Microbee Computers are Back

[terryc]

T.T. wrote:

Hence my argument that it isn't sutable for teach yourself.
Bring back COBOL.

Is this a horses for course comment?
Do you have any figures on the current percentage implementation of Cobol?

 

[T.T.]

"terryc" <newsninespam-spam@woa.com.au> wrote in message
news:jb3n1s$6tv$2@dont-email.me...

T.T. wrote:

Hence my argument that it isn't sutable for teach yourself.
Bring back COBOL.

Is this a horses for course comment?
Do you have any figures on the current percentage implementation of Cobol?
If it is above zero, it wouldn't be by much.

COBOL was such a forgiving language that, in its day, thousands of
programmers who should have starved to death paid off mortgages and raised
families.

 

[swanny]

On 30/11/2011 9:39 AM, terryc wrote:

swanny wrote:
On 30/11/2011 1:28 AM, terryc wrote:
keithr wrote:
On 29/11/2011 6:58 PM, terryc wrote:

The problem is that the language is so cryptic, you have to be well
versed to spot a simple mstake. similarly, it makes it very difficult
for people to teach themselves by reading the code.


My 2c is that if the language lends itself to faulty implementations,
then it isn't such a great language.

The language is not cryptic,

So non-cryptic that no one spotted a simple sign problem in 18 months.

easier to work through the coding flow than something as horrible as
Basic.

Shrug, your last comment is what matters "coding flow", not the language.

I learnt a great deal by reading the Unix SysV source code and using the
K&R C book.

You learnt programmng or C?

Was already studying programming, but the course was in Pascal which I
wasn't a big fan of. There were a few sys admins and and real
programmers there (keeping the old Vax 11/780 alive) who I got to know
and undertook C programming myself (not part of the Uni course). Unix
was written in C so it was a good place to start reading code.
Surprisingly the code was very well structured and documented
throughout, making it easy to understand.

In any case, you and I are not the target

audience I was asking about. I've worked with the Joe blow who walks in
off the street to the local computer club and wants to learn
programming. with old PCs, it was an easy matter of popping a prompt and
editing a simple basic program that gave them a clue and they were off.
No such ability these days.

Agree with that. The GUI sort of killed off the 'Ready>' prompt of the
old Z80 machines.

Poor programmers lead to faulty implementations, not the language.

That applies to any language.

Lol, reminded me of a Phd candidate(middile aged), with a successful
commercial program under their belt, who said he had spent all weekend
changing A CONSTANT. BLINK!

Some changes from K&R C to ANSI C.

I learnt about global declarations the first week of uni by reading
comments in discarded programmes in the terminal room (early Dec stuff)
or the good old finger crunching ICL(?) teletype.

I remember the teletype machines and the old thermal printer terminals.
Wrote a program that took over them from a remote console, allowing
copious BEL and FF characters to be sent to them, surprising the first
year undergrads

 

[atec77]

On 30/11/2011 8:59 AM, T.T. wrote:

"terryc"<newsninespam-spam@woa.com.au> wrote in message
news:jb3n1s$6tv$2@dont-email.me...
T.T. wrote:

Hence my argument that it isn't sutable for teach yourself.
Bring back COBOL.

Is this a horses for course comment?
Do you have any figures on the current percentage implementation of Cobol?
If it is above zero, it wouldn't be by much.
COBOL was such a forgiving language that, in its day, thousands of
programmers who should have starved to death paid off mortgages and raised
families.


It's so easy even I can use it

--









X-No-Archive: Yes

 

[swanny]

On 29/11/2011 4:29 PM, Krypsis wrote:

On 29/11/2011 1:59 PM, kreed wrote:
On Nov 29, 10:49 am, terryc<newsninespam-s...@woa.com.au> wrote:
swanny wrote:
On 28/11/2011 9:27 PM, aussiblu wrote:
Seehttp://www.microbeetechnology.com.au/premiumpluskit.htm

The world has moved on from the early 80's. Can't imagine why anyone
would bother. Nostalgic dust collectors?

Is there anything else around that the average joe blow could get their
hands on and program? Has some interesting I/O option for projects.


The problem with this system is that if they are only making 100 of
them, there isn't going to be much of a community to support and
innovate with them, especially with the newer side of it (coldfire
processor and its linux system)

That is where I see the major limitation. 100 is a miniscule number but
that may be just a number dragged out of thin air in order to test the
water. If the unit is successful, they may continue with other
production runs though there is no mention of that on the website.



There is the Maximite that is the same sort of thing, but does not
have the coldfire processor (or similar) so is not Linux compatible or
colour display.

Don't need a Coldfire processor to be run uCLinux. There's even an Atmel
dev board that runs it and has a complete toolchain for development
(NGW100 board). Has two 10/100 network ports.

Not Z80 though, but not much is these days.

The good thing with this is that it is fully open source, a shitload
of them have been sold, built, and there is a solid community
developing with them. Also VERY cheap, especially if you use an old
keyboard/monitor with it.


Finally, if you come up with a project that you like and want to use
full time, you only have to buy another kit for about $70 and use it
for that dedicated purpose. If it only needs to be simple, you can
just buy the chip and make up the hardware for a lot less.

I'd be more inclined to the Maximite to tinker with.

Or a PIC32 dev module, which is also good enough to run uCLinux.

 

[keithr]

On 30/11/2011 9:41 AM, terryc wrote:

T.T. wrote:

Hence my argument that it isn't sutable for teach yourself.
Bring back COBOL.

Is this a horses for course comment?
Do you have any figures on the current percentage implementation of Cobol?

You'd be surprised, the big banks still have a lot of COBOL embedded in

their back office applications

 

[terryc]

swanny wrote:

I learnt a great deal by reading the Unix SysV source code and using the
K&R C book.

Lol, just remembered what K&R was. I don't have the book, but a print
out of it that the uni offered during their summer trial.

You learnt programmng or C?

Was already studying programming, but the course was in Pascal which I
wasn't a big fan of. There were a few sys admins and and real
programmers there (keeping the old Vax 11/780 alive) who I got to know
and undertook C programming myself (not part of the Uni course). Unix
was written in C so it was a good place to start reading code.
70's, Newcastle, Engineering/Computing & Maths used Fortran(cards only

input) and Physics required Basic(dec terminal). Maths required Algol
and Cobol for some assisgnments.

I picked up my K&C C printout when the uni IT dept was running a summer
trial prior of providing C program support to the various departments
and course. Mutter, mutter, scored a full summer of work and t was over.

Surprisingly the code was very well structured and documented
throughout, making it easy to understand.

That is germain to my critiscism. By the time in the 90s while at
another course of study at UTS, engineering first taught Pascal and then
"C", the concept of "commenting" in C was well and truly banned. WTF, I
failed an assignment because I commented?

I had no problems with Fortran, Basic, Algol, Cobol, Assember(Dec PDP
70), etc as all the programs I got my hands on were fully commented.

It didn't help that the eng lecturer was using lectures to air his
bitches with the IT dept over who taught what and the tutors were a pair
of brown nosing accolytes who didn't help. Not surprising others passed
the assignments by submtting one from last year with the date changed as
the lab gear didn't work(hardware conflict) for the full year.

Then I ran into programmers at a linux users group that proudly chest
beat about not needing comments as "if you are not good enogh to read
the code, then you are not good enough to change my code". WTF. Then I
found that linux source code is very scarce on comments. I decided that
I had better things to do than learn C.

Agree with that. The GUI sort of killed off the 'Ready>' prompt of the
old Z80 machines.

lol, as a contractor I just loved popping terminals, editing the config
and fixing the "problem". Save so much time in this gui centric world.

I remember the teletype machines and the old thermal printer terminals.
Wrote a program that took over them from a remote console, allowing
copious BEL and FF characters to be sent to them, surprising the first
year undergrads

Woops, doing anything like that near the uni IT dept brought a
thundering of feet down the corridor. If you were remote, they just cut
off all the terminals in that location. Seems a few people used that
trick to collect logins and passwords.

The best use of the teletype, since it wasn't student mainframe access,
was to create paper tape pictures.

 

[terryc]

T.T. wrote:

"terryc" <newsninespam-spam@woa.com.au> wrote in message
news:jb3n1s$6tv$2@dont-email.me...
T.T. wrote:

Hence my argument that it isn't sutable for teach yourself.
Bring back COBOL.
Is this a horses for course comment?
Do you have any figures on the current percentage implementation of Cobol?
If it is above zero, it wouldn't be by much.

Mmmmm, I suspect a lot of people are fooled by the fact that it doesn't
figure much in contracts/jobs. i.e. all new code is written in something
else so cobol is dead.

COBOL was such a forgiving language that, in its day, thousands of
programmers who should have starved to death paid off mortgages and raised
families.

...then they retired and left a mountain of code to be maintained and a
single mention of exposure to cobol in my resume brought continual
offers of cobol contracts(until I removed it).

If it ain't broke, then you don't fix it* and that cobol code was doing
a lot of the back office grunt work and it stayed there until it
couldn't be patched anymore and they found the millions to replace it.

I would be too surprised that in the banking and financial services
world, that there isn't a mountain of it still sitting there quietly
processing data. It might be the real reson why we don't have instant
cheque clearance.

*I worked in one financial institution where they had a large
refidgerator sized small mainframe sitting there doing zip/zilch/nada
all day except for a single international dialup to collect a handful of
numbers each night. It wasn't just that any other machine in the data
centre or a PC could have easily taken on the job, but the fact that
they didn't have the OS security rating. so it stayed.

 

[John Whorfin]

Matt Starr will be very interested.

 

[keithr]

On 30/11/2011 2:47 PM, terryc wrote:

swanny wrote:

I learnt a great deal by reading the Unix SysV source code and using
the
K&R C book.

Lol, just remembered what K&R was. I don't have the book, but a print
out of it that the uni offered during their summer trial.

You learnt programmng or C?

Was already studying programming, but the course was in Pascal which I
wasn't a big fan of. There were a few sys admins and and real
programmers there (keeping the old Vax 11/780 alive) who I got to know
and undertook C programming myself (not part of the Uni course). Unix
was written in C so it was a good place to start reading code.
70's, Newcastle, Engineering/Computing & Maths used Fortran(cards only
input) and Physics required Basic(dec terminal). Maths required Algol
and Cobol for some assisgnments.

I picked up my K&C C printout when the uni IT dept was running a summer
trial prior of providing C program support to the various departments
and course. Mutter, mutter, scored a full summer of work and t was over.

Surprisingly the code was very well structured and documented
throughout, making it easy to understand.

That is germain to my critiscism. By the time in the 90s while at
another course of study at UTS, engineering first taught Pascal and then
"C", the concept of "commenting" in C was well and truly banned. WTF, I
failed an assignment because I commented?

I had no problems with Fortran, Basic, Algol, Cobol, Assember(Dec PDP
70), etc as all the programs I got my hands on were fully commented.

It didn't help that the eng lecturer was using lectures to air his
bitches with the IT dept over who taught what and the tutors were a pair
of brown nosing accolytes who didn't help. Not surprising others passed
the assignments by submtting one from last year with the date changed as
the lab gear didn't work(hardware conflict) for the full year.

Then I ran into programmers at a linux users group that proudly chest
beat about not needing comments as "if you are not good enogh to read
the code, then you are not good enough to change my code". WTF. Then I
found that linux source code is very scarce on comments. I decided that
I had better things to do than learn C.

I tried to modernise some code written like that, 2000 odd lines of
dense code without a single comment, randomly named variables, and lots
of little tricks with pointers.

After a couple of weeks I decided that there were better things to do
with my life.

 

[Jasen Betts]

On 2011-11-29, terryc <newsninespam-spam@woa.com.au> wrote:

keithr wrote:
On 29/11/2011 6:58 PM, terryc wrote:
swanny wrote:

Rubbish. C is an excellent language to program in. If the best you can
do is Basic then you probably need to learn how to program properly.

Typical comment I've come to expect. How did you learn it?

If C is so great, how come there was that security bug in randomisation
for so many months. It was a basic mistake and in any other language it
could have been picked up by a beginner reading te source code.

What a load of crap, that would be a problem with a particular
implementation not the language itself.

The problem is that the language is so cryptic, you have to be well
versed to spot a simple mstake. similarly, it makes it very difficult
for people to teach themselves by reading the code.

All useful languages are like that. (not just computer programming languages,
all.)

C is one of the smallestt languages. it's got fewer keywords and
other symbols than most other languages.

My 2c is that if the language lends itself to faulty implementations,
then it isn't such a great language.

that's like saying if aome automobiles crash then the concept of
automibiles is flawed

--
⚂⚃ 100% natural

--- Posted via news://freenews.netfront.net/ - Complaints to news@netfront.net ---

 

[Gary R. Schmidt]

On 30/11/2011 1:28 AM, terryc wrote:

keithr wrote:
On 29/11/2011 6:58 PM, terryc wrote:
swanny wrote:

Rubbish. C is an excellent language to program in. If the best you can
do is Basic then you probably need to learn how to program properly.

Typical comment I've come to expect. How did you learn it?

If C is so great, how come there was that security bug in randomisation
for so many months. It was a basic mistake and in any other language it
could have been picked up by a beginner reading te source code.

What a load of crap, that would be a problem with a particular
implementation not the language itself.

The problem is that the language is so cryptic, you have to be well
versed to spot a simple mstake. similarly, it makes it very difficult
for people to teach themselves by reading the code.
The problem was that a /temporary/ change - made to stabilise things

while something else was being tested - got put back *without* being
reviewed.

My 2c is that if the language lends itself to faulty implementations,
then it isn't such a great language.
It doesn't matter *what* language you use, if the code isn't reviewed

then using it could cause the proverbial monkeys to fly out your butt!

Yes, I use C, and prefer it. I learned it back in 1981, after learning
Pascal in 1980, and then various assemblers and COBOLs and FORTRANs and
DIBOL and SPL and SPAN and POWERflex and perl and SQL and C++ and ...

Cheers,
Gary B-)

--
When men talk to their friends, they insult each other.
They don't really mean it.
When women talk to their friends, they compliment each other.
They don't mean it either.