob Eager wrote:
On Thu, 20 Jun 2013 14:39:05 +0100, John Rumm wrote:

On 20/06/2013 10:47, Tim Streater wrote:
In article ,
Bob Eager wrote:

On Thu, 20 Jun 2013 09:58:28 +0100, polygonum wrote:

On 20/06/2013 09:47, Huge wrote:
On 2013-06-20, Gordon Henderson
wrote:

And while I do enjoy a bit of "retro" computing, I have no
yearning whatsoever to go back and write another BCPL program.

That's how I feel about the PDP-11 programming job that started
this thread.

Mind you, I was getting enquiries for MUMPS contracts until
quite recently. (I installed the Linux MUMPS port recently, out
of idle curiousity, and now wonder why I liked it...)

BTW, I emailed the guy to (politely) express my incredulity, but
I haven't had an answer.

I enjoyed my time using the VME/B SCL language. Would be very
happy to write some more things in it! Limited in some ways but
that made the challenges interesting. Plus a few bits and pieces
in Message Text Modules... And the odd extra functionality
crafted in Cobol. :-)

I never got to use SCL because we had VME/K (one of the few...).
But we dumped VME/K in favour of a third party operating system,
after three years.

Ever tried REXX? IBM's rough equivalent. I use it a lot.

Never really liked it. These days I use PHP and JavaScript. It's
far easier with those - no types, really, and some amount of string
handling.

Never liked weakly typed language myself... ;-)

BCPL was untyped!

Not too surprising when you look at how the original K&R style C
turned out... did not like that much either until it at least got as
far as ANSI and had typed parameters ;-)

Surely it had typed parameters before that....just not in the actual
parameter list.

Well they were typed when you passed em in, and typed when you got them
out, only no one was checking they actually matched ;-)

That's true. As a previous BCPL programmer, I didn't care!

But in BCPL you could even reassign the value of a label, so that it
didn't actually label the code it was next to...



--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

in 1236056 20130620 110343 Bob Eager wrote:
On Thu, 20 Jun 2013 10:47:40 +0100, Tim Streater wrote:

In article ,
Bob Eager wrote:

On Thu, 20 Jun 2013 09:58:28 +0100, polygonum wrote:

On 20/06/2013 09:47, Huge wrote:
On 2013-06-20, Gordon Henderson wrote:

And while I do enjoy a bit of "retro" computing, I have no yearning
whatsoever to go back and write another BCPL program.

That's how I feel about the PDP-11 programming job that started this
thread.

Mind you, I was getting enquiries for MUMPS contracts until quite
recently. (I installed the Linux MUMPS port recently, out of idle
curiousity, and now wonder why I liked it...)

BTW, I emailed the guy to (politely) express my incredulity, but I
haven't had an answer.

I enjoyed my time using the VME/B SCL language. Would be very happy
to write some more things in it! Limited in some ways but that made
the challenges interesting. Plus a few bits and pieces in Message
Text Modules... And the odd extra functionality crafted in Cobol. :-)

I never got to use SCL because we had VME/K (one of the few...). But we
dumped VME/K in favour of a third party operating system, after three
years.

Ever tried REXX? IBM's rough equivalent. I use it a lot.

Never really liked it. These days I use PHP and JavaScript. It's far
easier with those - no types, really, and some amount of string
handling.

I use it for command scripts, not web stuff!

My websites are 100% Rexx (the ooRexx version).

in 1236103 20130620 144957 John Rumm wrote:
On 20/06/2013 12:09, Bob Eager wrote:

I started with BASIC. Lost count of the number of different assemblers.

I remember having a slightly strange conversation with with an IBM type
at a recruitment exhibition once many years ago... they had a big sign
up saying Assembler programmers wanted. So I thought I would find out a
bit more, gave them a CV to look at, and after a few minutes perusal,
she said "but I can't see any assembler experience here?". So I directed
her at the section titled "Low level languages" and a moderately long
list of processor architectures. Realisation slowly dawned that to IBM
people "Assembler" was an IBM specific technology, and is apparently the
only low level language that exists - she had absolutely no concept that
assembler was a generic term, or that the concepts of machine languages
were similar across platforms or even it seems, that "other" computers
even had machine languages. ;-)

It's unfair to imply that she was a typical IBMer, and also naive of you to assume
that IBM only had/has one processor architecture.
In my almost 40 years with IBM I programmed at least a dozen IBM processor
types (1401, 1410, 1130, 1800, 360/370, system/3 etc as well as 3 or 4 support
microprocessors) and all the Intel, Motorola and Texas micros.

On Friday 21 June 2013 08:26 Huge wrote in uk.d-i-y:


http://me.veekun.com/blog/2012/04/09...of-bad-design/

Good grief.



This is my favourite quote:

=====
Now imagine you meet millions of carpenters using this toolbox who tell you
€œwell hey whats the problem with these tools? Theyre all Ive ever used
and they work fine!€� And the carpenters show you the houses theyve built,
where every room is a pentagon and the roof is upside-down. And you knock on
the front door and it just collapses inwards and they all yell at you for
breaking their door.
=====

--
Tim Watts Personal Blog: http://squiddy.blog.dionic.net/

http://www.sensorly.com/ Crowd mapping of 2G/3G/4G mobile signal coverage

Reading this on the web? See:
http://wiki.diyfaq.org.uk/index.php?title=Usenet

On 21/06/13 08:26, Huge wrote:
On 2013-06-20, Tim Watts wrote:
On Thursday 20 June 2013 15:26 The Natural Philosopher wrote in uk.d-i-y:

On 20/06/13 15:14, Huge wrote:
On 2013-06-20, Tim Streater wrote:
In article ,
(D.M.Chapman) wrote:

In article ,
Tim Streater wrote:

I use PHP for command scripts, too. Shell stuff is for the birds.
Urghhh! Perl is the correct answer for that stuff (with a bit of bash
at times)
No, perl is even worse than bash etc. Perl's designers need a good
smack.
That would be an ecumenical matter.

But anyone who thinks PHP is a good idea is on pretty dodgy ground. )


Id say its better than perl or bash...
Thanks for that. Now I know PHP sucks copiously.

http://me.veekun.com/blog/2012/04/09...of-bad-design/
Good grief.

The error he makes is to think that PHP is a programming language.
Its not. It is and always was a quick way to get an 'active web page'
up and running.

If any PHP 'program' I write is more than a couple of pages, or is
anything more than a basic way to interrogate a database and present the
results onscreen, I consider its the wrong tool and I reach for the c
compiler.

I feel the same way about bash.
Python looks interesting, but I don't have the time or inclination to
learn it.

It may be a hammer with two claws and no head, but if all you need a
hammer for is extracting nails, that's just brilliant.

I agree its error handling is dire.

YOu would have thoughtte fillowing would be just fine

function get_text($dir)
{
$filename=$dir."/readme.txt";
$text=file_get_contents($filename);
if($text) return $text;
else return "";
}

I.e. I want to find if a readme.txt file exists in a given directory,
and if so display its contents.

That,in C would be find. To open a nonextstent file is simply a quick
way to see if you can. And return a poiinter to it, or NULL if forANY
resasoin it fails.

PHP litters my logs with 'error: file doesn't exist' messages.

One day I will work out how to stop that. In a century or so.

But hey, its popular, there's a lot written in it, plenty of library
support and its not the worst 'language;' i've ever had to learn.
JavaScript takes that award.



--
Ineptocracy

(in-ep-toc-ra-cy) €“ a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishing number of producers.

On 21/06/13 10:16, Tim Streater wrote:
I
Hmmm, what a content-free site. He whinges about a lot of stuff I
never use, so it's just a big yawn.

+1.
He went looking for Pascal and found PEEK POKE and BASIC
Big deal.

It reminds me of that plaintive Edgar Broughton song 'what is a woman for?'

A question I shall probably take unanswered to my grave.

What in fact is PHP for? Does it actually make the job of constructing
YOUR website easier, or harder?

--
Ineptocracy

(in-ep-toc-ra-cy) €“ a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishing number of producers.

On 21/06/13 11:17, Tim Streater wrote:
In article ,
The Natural Philosopher wrote:

On 21/06/13 08:26, Huge wrote:
On 2013-06-20, Tim Watts wrote:
On Thursday 20 June 2013 15:26 The Natural Philosopher wrote in
uk.d-i-y:

On 20/06/13 15:14, Huge wrote:
On 2013-06-20, Tim Streater wrote:
In article ,
(D.M.Chapman) wrote:

In article
,
Tim Streater wrote:

I use PHP for command scripts, too. Shell stuff is for the
birds.
Urghhh! Perl is the correct answer for that stuff (with a bit
of bash
at times)
No, perl is even worse than bash etc. Perl's designers need a good
smack.
That would be an ecumenical matter.

But anyone who thinks PHP is a good idea is on pretty dodgy
ground. )


Id say its better than perl or bash...
Thanks for that. Now I know PHP sucks copiously.

http://me.veekun.com/blog/2012/04/09...of-bad-design/
Good grief.

The error he makes is to think that PHP is a programming language.
Its not. It is and always was a quick way to get an 'active web
page' up and running.

If any PHP 'program' I write is more than a couple of pages, or is
anything more than a basic way to interrogate a database and present
the results onscreen, I consider its the wrong tool and I reach for
the c compiler.

Well I used to know C, and wrote a largish stats gathering and command
multiplexer in it 25 years ago. But these days I can't be arsed to
worry about memory allocation and string handling. So it's PHP every
time, even to the extent of writing a bayesian spam filter in it.

I find string handling WORSE with PHP, than C actually. Sure the easy
things are easier with PHP., but take the case where you want to spilt
string into little strings.

In C, you can just assign a pointer to the start of each bit, and
replace the separator with a null byte, and without any extra memory
being used, that's how you split it up.

Its a bloody nightmare in PHP..and you end up with a huge amount of
redundant memory used up in string fragments you no longer need. You
cannot alter a string in PHP easily.

As far as memory allocation goes well I do it the old fashioned way. 99
times out of 100 I know the size, so its allocated at compile time. I
try not to use malloc/free.

And if its a small program that runs and exist, who need free() anyway?
Just allocate,use, and then close the program!

Ive alyws been amusied by people who can't get their heads round
pointers or typing, but can get their heads round operator overloading
and constructors and destructors.




--
Ineptocracy

(in-ep-toc-ra-cy) €“ a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishing number of producers.

On 21/06/13 11:19, Tim Streater wrote:
In article ,
The Natural Philosopher wrote:

On 21/06/13 10:16, Tim Streater wrote:
I
Hmmm, what a content-free site. He whinges about a lot of stuff I
never use, so it's just a big yawn.

+1.
He went looking for Pascal and found PEEK POKE and BASIC
Big deal.

It reminds me of that plaintive Edgar Broughton song 'what is a woman
for?'

A question I shall probably take unanswered to my grave.

What in fact is PHP for? Does it actually make the job of
constructing YOUR website easier, or harder?

Well oddly enough I only make minimal use of it on one website.

In which case there is probably a better language for you. I'd guess at
Python.


"While offering choice in coding methodology, the Python philosophy
rejects exuberant syntax, such as in Perl, in favor of a sparser,
less-cluttered grammar. As Alex Martelli put it: "To describe something
as clever is not considered a compliment in the Python culture."[28]
Python's philosophy rejects the Perl "there is more than one way to do
it" approach to language design in favor of "there should be one€”and
preferably only one€”obvious way to do it".

http://en.wikipedia.org/wiki/Python_...ng_language%29

I think if I wanted to actually write complex scripts and do maths, Id
look at python.

I did briefly look at it when looking for a language to build websites
with, but settled on PHP because it was very widely understood and
recognised, and I have been dealing with weird programming tools all my
life :-)

--
Ineptocracy

(in-ep-toc-ra-cy) €“ a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishing number of producers.

dave wrote:

I want to put Mini UNIX on one of mine...the 11/23. I also have an 11/84,
11/03, and a Micro PDP-11 (deskside, if you have a bloody big desk).

Why do you want to run Mini-Unix?
We ran Unix versions 5 and 6 on a pdp-11/23 for several years,
with 7 users and a printer.
I always thought the 11/23 was the best computer ever made.

On 21/06/13 12:32, Tim Streater wrote:
In article ,
The Natural Philosopher wrote:

On 21/06/13 11:17, Tim Streater wrote:
In article ,
The Natural Philosopher wrote:

If any PHP 'program' I write is more than a couple of pages, or
is anything more than a basic way to interrogate a database and
present the results onscreen, I consider its the wrong tool and I
reach for the c compiler.

Well I used to know C, and wrote a largish stats gathering and
command multiplexer in it 25 years ago. But these days I can't be
arsed to worry about memory allocation and string handling. So it's
PHP every time, even to the extent of writing a bayesian spam
filter in it.

I find string handling WORSE with PHP, than C actually. Sure the easy
things are easier with PHP., but take the case where you want to
spilt string into little strings.

In C, you can just assign a pointer to the start of each bit, and
replace the separator with a null byte, and without any extra memory
being used, that's how you split it up.

Its a bloody nightmare in PHP..and you end up with a huge amount of
redundant memory used up in string fragments you no longer need. You
cannot alter a string in PHP easily.

As far as memory allocation goes well I do it the old fashioned way.
99 times out of 100 I know the size, so its allocated at compile
time. I try not to use malloc/free.

And if its a small program that runs and exist, who need free()
anyway? Just allocate,use, and then close the program!

The program I was referring to was in the manner of a daemon. Running
under VMS on a MicroVAX 2, It typically ran for 9 months at a time,
and was only shut down then because of the site wide power outage
while they did maintenance on the 132kV line supplying power to the
site (they had to switch over to the backup 64kV line).

well daemons as opposed to cron invokedprograms do need to be careful
about memory allocation indeed.

And they had better not segfault. :-0)



--
Ineptocracy

(in-ep-toc-ra-cy) €“ a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishing number of producers.

On Fri, 21 Jun 2013 11:57:15 +0000, Huge wrote:

On 2013-06-20, Bob Minchin
wrote:
Huge wrote:
On 2013-06-18, Bob Minchin
wrote:

Mid 70s saw me programming PDP11 with ML/1 studying under its Author
Peter Brown and his wife Heather at University of Kent.
I'll still have my notes - "somewhere"

I'm not sure if we've had this conversation before, but we may have
been contemporaneous at UKC - I graduated in 1975 and spent far too
much time in the Computing Lab and nothing like enough in the Biology
Lab, which accounts for both the class of my degree and my current
career. Prof.
Brown took my undergrad computing course, which was optional and
designed for people who already knew how to program. (I'd learned
FORTRAN at school.)


I was there from 72-75 as an undergraduate.

The same time, then.

I was up the other end of the Natural Sciences intake spectrum. I
initially signed up for Electronics

Did you know John Scherrer? He was doing electronics at the same time as
you were. (Well done to his son, Jonathan, who has just got a First in
Computer Science at UKC!)

Indeed he has....as I heard last week (decision day was last Friday).

John later became our site engineer for the ICL 2960 mainframe.

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Fri, 21 Jun 2013 12:49:01 +0100, dave wrote:

On 19 Jun 2013 22:22:05 GMT, Bob Eager wrote:


I used to write assembler on those, then wrote an assembler, a
disassembler, an emulator and a small Basic interpreter. Still toying
with it from time to time. Somebody actually emailed me to thank me
for writing the assembler and suggesting a couple of tweeks.

JGH

Wrote in assembler on PDP11/10... 34. BASIC and FORTRAN and RATFOR.
Wrote libs for FORTRAN to call assem. Real Time RT-11 user
applications AR11 i/o. Also ran "mini" unix, C compiler etc (Not at
same time of course :-)

I want to put Mini UNIX on one of mine...the 11/23. I also have an
11/84,
11/03, and a Micro PDP-11 (deskside, if you have a bloody big desk).

Do hope you get mini unix up 'n running. I still have the letter from
Brian Kernighan during the days of paper mail (:-)) where he gave me
contact info. for getting the software (academic use). Alas I do not
have the magtape it was distributed on... but I guess is avaiable online
- unless Bell Labs say no of course :-)

I have the magtape image here.

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Fri, 21 Jun 2013 15:32:10 +0200, Timothy Murphy wrote:

dave wrote:

I want to put Mini UNIX on one of mine...the 11/23. I also have an
11/84,
11/03, and a Micro PDP-11 (deskside, if you have a bloody big desk).

Why do you want to run Mini-Unix?
We ran Unix versions 5 and 6 on a pdp-11/23 for several years,
with 7 users and a printer.
I always thought the 11/23 was the best computer ever made.

Mine doesn't have the memory management option chip.
--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On 21/06/2013 08:07, Bob Martin wrote:
in 1236103 20130620 144957 John Rumm wrote:
On 20/06/2013 12:09, Bob Eager wrote:

I started with BASIC. Lost count of the number of different assemblers.

I remember having a slightly strange conversation with with an IBM type
at a recruitment exhibition once many years ago... they had a big sign
up saying Assembler programmers wanted. So I thought I would find out a
bit more, gave them a CV to look at, and after a few minutes perusal,
she said "but I can't see any assembler experience here?". So I directed
her at the section titled "Low level languages" and a moderately long
list of processor architectures. Realisation slowly dawned that to IBM
people "Assembler" was an IBM specific technology, and is apparently the
only low level language that exists - she had absolutely no concept that
assembler was a generic term, or that the concepts of machine languages
were similar across platforms or even it seems, that "other" computers
even had machine languages. ;-)

It's unfair to imply that she was a typical IBMer,

At the time I assumed nothing - however having met many more like her
since I concluded (rightly or wrongly) that there was a certain isular
nature to some of their people's experience.

and also naive of you to assume
that IBM only had/has one processor architecture.

I don't think that I assumed that then, or since. I fully appreciate
that they have different platforms, which is partly why it struck me as
odd that this lass treated "Assembler" as is it were one and only one
IBM specific thing.

In my almost 40 years with IBM I programmed at least a dozen IBM processor
types (1401, 1410, 1130, 1800, 360/370, system/3 etc as well as 3 or 4 support
microprocessors) and all the Intel, Motorola and Texas micros.

Indeed, as I attempted to highlight to her, once you have encountered a
variety of low level languages, learning a new one is no big deal.

(learning the architecture that goes with it can be far more of a big
deal however!)

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

dave wrote:

Do hope you get mini unix up 'n running. I still have the letter from
Brian Kernighan ...

I've got a cheque for $2.56 from Donald Knuth for finding some mistakes in
one of his text books.

--
Jeremy C B Nicoll - my opinions are my own.

Email sent to my from-address will be deleted. Instead, please reply
to replacing "aaa" by "284".

John Rumm wrote:

Indeed, as I attempted to highlight to her, once you have encountered a
variety of low level languages, learning a new one is no big deal.

(learning the architecture that goes with it can be far more of a big deal
however!)

Yes, and the macros etc relevant to whatever you're writing the code for. I
worked as an MVS systems programmer for somme years, and learning enough
assembler to write decent code didn't take all that long. But reading the
relevant parts of the supervisor state macro manuals, and the JES2 manuals
etc and (when IBM still allowed customers to see it) looking at the source
for bits of the OS that one intended to alter.... that could take ages just
to understand enough.

--
Jeremy C B Nicoll - my opinions are my own.

Email sent to my from-address will be deleted. Instead, please reply
to replacing "aaa" by "284".

On Fri, 21 Jun 2013 23:16:59 +0100, Jeremy Nicoll - news posts wrote:

dave wrote:

Do hope you get mini unix up 'n running. I still have the letter from
Brian Kernighan ...

I've got a cheque for $2.56 from Donald Knuth for finding some mistakes
in one of his text books.

Nothing like that, but I once met Edsger Dijkstra!

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Fri, 21 Jun 2013 12:49:01 +0100, dave wrote:

Do hope you get mini unix up 'n running

I did get it running on a PDP-11/20 once...

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

Bob Martin wrote:

My websites are 100% Rexx (the ooRexx version).

Before I got ill and had to stop work, but after my MVS sysprog days, I ran
a small programming team that wrote ops automation (in NetView, SA/390 etc),
almost all written in REXX (and some Assembler). Now on my home computer I
too use ooREXX for the small number of programs I write at home.

--
Jeremy C B Nicoll - my opinions are my own.

Email sent to my from-address will be deleted. Instead, please reply
to replacing "aaa" by "284".

"Jeremy Nicoll - news posts" wrote
in message nvalid...
John Rumm wrote:

Indeed, as I attempted to highlight to her, once you have encountered a
variety of low level languages, learning a new one is no big deal.

(learning the architecture that goes with it can be far more of a big
deal
however!)

Yes, and the macros etc relevant to whatever you're writing the code for.
I
worked as an MVS systems programmer for somme years, and learning enough
assembler to write decent code didn't take all that long. But reading the
relevant parts of the supervisor state macro manuals, and the JES2 manuals
etc and (when IBM still allowed customers to see it) looking at the source
for bits of the OS that one intended to alter.... that could take ages
just
to understand enough.

What I always fancied was a disassembler which added comments

On 21/06/2013 23:34, bm wrote:
"Jeremy Nicoll - news posts" wrote
in message nvalid...
John Rumm wrote:

Indeed, as I attempted to highlight to her, once you have encountered a
variety of low level languages, learning a new one is no big deal.

(learning the architecture that goes with it can be far more of a big
deal
however!)

Yes, and the macros etc relevant to whatever you're writing the code for.
I
worked as an MVS systems programmer for somme years, and learning enough
assembler to write decent code didn't take all that long. But reading the
relevant parts of the supervisor state macro manuals, and the JES2 manuals
etc and (when IBM still allowed customers to see it) looking at the source
for bits of the OS that one intended to alter.... that could take ages
just
to understand enough.

What I always fancied was a disassembler which added comments

That would be easy... making the comments relevant, more difficult ;-)


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 21/06/2013 23:27, Bob Eager wrote:
On Fri, 21 Jun 2013 23:16:59 +0100, Jeremy Nicoll - news posts wrote:

dave wrote:

Do hope you get mini unix up 'n running. I still have the letter from
Brian Kernighan ...

I've got a cheque for $2.56 from Donald Knuth for finding some mistakes
in one of his text books.

Nothing like that, but I once met Edsger Dijkstra!

And did you ask him how he pronounces his name? ;-)


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On Sat, 22 Jun 2013 03:14:28 +0100, John Rumm wrote:

On 21/06/2013 23:27, Bob Eager wrote:
On Fri, 21 Jun 2013 23:16:59 +0100, Jeremy Nicoll - news posts wrote:

dave wrote:

Do hope you get mini unix up 'n running. I still have the letter from
Brian Kernighan ...

I've got a cheque for $2.56 from Donald Knuth for finding some
mistakes in one of his text books.

Nothing like that, but I once met Edsger Dijkstra!

And did you ask him how he pronounces his name? ;-)

I have no need...! .-)

I say it many times each year in my lectures...

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On 22/06/2013 03:08, John Rumm wrote:
On 21/06/2013 23:34, bm wrote:


What I always fancied was a disassembler which added comments

That would be easy... making the comments relevant, more difficult ;-)


:-)

There are times it would be good to find a sentient programmer who could
manage that.

Way back in my VME SCL days, I wrote a bit of code which would convert
the source code into diagnostic records which were then stored in the
object modules produced by the compiler.

People had a tendency to write little bits of SCL then forget where the
source was, or not have access, or look at the wrong version. That even
happened with some of the production SCL in the early days. Before the
SCL compiler was introduced, the object code was pretty much the source
slightly re-hashed - so everyone had got used to the idea that you could
always list it out again. And this effectively re-introduced the facility.

Can't say the whole idea was mine, but I implemented it as bone fide
diagnostic records - not simply tacking the source records onto the end
of the object file!

--
Rod

Bob Eager wrote:
John later became our site engineer for the ICL 2960 mainframe.
Back in the day, the campus computing power was an ICL 4130 with a PDP11
bolted on the front running interactive BASIC to 8 (or was it 10)
Teletypes scattered across the campus.
The system was known as KOS - (Kent Online System) and I think a clone
of it was also run at Reading too.

Bob

According to Google, York also ran KOS

On Sat, 22 Jun 2013 13:42:26 +0100, Bob Minchin wrote:

Bob Eager wrote:
John later became our site engineer for the ICL 2960 mainframe.
Back in the day, the campus computing power was an ICL 4130 with a PDP11
bolted on the front running interactive BASIC to 8 (or was it 10)
Teletypes scattered across the campus.
The system was known as KOS - (Kent Online System) and I think a clone
of it was also run at Reading too.

Bob

According to Google, York also ran KOS

They did. And Lancaster, Aberystwyth, and a couple of others.

It was however developed at Kent by Peter Brown, Heather Brown, Steve
Binns and Brian Spratt. I started using it in January 1971, but by June I
had stolen the source code. I got the hardware specs from ICL over the
summer and hacked rather a lot of it the following year...

The first hack was setting a bit in the multiplexer status word for a
terminal - which logged it out, allowing someone else to use it!
--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor

On Sat, 22 Jun 2013 13:42:26 +0100, Bob Minchin wrote:

Bob Eager wrote:
John later became our site engineer for the ICL 2960 mainframe.
Back in the day, the campus computing power was an ICL 4130 with a PDP11
bolted on the front running interactive BASIC to 8 (or was it 10)
Teletypes scattered across the campus.
The system was known as KOS - (Kent Online System) and I think a clone
of it was also run at Reading too.

Bob

According to Google, York also ran KOS

Strangely, I have a KOS manual within reach of me right now...beside my
desk.

--
Use the BIG mirror service in the UK: http://www.mirrorservice.org
My posts (including this one) are my copyright and if @diy_forums on
Twitter wish to tweet them they can pay me £30 a post
*lightning surge protection* - a w_tom conductor