In Tribute to Gary Kindall - A True Hero
Michael Hauben
in high school. One of the heroes I picked was Gary Kindall.
I am contributing this essay to help pay tribute to this
true hero of the personal computer movement and revolution. It
was printed in the Amateur Computerist Newsletter in the Fall
1989 issue (volume 2 no 4).
True Heroes
by Michael Hauben <hau...@columbia.edu>
A hero is not someone who only appears a hero in our
eyes, but one who achieves good for the common man, against
the will of the establishment. This person must be able to
able to stand up against the common bad, instead of pleasing
somebody. Often this person will either have all or most of
society going against him. The hero must be able to stand up
for what he believes in and not succumb to outside pressure.
Galileo Galilei is a good example of a hero. He challenged
the Roman Catholic Church by publishing his scientific findings,
which were against the then current and less accurate Aristotelian
science. He kept on studying and his mind was unchanged even
after the Inquisition challenged him. So in the end, when
the Inquisition tired to silence him by putting him under
house-arrest, he still got his writings out to the people.
Galileo would not be silenced by the Inquisition because in
his search for the truth he was not afraid to oppose
authority!
Gary Kindall is another hero that was part of a
heroic movement. He was the creator of CP/M (Control Program
for Microprocessors), the first operating system for
micro-computers. CP/M was developed during the beginning of
the micro-computer revolution. Before micro-computers were
developed, IBM (International Business Machines) and other
big companies produced the only computers available. The
computers that they marketed were incompatible mainframes
and mini-computers which only big businesses and big
universities could afford. What the micro-computer
revolution brought together was a community of people who
wanted to make the power of computers available to the common
person, rather than just to large institutions. This community
succeeded in building the personal computer. They insisted
that the various personal computers have open architectures and be
able to work similarly and run the same software. CP/M played
a big part in making personal computers this compatiblity
possible.
After a while, IBM wanted to get into the micro-computer
market, so they introduced a micro-computer that was incompatible
with everything else. No one would have anything to do with this
machine, so IBM was forced to come out with a more compatible and
open machine! IBM went to Gary Kildall and asked to use a version
of CP/M for the IBM PC (Personal Computer), but first they
wanted him to sign a non-disclosure agreement. He would not
agree to IBM's requirement of secrecy because that would be
breaking the micro-computer principle of keeping everything
open and available to all the people. So IBM did not use
CP/M. Instead they used an operating system that was almost
exactly like CP/M except it was developed especially for
IBM's PC. Gary Kindall and the people who brought the
micro-computer into existince are heroes. They released the
secret of computer power kept by IBM and other big companies
to all the people!
Sports heroes and celebrities are not true heroes
because they cannot bring about real changes. People like
Galileo and Kindall are real heroes because they made a
significant change in the world benefitting the common
person. All ages have untrue heroes, but true heroes are
rare.
--------------------------------------------------------------------------
Michael Hauben Columbia College'95 Editor of Amateur Computerist Newsletter
by day hau...@columbia.edu by night
<a href="http://www.cc.columbia.edu/~hauben/home.html">Netizen's Cyberstop</A>
Ed Moore
: open machine! IBM went to Gary Kildall and asked to use a version
^
^
Got the right name once. :-)
Michael Hauben
Gary Kildall's last name. The original piece was written many
years ago, and I was not careful enough in editing it for this
post. I am sorry for my mistake, and am thankful for the various
people who emailed me about my error.
-Michael
This is an essay I wrote about heroes when I was a freshman
in high school. One of the heroes I picked was Gary Kildall.
I am contributing this essay to help pay tribute to this
true hero of the personal computer movement and revolution. It
was printed in the Amateur Computerist Newsletter in the Fall
1989 issue (volume 2 no 4).
True Heroes
by Michael Hauben <hau...@columbia.edu>
A hero is not someone who only appears a hero in our
eyes, but one who achieves good for the common man, against
the will of the establishment. This person must be able to
able to stand up against the common bad, instead of pleasing
somebody. Often this person will either have all or most of
society going against him. The hero must be able to stand up
for what he believes in and not succumb to outside pressure.
Galileo Galilei is a good example of a hero. He challenged
the Roman Catholic Church by publishing his scientific findings,
which were against the then current and less accurate Aristotelian
science. He kept on studying and his mind was unchanged even
after the Inquisition challenged him. So in the end, when
the Inquisition tired to silence him by putting him under
house-arrest, he still got his writings out to the people.
Galileo would not be silenced by the Inquisition because in
his search for the truth he was not afraid to oppose
authority!
Gary Kildall is another hero that was part of a
IBM's PC. Gary Kildall and the people who brought the
micro-computer into existince are heroes. They released the
secret of computer power kept by IBM and other big companies
to all the people!
Sports heroes and celebrities are not true heroes
because they cannot bring about real changes. People like
Galileo and Kildall are real heroes because they made a
Rahul Dhesi
(Michael Hauben) writes
[ quite a bit of revisionism,
with too many inaccuracies to rebut point by point ]
>This community
>succeeded in building the personal computer. They insisted
>that the various personal computers have open architectures and be
>able to work similarly and run the same software. CP/M played
>a big part in making personal computers this compatiblity
>possible.
>After a while, IBM wanted to get into the micro-computer
>market, so they introduced a micro-computer that was incompatible
>with everything else.
Incompatibility was quite the norm, I believe. Who remembers the
COSMAC ELF, the MITS Altair, the Ohio Scientific (with THREE
microprocessors), the TRS-80, Exidy Sorcerer, Commodore 64, Sinclair,
the Apple II, and many others before and after them? CP/M machines
were compatible with other CP/M machines, but only if (a) the disk
formats were the same, and they usually were not, and if (b) CP/M
loaded in memory at the same address, which it often did not. CP/M was
a dead-end system, and its demise is often incorrectly blamed on IBM's
marketing.
CP/M had no concept of a file hierarchy (even in the mid-eighties, when
DEC was still pushing it over MS-DOS), and had no concept of any sort
of loadability of device drivers -- two features that made MS-DOS,
version 2.0 onwards, far superior to any version of CP/M. Even
something so basic as standardizing disk formats was something Digital
Research neglected to do, resulting in a Tower of Babel situation when
you wanted to simply move a file from one CP/M machine to another.
(The microcomputer world was not to know much about compatibilty and
standardization until the much-maligned IBM PC and clones gained a
dominant share in the hardware market and when the much-maligned
Microsoft gained a dominant share in the software market.)
CP/M was consistently priced at around $150, going down only slightly
when MS-DOS became available for $40.
>[Gary Kildall] He would not
>agree to IBM's requirement of secrecy because that would be
>breaking the micro-computer principle of keeping everything
>open and available to all the people. So IBM did not use
>CP/M. Instead they used an operating system that was almost
>exactly like CP/M except it was developed especially for
>IBM's PC.
Hm.
(I thought Kildall was away flying a plane when IBM came knocking,
but that could be a rumor.)
It's not clear to me that CP/M was any less secret than MS-DOS. In
neither case was source code available on terms that I would consider
reasonable, and in both cases there was just enough information
provided to add one's own device drivers and little more. In the
early days a signed license agreement was required *before* one could
purchase a copy of CP/M, while MS-DOS used a traditional 'shrink-wrap'
license with no legal force.
>People like
>Galileo and Kildall are real heroes because they made a
>significant change in the world benefitting the common
>person
Hm again. The world is now rid of CP/M, and I say good riddance. It
was a poor imitation of DEC's DOS/BATCH operating system and set back
the cause of friendly microcomputer use by perhaps a decade. It
saddled us with the broken OS that MS-DOS is, because MS-DOS was
designed to be compatible with CP/M for software porters, and future
versions of MS-DOS to a great extent had to remain compatible with
MS-DOS 1.0 to avoid confusing users. It was pure luck that gave CP/M
the market share that it had, not beauty or brilliance. Had Gary
Kildall paid more attention to TENEX or MULTICS when looking for
inspiration, the world might be different today and his estate
substantially larger.
--
Rahul Dhesi <dh...@rahul.net>
also: dh...@cirrus.com
John De Armond
>In <319ms2$8...@apakabar.cc.columbia.edu> hau...@bonjour.cc.columbia.edu
>(Michael Hauben) writes
>[ quite a bit of revisionism,
>with too many inaccuracies to rebut point by point ]
>>This community
>>succeeded in building the personal computer. They insisted
>>that the various personal computers have open architectures and be
>>able to work similarly and run the same software. CP/M played
>>a big part in making personal computers this compatiblity
>>possible.
Open Architecture in CP/M machines? Surely you jest.
>>After a while, IBM wanted to get into the micro-computer
>>market, so they introduced a micro-computer that was incompatible
>>with everything else.
ROFL!
>Incompatibility was quite the norm, I believe. Who remembers the
>COSMAC ELF, the MITS Altair, the Ohio Scientific (with THREE
>microprocessors), the TRS-80, Exidy Sorcerer, Commodore 64, Sinclair,
>the Apple II, and many others before and after them? CP/M machines
>were compatible with other CP/M machines, but only if (a) the disk
>formats were the same, and they usually were not, and if (b) CP/M
>loaded in memory at the same address, which it often did not. CP/M was
>a dead-end system, and its demise is often incorrectly blamed on IBM's
>marketing.
One of my favorite CP/M machines, the Zorba, WAS my favorite machine
because built into its BIOS was the capability to read other
disk formats. The last version of the little box before the company
went to that great disk disk drive in the sky would read/write
over 100 formats and that was just the 5-1/4 format. Probably
50 more if one hacked an 8" drive onto the internal (but never
wired out) connector and flipped a few bits in the BIOS source code
and recompiled. That was another MAJOR selling point of the Zorba -
BIOS source code plus the proprietary utilites to stick it on a boot
disk. Not that all this compatability did a lot of good in some
instances. All it took to throw a monkey wrench in the works was
one of the other computers to use a different FDD controller.
The Zorba used the then very popular Western Digital FDD controller.
I had the funny problem of, after getting a PC and a compatability
program for it called Uniform, of being able to read or write
Zorba disks on the PC only if they were formatted on the Zorba. Format
them on the PC (which used a different FDD controller) to Zorba
format and nuthin' doing.
Oh, and anyone remember the wonderful world of MODEM7 and all the dozens
of drivers, one for each machine architecture and UART?
Ah, them open systems were wonderful - mine was an open system - The
case stayed open a lot so I could jiggle the wires and make it run again.
>>[Gary Kildall] He would not
>>agree to IBM's requirement of secrecy because that would be
>>breaking the micro-computer principle of keeping everything
>>open and available to all the people. So IBM did not use
>>CP/M. Instead they used an operating system that was almost
>>exactly like CP/M except it was developed especially for
>>IBM's PC.
>Hm.
>(I thought Kildall was away flying a plane when IBM came knocking,
>but that could be a rumor.)
Same rumor I heard at the time.
>It's not clear to me that CP/M was any less secret than MS-DOS. In
>neither case was source code available on terms that I would consider
>reasonable, and in both cases there was just enough information
>provided to add one's own device drivers and little more. In the
>early days a signed license agreement was required *before* one could
>purchase a copy of CP/M, while MS-DOS used a traditional 'shrink-wrap'
>license with no legal force.
What I consider to be my crowning achievement was the writing of a
BIOS replacement for the Zorba to do some special goodies. This is
not because writing a BIOS for a Z-80 machine is so difficult but
because the DR docs were SOOOOO shitty.
>>People like
>>Galileo and Kildall are real heroes because they made a
>>significant change in the world benefitting the common
>>person
>Hm again. The world is now rid of CP/M, and I say good riddance. It
>was a poor imitation of DEC's DOS/BATCH operating system and set back
>the cause of friendly microcomputer use by perhaps a decade. It
>saddled us with the broken OS that MS-DOS is, because MS-DOS was
>designed to be compatible with CP/M for software porters, and future
>versions of MS-DOS to a great extent had to remain compatible with
>MS-DOS 1.0 to avoid confusing users. It was pure luck that gave CP/M
>the market share that it had, not beauty or brilliance. Had Gary
>Kildall paid more attention to TENEX or MULTICS when looking for
>inspiration, the world might be different today and his estate
>substantially larger.
So true. Anyone who thinks Kildall was a hero should reconsider.
A person who:
* Wrote CP/M on company time at Intel.
* Called assembler output listings "porting documentation".
* Was so obsessively secret that even hardware vendors had trouble
getting sufficient docs to write BIOSes in the beginning.
* Made no attempt to standardize disk formats, conceive of logical devices
or anything else then in common usage.
* Refused to sell retail.
* Refused to do ANYTHING for the hacker community that was creating
his market.
* Pissed IBM off to the extent it went elsewhere
And the worst sin of all:
* Made Bill Gates something other than the two-bit BASIC programmer
that he was.
Can hardly be called a hero. The industry's chief villian, perhaps, but
not a hero.
John
--
John De Armond, WD4OQC, Marietta, GA j...@dixie.com
Performance Engineering Mag. Unsolicited email published at my sole discretion
--
The government has 3 new savings bonds: The Steffie bond with no maturity,
the Gore bond with no interest and the Clinton bond with no principle.
John Wilson
>* Wrote CP/M on company time at Intel.
You mean the same way Gates wrote MBASIC-80 on a gov't-supported PDP10
that specifically prohibited private use?
>* Called assembler output listings "porting documentation".
>* Was so obsessively secret that even hardware vendors had trouble
> getting sufficient docs to write BIOSes in the beginning.
Ever tried to get the same docs from Microsoft? When I tried they told me
I could only get MS-DOS pre-ported from the "vendor". Well the vendor was
Intel and their hard-on for iRMX-86 was blocking their view of the market.
>* Made no attempt to standardize disk formats, conceive of logical devices
> or anything else then in common usage.
Lucky thing MS-DOS *did* standardize disk formats. We had the 8-sector
160KB and 320KB standards, then with DOS 2.0 the 9-sector 180KB and 360KB
standards, then DOS 3.X and the AT added the 15-sector 1.2MB standard,
the 9-sector 720KB 3.5" standard and finally the 18-sector 1.44MB standard.
Now the 2.88MB standard and the CD-ROM standard have appeared. You need
three drives and five bit rates just to be able to read all the "standard"
disks! They should have gotten Zorba to write the BIOS!
>* Refused to do ANYTHING for the hacker community that was creating
> his market.
What, like a barbecue? I can't think of what he should have done.
> And the worst sin of all:
>
>* Made Bill Gates something other than the two-bit BASIC programmer
> that he was.
We can't pin the blame for that on Kildall, that one's due to you and me
and everyone else who's ever paid full price for mediocre software.
>Can hardly be called a hero. The industry's chief villian, perhaps, but
>not a hero.
Hey, at least he didn't get as far as he did by bullying people like Gates
has. It's the IBM/Microsoft mediocrity that has made the microcomputer
industry practically stagnate for the last 13 years. The idea of selling
more stuff by offering better products at lower prices is totally beyond
either company. Microsoft sells stuff just by adding 17 to the version
numbers every few months, and IBM is still clinging to the same old market
that buys IBM machines regardless of quality just because they want to
be identified with a large company like IBM, as if having an AS/400 can
magically turn you into a multinational conglomerate.
What Kildall did was help make order out of chaos. The disk format
problem was a relatively minor deal (and hasn't gone away, see above and
also see those 5.25" mail-in cards that come with all your software).
What's happened since then is like a huge step backwards, instead of
standardizing the OS so that software can run anywhere, we've standardized
the hardware and let the software depend on the hardware as much as it
wants (or rather, as much as it has to since MS-DOS may be 2 or 3 times
as good as CP/M 2.2, but it certainly doesn't reflect the 13 years of
development time it's had and there are still a zillion things you have
to do yourself). Coming out with a completely new and different CP/M
machine used to be a piece of cake, but now look at us with two different
flavors of real mode and many kinds of A20 gate, and CGA compatibility
built into the EGA compatibility in the VGA compatilibity in your SVGA,
and the 16550A adding FIFOs to the 16450 upgrade to your 8250A UART.
The enhanced keyboard is doctored to send the same scan codes as the
84-key keyboard, so that the same outboard microprocessor can convert
those to the ones the old 83-key keyboard sent. This foolishness never
used to happen! You'd get burned developing for TRSDOS or HDOS or
whatever, so you'd switch over to CP/M-80 and modulo the terminal type
everything would be fine, pretty much. And I haven't noticed MS
releasing any version of MSDOS that works with dual instruction sets
or multiple terminals!
John Wilson
Greg Bridgewater
Subject: Re: In Tribute to Gary Kildall - A True Hero
Newsgroups: alt.folklore.computers,comp.misc,comp.os.cpm,alt.amateur-comp
In article <njx8#6...@dixie.com> you wrote:
> Rahul Dhesi <dh...@rahul.net> writes:
> >In <319ms2$8...@apakabar.cc.columbia.edu> hau...@bonjour.cc.columbia.edu
> >(Michael Hauben) writes
> >[ quite a bit of revisionism,
> >with too many inaccuracies to rebut point by point ]
I agree, lots of innaccuracies.
> >>This community
> >>succeeded in building the personal computer. They insisted
> >>that the various personal computers have open architectures and be
> >>able to work similarly and run the same software. CP/M played
> >>a big part in making personal computers this compatiblity
> >>possible.
> Open Architecture in CP/M machines? Surely you jest.
I believe he was trying to make the point that CP/M offered a 'common'
operating system for many flavors of otherwise incompatible machines.
This carried over into the IBM clones; if you will recall (I know my
memories are failing) many early 'MSDOS'/'IBM' clones did _not _ have
compatible hardware either, but the MSDOS allowed common functionality
again, as long as authors adhered to use of DOS or BIOS system calls and
did not try direct hardware manipulation. Of course speed criteria often
forced designers to do just that, especially for video manipulation.
For me personally, CP/M did exactly what I expected at the time, it
allowed me to load and execute the same programs on a flock of machines
composed of wildly differing hardware.
> >Incompatibility was quite the norm, I believe. Who remembers the
> >COSMAC ELF, the MITS Altair, the Ohio Scientific (with THREE
> >microprocessors), the TRS-80, Exidy Sorcerer, Commodore 64, Sinclair,
> >the Apple II, and many others before and after them? CP/M machines
> >were compatible with other CP/M machines, but only if (a) the disk
> >formats were the same, and they usually were not, and if (b) CP/M
> >loaded in memory at the same address, which it often did not. CP/M was
> >a dead-end system, and its demise is often incorrectly blamed on IBM's
> >marketing.
I believe IBM marketing had a great deal to do with the demise of CP/M.
They held out the promise of the 'god' of big computers providing
'blue'suit support to the small businessman on a 'reasonably priced' piece
of hardware (choke, really a piece of crap). I know few hobbiests that
ran out to purchase an IBM... remember it also had only 64K capability,
ran slower than Z80 machines, came with only a cassette port, basic and
not even a video card in it's base configuration. True to Invest Big
Money's philosophy, you had to "buy more memory" and add-ons to even have
a machine that could say hi to you. Selling that piece of (?)_ to
America had to be one of the biggest coups of all time!
> One of my favorite CP/M machines, the Zorba, WAS my favorite machine
> because built into its BIOS was the capability to read other
> disk formats. The last version of the little box before the company
> went to that great disk disk drive in the sky would read/write
> over 100 formats and that was just the 5-1/4 format. Probably
> 50 more if one hacked an 8" drive onto the internal (but never
> wired out) connector and flipped a few bits in the BIOS source code
> and recompiled. That was another MAJOR selling point of the Zorba -
> BIOS source code plus the proprietary utilites to stick it on a boot
> disk. Not that all this compatability did a lot of good in some
> instances. All it took to throw a monkey wrench in the works was
> one of the other computers to use a different FDD controller.
> The Zorba used the then very popular Western Digital FDD controller.
> I had the funny problem of, after getting a PC and a compatability
> program for it called Uniform, of being able to read or write
> Zorba disks on the PC only if they were formatted on the Zorba. Format
> them on the PC (which used a different FDD controller) to Zorba
> format and nuthin' doing.
Hmmm, I have one of those lying around here... would have been nice to
have gotten manuals or something when I salvaged it from an 'early'
death... might still be able to to something with it. What might you have?
> Oh, and anyone remember the wonderful world of MODEM7 and all the dozens
> of drivers, one for each machine architecture and UART?
Yes, but a _free_ piece of software that permitted exchange of
information beyond even what mainframes normally were doing!
> Ah, them open systems were wonderful - mine was an open system - The
> case stayed open a lot so I could jiggle the wires and make it run again.
Hmmm, and my 486-50's case still does... believe they are operating under
similar conditions :-)!
> >>[Gary Kildall] He would not
> >>agree to IBM's requirement of secrecy because that would be
> >>breaking the micro-computer principle of keeping everything
> >>open and available to all the people. So IBM did not use
> >>CP/M. Instead they used an operating system that was almost
> >>exactly like CP/M except it was developed especially for
> >>IBM's PC.
Strange isn't it that huge IBM had to go elsewhere for a simple operating
system?
> Same rumor I heard at the time.
> What I consider to be my crowning achievement was the writing of a
> BIOS replacement for the Zorba to do some special goodies. This is
> not because writing a BIOS for a Z-80 machine is so difficult but
> because the DR docs were SOOOOO shitty.
Then there were the tons of manuals for any IBM...
By the way, what did your BIOS do special and would you mind parting with
a copy? I've never been below making use of other folks' already (I hope)
debugged code rather reinvent it!
> >Hm again. The world is now rid of CP/M, and I say good riddance. It
> >was a poor imitation of DEC's DOS/BATCH operating system and set back
> >the cause of friendly microcomputer use by perhaps a decade. It
Is CP/M gone? Darn, I guess all my machines will cease to exist now.
That poor guy must really be desperate to come to a conference dedicated
to the CP/M operating system and 'bit'bang the system!
> >saddled us with the broken OS that MS-DOS is, because MS-DOS
Hmmm, why didn't he just offer a better mouse trap instead of complaining
about a decade late?
> So true. Anyone who thinks Kildall was a hero should reconsider.
> A person who:
> * Wrote CP/M on company time at Intel.
... and here I'd always thought he wrote it while he was employed by the
government (probably got that idea from the PL/M source files for CP/M)?
> * Called assembler output listings "porting documentation".
better than nothing :-).. I was able to use them to port when needed.
> * Made no attempt to standardize disk formats, conceive of logical devices
> or anything else then in common usage.
Believe that was his intent... he provided a base operating system that
ran through a BIOS interface built to his specifications. Don't believe
he ever had any intent to get into hardware.
> * Refused to sell retail.
I tried to buy direct, and could have, but the price was outlandish.
> And the worst sin of all:
> * Made Bill Gates something other than the two-bit BASIC programmer
> that he was.
Didn't know that he ever became more than that. Nothing that I ever
purchased from MicroSlop ever was bug free, nor were free fixes offered.
Buy the next version was the motto. Then again, look at the original
BASIC compiler for MSDOS, it contained the _same_ bugs as the early CP/M
versions. He got to sell his 'bugware' twice! and the upgrades!
> Can hardly be called a hero. The industry's chief villian, perhaps, but
> not a hero.
Oh come on now, think about it... what did you do to fix the situation?
Personally, I used the system, still do, and fully intend to continue.
That does not mean it is the best, I also use UNIX and NWDOS, etc. and a
whole flock of other than 8080 compatible processors.
--
Greg Bridgewater
voice (412)363-9713
internet: greg...@telerama.lm.com
--
Greg Bridgewater
voice (412)363-9713
internet: greg...@telerama.lm.com
John De Armond
>In article <njx8#6...@dixie.com>, John De Armond <j...@dixie.com> wrote:
>>* Wrote CP/M on company time at Intel.
>You mean the same way Gates wrote MBASIC-80 on a gov't-supported PDP10
>that specifically prohibited private use?
Was there any purpose to your post other than to pontificate. Are you
sufficiently reading impaired that you somehow interpreted
my article as some sort of defense of microsoft? If so, wow!
>Lucky thing MS-DOS *did* standardize disk formats. We had the 8-sector
>160KB and 320KB standards, then with DOS 2.0 the 9-sector 180KB and 360KB
>standards, then DOS 3.X and the AT added the 15-sector 1.2MB standard,
>the 9-sector 720KB 3.5" standard and finally the 18-sector 1.44MB standard.
>Now the 2.88MB standard and the CD-ROM standard have appeared. You need
>three drives and five bit rates just to be able to read all the "standard"
>disks! They should have gotten Zorba to write the BIOS!
Why? Plug the appropriate hardware in (you aren't REALLY saying
that microsoft should somehow have forced a HARDWARE standard on
people, are you?), flip a few bits in DOS and you can read any of these.
I'm sure the subtlety between support for multiple hardware and
the LACK of support for a common format on a GIVEN hardware (DSDD
5.25" floppy, for the most part) went right over your head. I'm
still trying to figure out what your beef is. Do you argue that
Microsoft should NOT support all the different drives available?
If so, wow!
>>* Refused to do ANYTHING for the hacker community that was creating
>> his market.
>What, like a barbecue? I can't think of what he should have done.
I can. Retail sales of generic CP/M for a reasonable price, decent
documentation, porting tools, someone on the other end of the phone
who didn't snarl. In that era, since hackers == VARs and systems
integrators, they were cutting their own throats but NOT supporting
the hacker. All DR could see was the dollars they were making today;
they couldn't see past the ends of their noses, must less into
the future.
Hell, he could have even licensed the source code. Not that it was any
great secret. C.C Calkin's "disassembler" that produced commented,
M-80 assemblable source code worked fine. Kildell could have
benefitted from the same community effort as the budding Unix community
was. I hacked all sorts of nifty stuff into CP/M once I bought
Calkin's product including the couple of byte fix to the dreaded
BDOS error after changing floppies but not hitting ^C. Just made
it do a drive reset like DR would have done had it not had its head
where the sun never shines. Leaving this "feature" in, which
crippled any serious business applications under CP/M was a sin.
>> And the worst sin of all:
>>
>>* Made Bill Gates something other than the two-bit BASIC programmer
>> that he was.
>We can't pin the blame for that on Kildall, that one's due to you and me
>and everyone else who's ever paid full price for mediocre software.
Huh? Had Kildall not had his head up his ass when dealing with IBM,
there never would have been an MS-DOS. Gates would still be spending
his time writing articles and letters calling everyone thieves for
stealing his BASIC-80. Of course, the thought of CP/M-486 and
fourty seven thousand files in a flat directory is truly frightening.
People bought PC-DOS instead of CP/M-86 because the former cost about
$40 while the latter cost almost $150. With little difference
in functionality, who would pay more?
John Wilson
>Was there any purpose to your post other than to pontificate.
You're so right, that was the purpose.
>Why? Plug the appropriate hardware in (you aren't REALLY saying
>that microsoft should somehow have forced a HARDWARE standard on
>people, are you?), flip a few bits in DOS and you can read any of these.
Make up your mind, you shot down CP/M for not standardizing, and now
DOS is a hero for supporting multiple formats. Lucky you don't like MS.
>I'm sure the subtlety between support for multiple hardware and
>the LACK of support for a common format on a GIVEN hardware (DSDD
>5.25" floppy, for the most part) went right over your head.
Yep, just like CP/M's almost universal 8" 3740 support went over yours.
Besides, no one has real 360KB floppy drives any more and the disks don't
work reliably on 1.2MB drives. And I don't know what the hell was up
with the PC-DOS 4.01 distribution, but you couldn't read the floppies
from a 3.3 machine and if you booted from them you couldn't read your
3.3 hard disk. But it's CP/M that's bad!
>I'm still trying to figure out what your beef is.
My beef is that you've totally missed the point of what a positive impact
CP/M had on the micro market, in spite of its many technical failings.
It brought order out of chaos without sacrificing innovation. These
days you can't touch the PC architecture w/o breaking something.
Yet you flame off just because it didn't have tree directories and
Kildall had better things to do than get sat on by IBM. Would *you*
have thrown yourself at IBM's mercy?
>I can. Retail sales of generic CP/M for a reasonable price, decent
>documentation, porting tools, someone on the other end of the phone
>who didn't snarl.
What else is new? MS provides none of these things. There is no generic
MS-DOS, the way to build your own IO.SYS file and patch PRINT.COM to
generate EOIs on your machine isn't documented anywhere. Porting is not
supposed to be your business. And I know all this because I was told so
on the phone by a snarling MS employee. Why didn't you write your own OS
and give us all this back when the Charlie Chaplin lookalike was
encouraging everybody to submit programs?
>Hell, he could have even licensed the source code.
Why?! Having a zillion customized versions of CP/M floating around would
have undermined the whole idea of providing a common OS.
>Kildell could have
>benefitted from the same community effort as the budding Unix community
>was.
Exactly my point! Nothing runs right out of the box on UNIX (unless someone
has already ported it to your particular mutation of UNIX).
>People bought PC-DOS instead of CP/M-86 because the former cost about
>$40 while the latter cost almost $150.
Oh please. When I bought my PC in 1983 the base system box with 64KB, no
video and a single-sided 5.25" floppy (which I had to replace with a DS
one because I couldn't read any of the commercial software I bought) cost
$1810. MDA & 5151 monitor was something like $500 more. The cost of the
OS was absolutely insignificant next to the hardware, I bought DOS 2.0 for
$60 because I, just like everyone else, thought that it was the OS that
IBM considered to be the "standard" PC OS. Anyway the price for CP/M-86
was lowered to $69 within a few months, to no avail.
John Wilson
Frank McConnell
>Of course, the thought of CP/M-486 and
Wouldn't need it. CP/M-86 for the IBM PC boots just fine in darn near
anything PC-compatible, including my '486 box and underneath of OS/2.
>fourty seven thousand files in a flat directory is truly frightening.
Indeed. While I've used and enjoyed CP/M, and someday will probably
get around to reassembling some of my older boxes for the pleasure of
playing with it again, I am absolutely certain there are things I
would rather have running on my desktop today.
I've often wondered, had CP/M-86 been the OS of choice for the PC
platform back then, would we have been pressured by its limitations
into replacing it with something better sooner? I'd like to think so.
-Frank McConnell
Steve Platt
>Incompatibility was quite the norm, I believe. Who remembers the
>COSMAC ELF, the MITS Altair, the Ohio Scientific (with THREE
>microprocessors), the TRS-80, Exidy Sorcerer, Commodore 64, Sinclair,
>the Apple II, and many others before and after them? CP/M machines
>were compatible with other CP/M machines, but only if (a) the disk
>formats were the same, and they usually were not, and if (b) CP/M
>loaded in memory at the same address, which it often did not. CP/M was
>a dead-end system, and its demise is often incorrectly blamed on IBM's
>marketing.
Some comments from someone who was there (programming device drivers
from CP/M 1.4 onward)... a) If you had 8" floppy disks, you *were*
capable of reading the 243k format; if you had 5/25: flopies, many
free BIOS table loaders were available to easily allow you to read
mulitple formats. b) CP/M programs *were* address independent, unless
written by a poor programmer. CP/M could load anywhere; it patched
low memory (0000-0002 and 0005-0007 (hex)) with JMPs to itself. All
user programs CALL'ed (or JMP'ed) through those locations. At the
time, it was a microcomputer innovation.
>CP/M had no concept of a file hierarchy (even in the mid-eighties, when
>DEC was still pushing it over MS-DOS), and had no concept of any sort
>of loadability of device drivers -- two features that made MS-DOS,
>version 2.0 onwards, far superior to any version of CP/M. Even
>something so basic as standardizing disk formats was something Digital
>Research neglected to do, resulting in a Tower of Babel situation when
>you wanted to simply move a file from one CP/M machine to another.
By the mid-80's CP/M was dead.In the early 80's CP/M had "user areas",
an extremely primitive user partitioning. Kinda like the Mac MFS
(only one level of folders), but worse. Also, keep in mind that Microsoft
(the software mfgr) didn't standardize disk formats, IBM (the hardware
vendor) did. Five times, by my latest count, on 5.25" media.
Somewhere, I may still have a copy of MS-DOS for an old Godbout
S-100 computer running both 8088 and 8085 processors. On 8" floppies.
>(The microcomputer world was not to know much about compatibilty and
>standardization until the much-maligned IBM PC and clones gained a
>dominant share in the hardware market and when the much-maligned
>Microsoft gained a dominant share in the software market.)
>CP/M was consistently priced at around $150, going down only slightly
>when MS-DOS became available for $40.
...yet I never paid more than around $95 for any of my copies...
(stuff deleted)
>It's not clear to me that CP/M was any less secret than MS-DOS. In
>neither case was source code available on terms that I would consider
>reasonable, and in both cases there was just enough information
>provided to add one's own device drivers and little more. In the
>early days a signed license agreement was required *before* one could
>purchase a copy of CP/M, while MS-DOS used a traditional 'shrink-wrap'
>license with no legal force.
I never did. Not for 1.4, 2.0, 2.2, Concurrent CP/M, their Macro
Assembler, etc....
>Hm again. The world is now rid of CP/M, and I say good riddance. It
>was a poor imitation of DEC's DOS/BATCH operating system and set back
>the cause of friendly microcomputer use by perhaps a decade. It
>saddled us with the broken OS that MS-DOS is, because MS-DOS was
>designed to be compatible with CP/M for software porters, and future
>versions of MS-DOS to a great extent had to remain compatible with
>MS-DOS 1.0 to avoid confusing users. It was pure luck that gave CP/M
>the market share that it had, not beauty or brilliance. Had Gary
>Kildall paid more attention to TENEX or MULTICS when looking for
>inspiration, the world might be different today and his estate
>substantially larger.
>--
>Rahul Dhesi <dh...@rahul.net>
>also: dh...@cirrus.com
Rahul, where is your sense of the times? Yes, CP/M was even more of
a glorified program loader than MS/DOS was. But remember,
*** it ran in 4k on a 16k machine ***
That's 4 kilobytes. As in around 2000 instructions Command processor,
file system manager, console and printer IO with virtual-to-physical
device redirection, etc.
Sure, it could have been Multics; with memory needs, S-100 machines would
never have been affordable by small businesses (figure 64K for the OS,
another 64k for the apps, at $25-50/kb, do the math...). Without
the primordeal demand, IBM wouldn't have built the PC. and where would
we be now?
I'm not a flag-waving Gary Kildall fan, but you have to give him credit
for providing an advanced product at the time. If he had been
Gates-hungry, who knows?
Steve Platt
Jay S. Rouman
Frank McConnell <f...@aphasia.us.com> wrote:
>I've often wondered, had CP/M-86 been the OS of choice for the PC
>platform back then, would we have been pressured by its limitations
>into replacing it with something better sooner? I'd like to think so.
ZCPR3 provided named directories (but no tree structure), an excellent
help system, termcaps, and a library that made assembly language
nearly high-level. All this was totally PD and used CP/M 2.2 as a
base. Of course, it got run over by the PC steamroller, but provides
a lot of fuel for "what if" musings.
--
Jay Rouman (j...@dexter.mi.org j...@umcc.umich.edu NIC Handle: JSR)
Paul Schlyter
Greg Bridgewater <greg...@telerama.lm.com> wrote:
>> * Made no attempt to standardize disk formats, conceive of
>> logical devices or anything else then in common usage.
>
> Believe that was his intent... he provided a base operating system that
> ran through a BIOS interface built to his specifications. Don't believe
> he ever had any intent to get into hardware.
with one single disk format: 8" single-side single-density IBM
compatible disk format (yes, those floppy disks once were the norm).
CP/M v2 added the ability to adapt the BIOS to different disks format,
most notably all those different 5.25" disk formats that existed at that
time.
In addition, the I/O byte in CP/M v2 was a rudimentary first attempt to
implement logical devices. It was of course far from perfect, but if
supported in the BIOS, the TTY:, LPT:, PUN: and RDR: logical devices
could each be re-directed to up to 4 different physical devices by simply
changing the value of the I/O byte, at absolute address 0003h.
--
----------------------------------------------------------------
Paul Schlyter, SAAF (Swedish Amateur Astronomer's Society)
Nybrogatan 75 A, S-114 40 Stockholm, Sweden
InterNet: pau...@saaf.se p...@ausys.se
Greg Bridgewater
In article <31jl7f$l...@electra.saaf.se> you wrote:
> In article <31gja5$l...@epicycle.lm.com>,
> Greg Bridgewater <greg...@telerama.lm.com> wrote:
>
> >> * Made no attempt to standardize disk formats, conceive of
> >> logical devices or anything else then in common usage.
> >
> > Believe that was his intent... he provided a base operating system that
> > ran through a BIOS interface built to his specifications. Don't believe
> > he ever had any intent to get into hardware.
>
> In CP/M v1 all disks format WERE compatible, since it only worked
> with one single disk format: 8" single-side single-density IBM
> compatible disk format (yes, those floppy disks once were the norm).
Yep, I remember them well as the standard, and adhered to it through about
v1.3; however, my v1.4 also ran 5.25! All things were possible, just not
as easily as in v2
>
> CP/M v2 added the ability to adapt the BIOS to different disks format,
> most notably all those different 5.25" disk formats that existed at that
> time.
>
> In addition, the I/O byte in CP/M v2 was a rudimentary first attempt to
> implement logical devices. It was of course far from perfect, but if
> supported in the BIOS, the TTY:, LPT:, PUN: and RDR: logical devices
> could each be re-directed to up to 4 different physical devices by simply
> changing the value of the I/O byte, at absolute address 0003h.
>
If the guy who wrote the BIOS adhered to CP/M guidelines, a lot of
features were possible. Thanks for reminding me of the I/O byte. I was
trying to point out that Kildall provided an operating system with a
whole lot of hooks on which the system designer could build. But that he
left that work up to the hardware builders for compliance. I believe he
did the right thing. I _liked_ (and still do) CP/M... it did exactly
what it was supposed to do!
Richard Plinston
>were compatible with other CP/M machines, but only if (a) the disk
>formats were the same, and they usually were not, and if (b) CP/M
You would wrong if you thought that all MS-DOS disk formats were the
same. Get a copy of FBNs ALIEN disk handler program. It caters for
hundreds of MS-DOS formats of diskette, along with the even greater
number of CP/M formats admittedly. In the early years of MS-DOS
manufacturers were making their own designs of machines and disk
formats were as varied as ever.
>loaded in memory at the same address, which it often did not. CP/M was
>a dead-end system, and its demise is often incorrectly blamed on IBM's
>marketing.
>
Which is why DRI developed Concurrent-CP/M back in 1982. This developed
into Concurrent-DOS and lately DR-Multiuser-DOS (now a Novell product).
While MS-DOS won on the single user, single tasking front, DRI developed
multi-user, multi-tasking and networked systems. In Europe in the
mid-80s CDOS had about half of the small multi-user system market via
companies such as ICL, Compupro, Phillips, and others.
CP/M-86 (and 8bit) ceased development in the early 80s in favour of
following more advanced systems, with corressponding smaller markets
and visibility. They also relied heavily on OEM and so they were less
visible, most ICL users probably never heard of CP/M or DRI. They used
ICL Concurrent-DOS (with full hierarchical disk systems and device
drivers).
>CP/M had no concept of a file hierarchy (even in the mid-eighties, when
>DEC was still pushing it over MS-DOS), and had no concept of any sort
>of loadability of device drivers -- two features that made MS-DOS,
>version 2.0 onwards, far superior to any version of CP/M. Even
Only if you restrict this harranging to CP/M-80 and CP/M-86.
Concurrent-CP/M-86, Concurrent-DOS (version 4 of CCP/M) and later
along with DR-NET kept ahead of MS-DOS in many areas and had
multi-tasking and multi-user since they replaced the original
products in 1982.
So much ahead that DR-DOS 5 was built as a single tasking version
of CDOS and it took MS 2 years to catch up. DR-DOS 6 was built
from DR-Multiuser-DOS putting it ahead of MS once again.
DRMDOS is still being developed by Novell and its OEMs: DataPac in
Australia with System Manager, IMS in UK with DOS-386-Professional,
CCI in the US with CCI-DOS. They still run CP/M-86 programs too.
Richard Plinston
>>* Wrote CP/M on company time at Intel.
>
>You mean the same way Gates wrote MBASIC-80 on a gov't-supported PDP10
>that specifically prohibited private use?
Gary wrote CP/M *for* Intel, but when they didn't want it they were
happy to let him take it away when he left.
>
>>* Made no attempt to standardize disk formats, conceive of logical devices
>> or anything else then in common usage.
>
>Lucky thing MS-DOS *did* standardize disk formats. We had the 8-sector
>160KB and 320KB standards, then with DOS 2.0 the 9-sector 180KB and 360KB
>standards, then DOS 3.X and the AT added the 15-sector 1.2MB standard,
>the 9-sector 720KB 3.5" standard and finally the 18-sector 1.44MB standard.
>Now the 2.88MB standard and the CD-ROM standard have appeared. You need
>three drives and five bit rates just to be able to read all the "standard"
>disks! They should have gotten Zorba to write the BIOS!
>
They were only the formats for the IBM machines. Sharp, Sanyo, Wang,
ACT, Apricot, and all the other MS-DOS machines also had their own
formats. Some were 80 track DD 700 - 800 Kb. All incompatible.
>What Kildall did was help make order out of chaos. The disk format
>problem was a relatively minor deal (and hasn't gone away, see above and
>also see those 5.25" mail-in cards that come with all your software).
>everything would be fine, pretty much. And I haven't noticed MS
>releasing any version of MSDOS that works with dual instruction sets
>or multiple terminals!
>
They once tried with MS-DOS 4.0/4.1 (not the later PC-DOS 4.00/MS-DOS
4.01) to make a multitasking OS. It was a failure though Wang and ICL
released version of it. They then stulsified with only minor and
bug-fix (or bug-introducing with 4.00) until DRI awoke the DOS market
again with DR-DOS 5 and then DR-DOS 6.
Œ#B€ ã¦Ì 9 dä€qC ”0tÐè
Holger Petersen
>In article <31gja5$l...@epicycle.lm.com>,
>Greg Bridgewater <greg...@telerama.lm.com> wrote:
>
>>> * Made no attempt to standardize disk formats, conceive of
>>> logical devices or anything else then in common usage.
>>
>> Believe that was his intent... he provided a base operating system that
>> ran through a BIOS interface built to his specifications. Don't believe
>> he ever had any intent to get into hardware.
>
>In CP/M v1 all disks format WERE compatible, since it only worked
>with one single disk format: 8" single-side single-density IBM
>compatible disk format (yes, those floppy disks once were the norm).
>
>CP/M v2 added the ability to adapt the BIOS to different disks format,
>most notably all those different 5.25" disk formats that existed at that
>time.
Just a small note:
I bought my Copy of CP/M 1.4 in dezember 1979 [somewere in LA] bundled
with the "SD-Sales Versafloppy". (I did not have to sign any license)
It came with both a 8" _and_ 5.25" boot-Disk.
But Yes, later the confusion in smaller Disk-Formats lead to one article
of me in a german computer magazine and a lot of income by converting...
>Paul Schlyter, SAAF (Swedish Amateur Astronomer's Society)
Holger Petersen (dg3lp)
Scott - Maxwell
>>>
>>>* Made Bill Gates something other than the two-bit BASIC programmer
>>> that he was.
>
>>We can't pin the blame for that on Kildall, that one's due to you and me
>>and everyone else who's ever paid full price for mediocre software.
>
>Huh? Had Kildall not had his head up his ass when dealing with IBM,
>there never would have been an MS-DOS. Gates would still be spending
>his time writing articles and letters calling everyone thieves for
>stealing his BASIC-80. Of course, the thought of CP/M-486 and
>fourty seven thousand files in a flat directory is truly frightening.
>
we can do what we want with what you tell us" NDA. Gates actually bathed and
did whatever IBM wanted so he got to do the OS.
I guess Bill's ego was a bit bruised after people had the "nerve" to fix all
the bugs in his BASIC and distribute it. :-)
>John De Armond, WD4OQC, Marietta, GA j...@dixie.com
>Performance Engineering Mag. Unsolicited email published at my sole discretion
/===========================================================\
| Scott Maxwell - sco...@cup.portal.com |
| Amiga 1000, 2000, 1200, Atari 800, 800XL, 1200XL, 130XE, |
| Pet, SuperPET, Vic, 64, +4, 128D, TI, //e, //c+, Nixdorf |
| I have WAY too many computers! |
\===========================================================/
Scott - Maxwell
>
>My beef is that you've totally missed the point of what a positive impact
>CP/M had on the micro market, in spite of its many technical failings.
>It brought order out of chaos without sacrificing innovation. These
>days you can't touch the PC architecture w/o breaking something.
>Yet you flame off just because it didn't have tree directories and
>Kildall had better things to do than get sat on by IBM. Would *you*
>have thrown yourself at IBM's mercy?
>
function (number escapes me) ended with a $. Gates cannot tell you why that
same function in MS-DOS still ends in a $.
>John Wilson
John Wilson
Scott - Maxwell <Sco...@cup.portal.com> wrote:
>Gates owes quite a bit to CP/M. Kildall could tell you why a certain
>function (number escapes me) ended with a $. Gates cannot tell you why that
>same function in MS-DOS still ends in a $.
OK I'll bite: why? Always seemed to me that NUL would make sense
(like RT-11 uses for .PRINT). The DOS function is 09h, "print string".
John
Richard Plinston
>the future.
DRI just saw a different future than the Gates steamroller brought to
us.
They had built MP/M as a multi-user multi-tasking system in the late
70s, along with the original DR-NET and built on these to move to
client/server, networked, workgroups, multitasking, real-time
systems. The market is moving to these now, several years later and
is looking at alternative to M$.
>
>Hell, he could have even licensed the source code. Not that it was any
>great secret. C.C Calkin's "disassembler" that produced commented,
>M-80 assemblable source code worked fine. Kildell could have
It was alledged that was how the original QDOS (forerunner of MS-DOS)
was produced. DRI was about to sue, but IBM got an agreement that they
would sell CP/M-86 only if he withdrew the case. IBM then stiffed
DRI on price.
>BDOS error after changing floppies but not hitting ^C. Just made
>it do a drive reset like DR would have done had it not had its head
>where the sun never shines. Leaving this "feature" in, which
>crippled any serious business applications under CP/M was a sin.
>
>>> And the worst sin of all:
>>>
MS-DOS 1 'fixed' this feature by not noticing the disk cahnge. It
would then write all over the new disk based on the old disk's
directory. In fact the ^C was not always necessary as a program
could do the reset too. At least the 'BDOS error' could be fixed
by putting the original disk back.
There is a related problem with MS-DOS. Take a diskette from the
drive, add a file to it on another machine and return it to the
original - DOS won't see the new file, and will happily overwrite
the area it is in. CP/M would notice the change, it may give an
'error' until it is reset, but at least it notices. Just choose
the poison you want.
>>>* Made Bill Gates something other than the two-bit BASIC programmer
>>> that he was.
>
>>We can't pin the blame for that on Kildall, that one's due to you and me
>>and everyone else who's ever paid full price for mediocre software.
>
>Huh? Had Kildall not had his head up his ass when dealing with IBM,
>there never would have been an MS-DOS. Gates would still be spending
>his time writing articles and letters calling everyone thieves for
>stealing his BASIC-80. Of course, the thought of CP/M-486 and
>fourty seven thousand files in a flat directory is truly frightening.
>
Remember when MS-DOS could only support 32MByte drives ? One had to
get Wyse or Compaq versions until MS finally 'broke the barrier'.
It was forced to by the competition. 'CP/M-486' is actually called
(in its latest version) DR-Multiuser-DOS 7. It still runs all
CP/M-86 and MP/M-86 programs, I still use many and even write them.
It also runs DOS and DPMI compliant extended with full multi-tasking
Rahul Dhesi
>Gates owes quite a bit to CP/M. Kildall could tell you why a certain
>function (number escapes me) ended with a $. Gates cannot tell you why that
>same function in MS-DOS still ends in a $.
Were you referring to strings being terminated with $, or the function
name itself (e.g. CLOCK$) ending with $? And does Kildall know which
you were referring to? (I'm sure Gates doesn't.)
John D. Baker
>'CP/M-486' is actually called
>(in its latest version) DR-Multiuser-DOS 7.
A few members of our surviving CP/M-Houston Users' Group (Houston, TX
and vicinity) delight in referring the DR-DOS vv. 5, 6, 7 as "CP/M 5,
6, 7" as appropriate. One fellow even patched the sign-on message
of his copy of DR-DOS 6 to read "CP/M 6" :-)
John D. Baker ->A TransWarp'802'd Apple //e CardZ180 Z-System nut //
Internet: jdb...@tamsun.tamu.edu, @blkbox.com, jdb...@taronga.com
BBS: JOHN BAKER on PIC of the Mid-Town [(713) 961-5817] 1:106/31,
Z-Node #45 [(713) 937-8886], The Vector Board [(716) 544-1863]
Mike Freeman
In article <1994Aug4.1...@belvedere.sbay.org>,
David E. Fox <ro...@belvedere.sbay.org> wrote:
>
> jmp PrintString
>
Make that
jp PrintString
or, better yet,
jr PrintString
We're less than 255 bytes away so a relative jump is faster.
Also, "jmp" is Intel mnemonics; "jp" is Zilog and John started out
with Zilog.
[Nit-picking mode off]
--
Mike Freeman | Internet: mi...@pacifier.com
GEnie: M.FREEMAN11 | Amateur Radio Callsign: K7UIJ
... Ask not for whom the <^G> tolls.
Holger Petersen
>OK I'll bite: why? Always seemed to me that NUL would make sense
>(like RT-11 uses for .PRINT). The DOS function is 09h, "print string".
I never see any sense in a special Character to "End the Text".
Did Yoe ever try to do Graphik to a printer? You'll need _all_ Codes!
Old BASIC's had a Byte-Counter for strings. Digital-Research CBasic (CB80)
extended that to an INT (but signed...)
>John
Holger
arog on BIX
>Richard_...@kcbbs.gen.nz (Richard Plinston) writes:
John, lighten up just a wee_tad on these folks that have
never seen a 'skeletal bios' source... but then they
never heard about the System Builder's Kits from DR either.
I've not bothered to get any of the more recent ones since
Cp/.... er, DR Dos 5. When that hit my desk and I looked at
the files on the disk... Gawd, did it bring back fun memories
of all night bios.hacking.
... and when I get the earth.quake.mess *finally* cleaned up
(I'm in Granada Hills... and was up working when it hit)
then the C/Pro with the dual.processor card and its manuals
gets set back up. I *Want* DR Dos 5 on my C/Pro. Its also
proly the only way that I'll ever stop thinking in Zilog
nmonics. 8)
............................................................
Alan Ogden, Moderator of the 'nos' conf on BIX which was the cpm conf.
ar...@BIX.com
Scott - Maxwell
>(like RT-11 uses for .PRINT). The DOS function is 09h, "print string".
>
that no one at MS could say why the $ was there. If it was for a null don't
you think they would have said so?
Scott - Maxwell
Richard Plinston
>
>>Richard_...@kcbbs.gen.nz (Richard Plinston) writes:
>
>>>'CP/M-486' is actually called
>>>(in its latest version) DR-Multiuser-DOS 7.
>
>>A few members of our surviving CP/M-Houston Users' Group (Houston, TX
>>and vicinity) delight in referring the DR-DOS vv. 5, 6, 7 as "CP/M 5,
>>6, 7" as appropriate. One fellow even patched the sign-on message
>>of his copy of DR-DOS 6 to read "CP/M 6" :-)
It is a pity that they removed the 'native' CP/M-86 support from CDOS/
DR-Multiuser-DOS in building DR-DOS. They had left it in when they
made the earlier DOS+. But all is not lost. The latest versions of
OEM DR-Multiuser-DOS are available as single-user versions, though
still not cheap.
DataPac's System Manager and IMS's and CCI's version of Multiuser-DOS
all give full preemtive multi-tasking, full support for DOS including
DPMI and still include the full CP/M-86 and MP/M-86 API as their
'native' mode support.
I can supply info on where to get these if you are interested.
Also CONUG / INDRUS is still operating. This was originally
'CONcurrent-dos User Group' specifically for CCP/M and CDOS, but
is now 'INdependent DR User Society' and includes coverage of DR-DOS,
NW-DOS and Personal Neteawe as well as continuing with DR-Multiuser-DOS
and its derivatives.
Richard Plinston
>Also CONUG / INDRUS is still operating. This was originally
>'CONcurrent-dos User Group' specifically for CCP/M and CDOS, but
>is now 'INdependent DR User Society' and includes coverage of DR-DOS,
>NW-DOS and Personal Neteawe as well as continuing with DR-Multiuser-DOS
>and its derivatives.
>
A couple of requests have come to me for this so:
INDRUS
PO Box 890086
Oklahoma City, OK 73189
FAX (405) 691-3555
European Affiliate:
INDRUS
c/o Peter vd Berg
Synetics Inc
Julianastraat 40
2771 DX Boskoop
Netherlands
INDRUS Bulletin Boards:
Eastern (407) 725-8978 (9600 v32)
Western (415) 873-6256 "
Germany 611-562047 1200/2400
Netherlands 31-1727-18517 1200/2400
Central (405) 691-4187 (9600 v32)
Membership is $US25 per year, $US35 for overseas, $US60 overseas
airmail. 6 issues per year of INDRUS Resource. VISA acceptable.
Call Central BBS for Membership or FAX or mail.
cheers
Richard N. Turner
Page 133 of the hardcover edition.
--
RNT
Chuck McManis
opportunity to look at some "alleged" MS-DOS source code (I say alleged
because while it appeared to be genuine, it was just sitting on a disk
of unknown origin so I couldn't trace it back to Bellvue.) Anyway, as
an avid CP/M hacker at the time the command.com code was eerily familiar
but twisted, then the pieces fell into place, I was looking at 8080 code
that had been run through the infamous Intel 8080 to 8086 translator
program (remember that one?). Given that evidence I was convinced that
much of MS-DOS was simply CP/M run through the translator and then
patched up.
--
--Chuck McManis All opinions in this message/article are
FirstPerson Inc. those of the author, who may or may not
Internet: cmcm...@firstperson.COM be who you think it is.
Sten Drescher
CM> Given that evidence I was convinced that much
CM> of MS-DOS was simply CP/M run through the translator and then
CM> patched up.
What version of DOS was this? I believe that DOS 1 was strongly
based on CP/M, but DOS 2 was rewritten.
--
Sten Drescher 2709 13th St #1248
s...@floyd.brooks.af.mil Brooks AFB, TX 78235
RIPEM and PGP public keys available via finger
#include <disclaimer.h>
"That's not funny."
-George Stephanopoulos, when Leon Panetta said about a group of
visiting children: "I thought they were the White House staffers,"
quoted in Newsweek.