Computer Chronicles episode
Eduardo Cavazos
In this episode of Computer Chronicles:
http://www.archive.org/details/Programm1984
Elizabeth Rather demos a Forth system. What Forth was it? PolyForth?
I'd love to give PolyForth for DOS a whirl in either Dosbox or Qemu/
MSDOS. Is it available anywhere?
As an aside, Pygmy runs well in Dosbox. Any other good old DOS Forths
you'd recommend?
Ed
Andrew Haley
> In this episode of Computer Chronicles:
>
> http://www.archive.org/details/Programm1984
>
> Elizabeth Rather demos a Forth system. What Forth was it? PolyForth?
Thanks for the link.
That's polyFORTH II.
As an aside, there were quite a lot of significant differences from
polyFORTH to polyFORTH II. I'm pretty sure a contemporary user (of
ANS Forth, say) would find polyFORTH II much more comfortable to use
than they would polyFORTH.
Andrew.
Doug Hoffman
> Hello,
>
> In this episode of Computer Chronicles:
>
> http://www.archive.org/details/Programm1984
>
> Elizabeth Rather demos a Forth system.
Yes, thanks for the link.
As one might expect, based on her posts here, Elizabeth delivered a
polished, clear presentation and projects a pleasant yet professional
image. Well done, Elizabeth. (and no, I'm not looking for a job at
Forth Inc.)
I wasn't happy that Kildall derailed things a bit by only asking about
the use of assembler in the demo program. Nothing wrong with his
question but he could have asked many other questions that would have
given the viewer better insight into Forth. But Elizabeth fielded the
question well and was even able to immediately pull up to the screen a
code definition. Maybe that was rehearsed?
-Doug
Elizabeth D Rather
Amazing! I remember that show. It was polyFORTH. I gave a full copy of
polyFORTH to Gary Kildall, hoping he would find it interesting. No idea
if he ever tried it.
polyFORTH hasn't been actively sold by FORTH, Inc. for over 10 years.
It was really great in its day, but doesn't really have much of a place
in today's software market. We offer systems for Windows and Linux, as
well as interactive cross-compilers for most popular microcontrollers.
Cheers,
Elizabeth
--
==================================================
Elizabeth D. Rather (US & Canada) 800-55-FORTH
FORTH Inc. +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
http://www.forth.com
"Forth-based products and Services for real-time
applications since 1973."
==================================================
Elizabeth D Rather
> On 8/9/10 5:46 AM, Eduardo Cavazos wrote:
>> Hello,
>>
>> In this episode of Computer Chronicles:
>>
>> http://www.archive.org/details/Programm1984
>>
>> Elizabeth Rather demos a Forth system.
>
> Yes, thanks for the link.
>
> As one might expect, based on her posts here, Elizabeth delivered a
> polished, clear presentation and projects a pleasant yet professional
> image. Well done, Elizabeth. (and no, I'm not looking for a job at Forth
> Inc.)
Thank you!
> I wasn't happy that Kildall derailed things a bit by only asking about
> the use of assembler in the demo program. Nothing wrong with his
> question but he could have asked many other questions that would have
> given the viewer better insight into Forth. But Elizabeth fielded the
> question well and was even able to immediately pull up to the screen a
> code definition. Maybe that was rehearsed?
No, nothing was rehearsed. We all talked for a bit about how the show
would proceed, but there was no rehearsal. However, I had demo'd
polyFORTH II quite a few times for other people at that point :-)
forther
> We offer systems for Windows and Linux
SVFIG
> you'd recommend?
>
> Ed
FIG Forth
There is an enhanced one here: (which I haven't tried)
<<<http://www.cdrom.com/free/Development-Forth-programming-language/
Fig-Forth-v2/63060.html>>>
You're also welcome at <<<http://www.forth.org/compilers.html>>> where
you can search for DOS.
If you can't find a working FIG Forth, please email me:
forther_at_comcast_dot_net
Aleksej Saushev
>> As an aside, Pygmy runs well in Dosbox. Any other good old DOS Forths
>> you'd recommend?
>
> FIG Forth
> There is an enhanced one here: (which I haven't tried)
> <<<http://www.cdrom.com/free/Development-Forth-programming-language/
> Fig-Forth-v2/63060.html>>>
>
> You're also welcome at <<<http://www.forth.org/compilers.html>>> where
> you can search for DOS.
>
> If you can't find a working FIG Forth, please email me:
> forther_at_comcast_dot_net
FIG Forth can't be good for too many reasons. Bury the dead.
--
HE CE3OH...
Jerry Avins
You help to keep it alive by citing its standard as current truth.
Jerry
--
Engineering is the art of making what you want from things you can get.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
John Passaniti
> In this episode of Computer Chronicles:
Someone call Mavis Beacon! There are some quick brown foxes that need
to jump over some lazy dogs!
Albert van der Horst
mina is the 16 bits version of ciforth. wina (32 bits) also runs in
(requires) a Dos box. They have the advantage of ISO compatibility and
the usage of files. You will be hard pressed to find other DOS Forths
that come with a pdf of 100+ pages and three separate clickable
indices. And if you get bored, you can upgrade to a 64 bit Forth that
can be configured to a 2 Gbyte dictionary space (on linux).
There is a whole slew of FIG material, some of it one conserved on
my site. The problem is to assess the quality. I use it mainly to
have examples of how to implement low level words, calculation words
like ``M/MOD'' or control words like ``(DO)''.
Do you want the experience of blocks on a floppy disk, and no
source control and clumsy editing?
(No advice here). Although mina/wina has a nice screen editor with
wordstar commands.
Do you want a single source file in standard assembler format?
All of the ciforth family are like that.
Do you want a plain interaction via console, redirectable input and
output, no pull down menu's?
Then all ciforth's will do. But the DOS experience is available
for all Linux forth's, including lina and notably gforth.
>
>Ed
Groetjes Albert
P.S. lina wina mina are on the site in my sig.
--
--
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- being exponential -- ultimately falters.
albert@spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst
Hugh Aguilar
I liked UR/Forth; it had a pretty good development environment and a
good block editor, and it made much better use of the 8088 memory
system than PolyFORTH did. I wrote a 65C02 cross-compiler under 16-bit
UR/Forth and the MiniForth cross-compiler (MFX) under 32-bit UR/Forth.
I don't think MS-DOS versions are still available though. I know of a
couple of bugs in the software and can provide fixes if anybody is
interested. I disassembled a lot of UR/Forth and know the inner
workings reasonably well.
I think most people in those days used FPC for MS-DOS; I didn't like
it as well as UR/Forth, but the price was right (free). TCOM is a good
non-interactive compiler for building .com programs. I have used the
non-interactive ForthCMP too, and it is quite good (and is ANS-Forth),
but it does cost $50. I have never heard of anybody using PolyFORTH;
it was astronomically expensive, and its 64K limitation made is
useless for anything except toy programs such as E.R. demo'd on that
TV show. It was also ITC, which is pretty slow.
What exactly is your interest in MS-DOS???
You are likely best off to develop your program using a 32-bit
interactive Forth system (such as Gforth) and then port it to ForthCMP
to compile a fast 16-bit version.
Hugh Aguilar
> On 8/9/10 3:03 AM, Doug Hoffman wrote:
> > As one might expect, based on her posts here, Elizabeth delivered a
> > polished, clear presentation and projects a pleasant yet professional
> > image. Well done, Elizabeth. (and no, I'm not looking for a job at Forth
> > Inc.)
>
> Thank you!
That is some pretty disgusting brown-nosing, even by C.L.F. standards.
Forth Inc. isn't even offering employment at this time, so what is the
point?
I remember 1984; that was the year I graduated from High School, and
also started programming in Forth using Tom Zimmer's Forth cartridge
for the VIC-20. This was also the year when I stopped watching
television. I believed that, being 18, I was an adult and it was time
to put away childish pursuits, so I vowed to never watch TV again. I
am now 44 and I have kept this promise. I believe that television
cheapens the world; nothing good will ever come from the flickering
demon. I'm not at all surprised that somebody like Elizabeth Rather
would appear on television. She also supported Passaniti in attacking
my software in the most vulgar manner, which is yet another example of
cheapening the world:
http://groups.google.com/group/comp.lang.forth/browse_thread/thread/c37b473ec4da66f1
P.S. --- I upgraded my novice package (which still includes symtab):
http://www.forth.org/novice.html
Albert van der Horst
Eduardo Cavazos <wayo.c...@gmail.com> wrote:
Oops. And there there is CHForth. It is still more or less the
official Forth of the Dutch FIG Chapter. It is DOS-only and the base
of several tools such as the server for e.g. ByteForth.
See www.forth.hcc.nl
>
>Ed
Paul Rubin
> In this episode of Computer Chronicles:
> http://www.archive.org/details/Programm1984
> Elizabeth Rather demos a Forth system. What Forth was it? PolyForth?
I liked the Forth segment better than the other segments, though the
Cobol one was also kind of interesting. Surprisingly the Logo one
didn't make much impression (maybe I'll watch it again). I couldn't
help chuckling at the now-dated appearances of the guys with their
1980's haircuts and clothing, but Elizabeth looked just fine.
Hugh Aguilar
> I wasn't happy that Kildall derailed things a bit by only asking about
> the use of assembler in the demo program. Nothing wrong with his
> question but he could have asked many other questions that would have
> given the viewer better insight into Forth.
In 1984, maybe 99% of all commercial software was written in assembly
language; either 65c02 or Z80. There may have been a very tiny handful
of programs written in high-level language (usually Pascal), but high-
level languages were pretty much just something for children to learn
the basics of programming on before graduating to assembly language.
At that time, Forth could have been successful if it had been marketed
as the ultimate macro-assembler. Forth assemblers really are way more
powerful than traditional macro assemblers with their prickly macro
"languages" that are little more than string expansion with a few
conditionals and maybe a very crude looping mechanism (repeat x times,
with x being a constant).
Forth Inc. largely pulled the rug out from under the Forth community
by marketing Forth as a high-level language when nobody wanted high-
level languages, and also by pushing indirect-threaded-code (ITC)
rather than subroutine threading. ITC kills Forth as a high-level
language because it is too slow. It also kills Forth as a macro
assembler because all you can do is write code words (similar to what
is done in Pascal); you can't write macros that generate machine
language and which can be used inside of colon words (as you can with
subroutine threading). By 1984, the low-memory computers of the 1970s
were a thing of the past; most personal computers had 64K or (with
bank switching) 128K of memory. ITC might have been an option on the
VIC-20 that came with 5K of RAM and supported 8K ROM cartridges, but
it was no longer necessary on the Commodore-64 that had 64K, or the
Apple-][ that had 128K of RAM plus a 256K RAM-disk, and not on the CP/
M computers that also had abundant memory.
Gary Kildall's question was completely relevant to 1984 technology.
The guy did write CP/M; he knew a lot about computer programming. He
knew that Z80 assembly (and 8088 on the horizon) was the *only*
important issue to programmers. By comparison, Elizabeth Rather was
completely out of touch with reality. She had the perfect opportunity
on that show to present Forth in a positive light, being asked the
perfect question (which less informed interviewers would not have
asked), and she completely dropped the ball. She tried to present
Forth as being a high-level language, and she ended up putting Forth
in the same category as Personal COBOL --- a joke. At that time, the
only high-level language that was getting any traction at all was
Pascal (more than C in those days), but Forth was far superior to
Pascal because of its ability to mix with assembly language.
If Forth had been marketed properly --- and that ridiculous ITC
disposed of --- Forth could have become the dominate language.
Instead, Forth became a complete failure that essentially nobody has
even heard of. The vast majority of people who have heard of Forth
even today believe that it is a purely interpreted language "like
BASIC." I blame Elizabeth Rather for Forth's failure. She has been
living in the 1970s her entire life. Even in 1984 she was out of date
with current technology. By 2010, the whole backwards-looking fiasco
has gone beyond weird.
BTW, when I was a senior in H.S. I taught Logo at a private school for
12-14 year olds. It is a toy version of Lisp that is oriented around
making funky designs using turtle graphics (somewhat like a
computerized spirograph). Shortly after H.S. I also wrote a four-
dimensional turtle graphics program in SuperForth for the
Commodore-64. It allowed travel in W, X, Y, and Z axis, and it
presented a three-dimensional view of the four-dimensional object ---
complete with perspective going to an imaginary horizon point. You
could construct a hypercube and then move your turtle around it to
view the three-dimensional protrusion from different angles. I though
it was pretty cool! Nobody else cared in the slightest. The only thing
people were spending money on was spreadsheets and, to a lesser
extent, word-processors and databases.
Andreas
> On Aug 9, 7:03 am, Doug Hoffman<glide...@gmail.com> wrote:
>> I wasn't happy that Kildall derailed things a bit by only asking about
>> the use of assembler in the demo program. Nothing wrong with his
>> question but he could have asked many other questions that would have
>> given the viewer better insight into Forth.
>
> In 1984, maybe 99% of all commercial software was written in assembly
Gosh, you must have had -200 dioptries then.
Elizabeth D Rather
from
http://en.wikipedia.org/wiki/BASIC_programming_language#Explosive_growth:_the_home_computer_era:
"In 1975, MITS released Altair BASIC, developed by Bill Gates and Paul
Allen as the company Micro-Soft, which grew into today's corporate
giant, Microsoft. The first Altair version was co-written by Gates,
Allen, and Monte Davidoff. Versions of Microsoft BASIC (also known then,
and most widely as M BASIC or MBASIC) was soon bundled with the original
floppy disk-based CP/M computers, which became widespread in small
business environments. As the popularity of BASIC on CP/M spread, newer
computer designs also introduced their own version of the language, or
had Micro-Soft port its version to their platform.
When three major new computers were introduced in what Byte Magazine
would later call the "1977 Trinity",[5] all three had BASIC as their
primary programming language and operating environment. Commodore
Business Machines paid a one-time fee for an unlimited license to a
version of Micro-Soft BASIC that was ported to the MOS 6502 in their PET
computer, while Apple II and TRS-80 both introduced new, largely similar
versions of the language. ... Most of the home computers of the 1980s
had a ROM-resident BASIC interpreter, allowing the machines to boot
directly into BASIC. Because of this legacy, there are more dialects of
BASIC than there are of any other programming language."
Brad
>
> > Elizabeth Rather demos a Forth system. What Forth was it? PolyForth?
>
>
> polyFORTH to Gary Kildall, hoping he would find it interesting. No idea
> if he ever tried it.
>
SwiftX's interactivity. I'm sure it would be fun to watch and probably
fun to make.
-Brad
Alex McDonald
> On Aug 9, 7:03 am, Doug Hoffman <glide...@gmail.com> wrote:
>
> > I wasn't happy that Kildall derailed things a bit by only asking about
> > the use of assembler in the demo program. Nothing wrong with his
> > question but he could have asked many other questions that would have
> > given the viewer better insight into Forth.
>
> In 1984, maybe 99% of all commercial software was written in assembly
> language; either 65c02 or Z80. There may have been a very tiny handful
> of programs written in high-level language (usually Pascal), but high-
> level languages were pretty much just something for children to learn
> the basics of programming on before graduating to assembly language.
Snort. You're serious? In 1984, COBOL, FORTRAN and other high level
languages were overwhelmingly the languages of choice for commercial
software. There are reckoned to be 200 billion lines of COBOL running
commercial applications today. C dominated in the systems programming
arena.
<complete bollox snipped>
Hugh Aguilar
I was talking about personal computers! I'm sure there was a lot of
COBOL on mainframes, and Fortran on mini-computers, and C on Unix
stations. In the personal computer world though, it was all 65c02 and
Z80 assembly language --- and also in the embedded-controller world.
It is true that every personal computer had a BASIC interpreter built-
in, but those were just for children to play with. They would
interpret the source-code at run-time with only some minor
"tokenizing" done at compile-time; you needed a calender to benchmark
them. There was some use of Pascal on the Apple computer, and CBASIC
on CP/M, but that was mostly for small in-house programming or
hobbyist work. I'm not aware of *any* shrink-wrapped commercial
software that was written in anything other than assembly language.
Those 8-bit computers were *slow* --- even the assembly-language
programs dragged.
Albert van der Horst
Hugh Aguilar <hughag...@yahoo.com> wrote:
<SNIP>
>
>In 1984, maybe 99% of all commercial software was written in assembly
>language; either 65c02 or Z80. There may have been a very tiny handful
I don't know in what world you're living. By 1984 I had been involved
in two large projects, calculating the output of the complete oil
pipe network of the Brent fields in the North-Sea and the support of
Dijkbouw Oosterschelde, both in FORTRAN (plus several smaller objects).
By 1984 I was a member of a team implementing a general purpose
simulation language in Unix and C.
There was virtually no assembly language used in large companies.
>of programs written in high-level language (usually Pascal), but high-
There was much talk and advocacy within Shell promoting Pascal over
FORTRAN, and little activity.
<SNIP>
Sp...@controlq.com
> In 1984, maybe 99% of all commercial software was written in assembly
> language; either 65c02 or Z80. There may have been a very tiny handful
> of programs written in high-level language (usually Pascal), but high-
> level languages were pretty much just something for children to learn
> the basics of programming on before graduating to assembly language.
Hugh, perhaps you should engage the services of an editor, or at least pay
someone to check your facts before you waste your time writing all the
reams of fiction you spew. There are machines out there which are
archiving your words, and there are impressionable minors who might
stumble upon, read and (heh heh) actually believe your bullshit.
Sheesh!
ron
> I was talking about personal computers! ...
> In the personal computer world though, it was all 65c02 and
> Z80 assembly language --- and also in the embedded-controller world.
Fascinating. I suppose my then-employer was mistaken. You see, all
the products we worked on were in "C" (except the one I worked on,
which was in assembly language). And that was on PCs, not mainframes
or unix-boxen.
Do you ever fact-check before posting?
MarkWills
With regard to *personal computers*, Hugh is exactly right.
From approx 1980 onwards there was an explosion of 'home computers' on
both sides of the Atlantic:
* Texas Instruments TI-99/4 and later TI-99/4A
* Commodore Vic 20
* Commodore 64
* Commodore 16
* Commodore Plus/4
* Commodore 128
* TRS80
* Dragon 32
* Dragon 64
* Memotech MTX512
* Lynx 96
* Sinclair ZX80
* Sinclair ZX81 (Timex in the USA)
* Sinclair ZX Spectrum
* Sinclair QL
* Mattel Aquarius
* Oric 1
* Oric Atmos
* BBC Model A
* BBC Model B
* Acorn Electron
* Atari 400
* Atari 800
* Atari 600XL
* Atari 800XL
* Amstrad CPC464
* Amstrad CPC6128
...and those are just the ones I remember from my childhood. They had
one thing in common: They *all* booted from on-board ROM, and featured
a BASIC interpreter in ROM, which was the killer-app of the day. If
you couldn't tinker in BASIC, if you couldn't type in those programs
from all those magazines, nobody was interested.
However, pretty much all commercial programs available for the above
(maybe as much as 99%?) were written in assembly language, and
supplied either on cassette, disk, or solid-state cartridge.
The first "home" machines to break this mould that I can think of
right now were the Atari ST and the Commodore Amiga. They did not
feature built-in BASIC. You had to load it from disk. However, both of
those machines were "cross-over" machines and traversed the home and
the then fledgling small-business market. At least in the UK it did.
I remember very well writing BASIC programs to POKE machine code op-
codes into RAM on my ZX Spectrum. You poked them in, and ran them. If
it worked, fine. If not, pull the power cord (no reset button!).
Mark
PS: Even the Commodore PET featured Micro-Soft Basic. In ROM.
Andrew Haley
> The first "home" machines to break this mould that I can think of
> right now were the Atari ST and the Commodore Amiga. They did not
> feature built-in BASIC. You had to load it from disk. However, both
> of those machines were "cross-over" machines and traversed the home
> and the then fledgling small-business market. At least in the UK it
> did.
>
> I remember very well writing BASIC programs to POKE machine code op-
> codes into RAM on my ZX Spectrum. You poked them in, and ran
> them. If it worked, fine. If not, pull the power cord (no reset
> button!).
>
> PS: Even the Commodore PET featured Micro-Soft Basic. In ROM.
In 1977. We're talking about the IBM PC in 1984, the year of the
Macintosh. By then there was a great deal of choice of programming
languages.
Andrew.
MarkWills
wrote:
>
> In 1977. We're talking about the IBM PC in 1984, the year of the
> Macintosh. By then there was a great deal of choice of programming
> languages.
>
> Andrew.
Sure. But nobody had an IBM PC ("Personal Computer") in 1984. Not in
the UK *home* anyway. The best you could hope for would be a clone,
something like the Amstrad PC1640 which came with GEM. If you wanted
MS-DOS you had to go an buy it yourself!
PC's didn't enter the UK home sector until ~1990 (IIRC) when the 286
came out. I remember selling MITAC 286 PCs around that time. Before
that, it was XTs, which were just rubbish!
I do agree though, on the PC platform, plenty of languages. I remember
Microsoft Quick C, and IIRC Microsoft also had a version of Cobol (DOS
based). Of course Borland had Pascal (Turbo Pascal?) etc. I remember I
used to use MS QuickBasic a lot.
However, as I was saying, it "home" computers that graced the living
rooms (plugged into the family TV) and adolescent bedrooms of the
time! A generation of 'back-bedroom programmers' fuelled the home-
computer revolution in the UK. With publishers such as Bug Byte,
Quicksilva, Imagine, Mastertronic et al making millionaires out of 15
year old school boys. It was a crazy time!
Regards
Mark
Andrew Haley
> On 13 Aug, 15:20, Andrew Haley <andre...@littlepinkcloud.invalid>
> wrote:
>>
>> In 1977. We're talking about the IBM PC in 1984, the year of the
>> languages.
>
Have you actually watched the episode? Clearly someone had an IBM PC.
> Not in the UK *home* anyway.
And what is the relevance of the UK home to this?
Andrew.
Sp...@controlq.com
> On 13 Aug, 01:31, S...@ControlQ.com wrote:
>> On Thu, 12 Aug 2010, Hugh Aguilar wrote:
>>> In 1984, maybe 99% of all commercial software was written in assembly
>>> language; either 65c02 or Z80. There may have been a very tiny handful
>>> of programs written in high-level language (usually Pascal), but high-
>>> level languages were pretty much just something for children to learn
>>> the basics of programming on before graduating to assembly language.
>>
>> Hugh, perhaps you should engage the services of an editor, or at least pay
>> someone to check your facts before you waste your time writing all the
>> reams of fiction you spew. There are machines out there which are
>> archiving your words, and there are impressionable minors who might
>> stumble upon, read and (heh heh) actually believe your bullshit.
>>
>> Sheesh!
>
> With regard to *personal computers*, Hugh is exactly right.
Refer to the term "commercial". Most commercial software publishers at
the time (early 80's) were focused upon the mainframe and minicomputers
(Vax/VMS, Data General, Prime, HP, Apollo, and about a gazillion MC-68K
S-100 bus based boxen running various early forms of Unix or Xenix -- but
most of those went out of business quite quickly). Languages were
primarily FORTRAN and COBOL, various higher level system languages (SPL,
PL/1) and odd languages like Prolog, Lisp, SmallTalk, APL, Ada
proliferated.
No doubt that this is true, but the personal computer market was not the
immediate target for the very large software publishers, and only
reluctantly (after the success of VisiCalc and dBase legitimized the
market) did they even move to support the IBM-PC.
Please note that your accurate and prolific list of incompatible and
underpowered toys was one of the reasons that commercial developers
largely stayed away. Even basic was dissimilar on each box, and with the
lack of commercial grade language compilers, it was not until IBM came out
with the PC that the commercial micro market began.
I will say, that in the months leading up to the IBM PC announcement,
there was (hopeful) speculation that the PC would be based upon the
Motorola 68000 family, owing largely to the 24bit address bus .. but alas,
the marketing weenies used the Intel 8088 with a segmented architecture,
and worse, they mated it with a hi-jacked operating system (the history of
Digital Research might have been quite different were it not for a meeting
gone awry -- I seem to recall it was Gerry's wife who threw the IBM guys
out, but that story is a little fuzzy).
> However, pretty much all commercial programs available for the above
> (maybe as much as 99%?) were written in assembly language, and
> supplied either on cassette, disk, or solid-state cartridge.
Sorry, that was not on my radar. I was doing commercial software
development using FORTRAN/SPL and working on HP Mini's at the time. Reel
to reel 1600bpi tape was common, 6250bpi less so, and various cartridge
tapes were just coming into vogue for software distribution.
>
> The first "home" machines to break this mould that I can think of
> right now were the Atari ST and the Commodore Amiga. They did not
> feature built-in BASIC. You had to load it from disk. However, both of
> those machines were "cross-over" machines and traversed the home and
> the then fledgling small-business market. At least in the UK it did.
Atari ST and Amiga were both 16bit (MC68000) based, and had a great deal
of potential. They were focused primarily upon games play, however, and
lacked suitable hard drives for commercial work.
>
> I remember very well writing BASIC programs to POKE machine code op-
> codes into RAM on my ZX Spectrum. You poked them in, and ran them. If
> it worked, fine. If not, pull the power cord (no reset button!).
>
> Mark
>
> PS: Even the Commodore PET featured Micro-Soft Basic. In ROM.
And the SuperPet was 6809 based ... (as I believe the RadioShack CoCo --
colour computer was) a much improved 8-bit processor (more index registers
IIRC).
Cheers,
Rob Sciuk
Thomas Pornin
> From approx 1980 onwards there was an explosion of 'home computers' on
> both sides of the Atlantic:
> However, pretty much all commercial programs available for the above
> (maybe as much as 99%?) were written in assembly language
I am not so sure. I encountered quite a few games which were sold for
actual money (so they were "commercial") and were written in BASIC
(possibly with one or two graphical routine written in assembly, but the
core of the code was in BASIC). RAM was tight and, thanks to the BASIC
interpreter already in ROM, a BASIC program was a compact representation
(usually more compact than the equivalent assembly code).
No Pascal or C, though.
--Thomas Pornin
Richard Owlett
> On 13 Aug, 01:31, S...@ControlQ.com wrote:
>> On Thu, 12 Aug 2010, Hugh Aguilar wrote:
>>> In 1984, maybe 99% of all commercial software was written in assembly
>>> language; either 65c02 or Z80. There may have been a very tiny handful
>>> of programs written in high-level language (usually Pascal), but high-
>>> level languages were pretty much just something for children to learn
>>> the basics of programming on before graduating to assembly language.
>>
>> Hugh, perhaps you should engage the services of an editor, or at least pay
>> someone to check your facts before you waste your time writing all the
>> reams of fiction you spew. There are machines out there which are
>> archiving your words, and there are impressionable minors who might
>> stumble upon, read and (heh heh) actually believe your bullshit.
>>
>> Sheesh!
>
> With regard to *personal computers*, Hugh is exactly right.
Malarkey!
Define "personal".
In mid/late 70's I personally knew 2 individual who had recycled
PDP-8's at home (8I's IIRC) and one individual with a PDP-11/45
in a home office. (He had a field service contract.) (My
mail-stop at time was ML5-5 ;)
If you class max of TWO users as "personal", I knew of a KI-10
being used by two MIT post-docs doing large vocabulary discrete
speech recognition research.
~85 I was toting my Kaypro-10 {with Wordstar, Calcstar, dBaseII)
to a construction job site. At the time my home machine (S100,
Z80, 32k, 2-8" floppies) had FORTRAN, COBOL, and FORTH (LMI?).
By 1990 I was on my 3rd PC/"PC compatible".
And I was always one of last of my peers getting "modern" hardware ;)
Or are you choosing to define "personal" as what *YOU* had or
might afford?
Hugh Aguilar
wrote:
How many people could afford a Macintosh in 1984? They were only used
for desktop publishing and were very expensive. Also, all Mac software
had to be developed on the Apple Lisa (and a Lisa was *extremely*
expensive); there were no compilers that ran on the Mac itself. The
Mac was used only by non-technical California-type people, most of
whom were employed at printing out glitzy sales brochures.
The IBM PC may have existed in 1984, but it was also extremely
expensive. It wouldn't be until much later when IBM lost legal control
over the BIOS that inexpensive clones began to be produced in Taiwan.
After I got out of H.S. I got a job at the Santa Barbara Public Works.
They used a multi-tasking CP/M computer. They used Dbase-II, although
I didn't get involved in that. All the secretaries typed out letters
and other documents on WordStar. I learned WordStar from one of the
secretaries; that was the first word-processor I had used. At home I
had "Paperback Writer" on my Commodore 64. The C64 video was only 40
characters wide. To get a full page displayed, the software had to
generate the text as graphics, which was horribly slow. By comparison,
the CP/M video (on the KayPro) was 64 characters wide which was enough
to display a page of text (with a two character margin on each side).
In 1984 the Atari ST was available for $1000, but it was mostly used
for music (it had MIDI built-in) and games. At that time the Commodore
Amiga was still vapor-ware. When it came out later it was used mostly
for games, but somewhat also for music. The Atari ST, Commodore Amiga
and MS-DOS XT computers were the transition from the dark-ages of the
8-bit 65c02 and Z80 computers to the middle-ages of the 16-bit 8088
and 68000 computers. Only with the 16-bit computers did high-level
languages become viable (Jforth on the Amiga). On the 65c02 and Z80
though, as I said, assembly language was mandatory; those 8-bit
processors really didn't support high-level languages efficiently
enough to be viable.
The Amiga was a good game machine, what with hardware support for
sprites, but it had low-resolution graphics. People wanted high-
resolution graphics for supporting GUI software and for displaying
digital photographs. I considered this to be a big mistake at the
time, and I still do. The GUI concept was originally developed on the
Mac for use by graphical artists who had no technical inclination
whatsoever. It was a mistake for the GUI to become general-purpose.
Even today I absolutely refuse to use word-processors. The last word-
processor I actually used seriously was WordStar on the CP/M computer.
I experimented a little with WordPerfect but didn't like it. I've
never used Microsoft Word and have no intention of ever doing so. I
just use LaTeX for everything.
Paul Rubin
> In 1977. We're talking about the IBM PC in 1984, the year of the
> Macintosh. By then there was a great deal of choice of programming
> languages.
From http://en.wikipedia.org/wiki/BDS_C :
BDS C (or the BD Software C Compiler) ... [was] first released in 1979.
A number of important commercial CP/M products were written in the
BDS C subset of the C language, including Mince and Scribble from
Mark of the Unicorn, and most of the software in the Perfect
Software suite including Perfect Writer, PerfectCalc, PerfectSpeller
and PerfectFiler (which suite was bundled with the Kaypro).
Prior to BDS there were apparently heavier-weight C compilers used for
development.
Hugh Aguilar
> According to MarkWills <markrobertwi...@yahoo.co.uk>:
Those ROM-based BASICs interpreted the source-code. It was impossible
to distribute a BASIC program without also giving the customer the
source-code.
I had a text-adventure game for the VIC-20 that was written in BASIC.
I doubt that I paid for it; it may have given away as a freebie,
although I don't remember the details. I do remember that I couldn't
figure out how to get past a certain point in the game, so I cheated
by looking through the source-code to find out what I was supposed to
type.
The most charitable thing that can be said about those ROM-based BASIC
interpreters was that they were "educational." They had no practical
value. Essentially all software was written in assembly language.
Hugh Aguilar
> Andrew Haley <andre...@littlepinkcloud.invalid> writes:
> > In 1977. We're talking about the IBM PC in 1984, the year of the
> > Macintosh. By then there was a great deal of choice of programming
> > languages.
>
>
> BDS C (or the BD Software C Compiler) ... [was] first released in 1979.
>
> A number of important commercial CP/M products were written in the
> BDS C subset of the C language, including Mince and Scribble from
> Mark of the Unicorn, and most of the software in the Perfect
> Software suite including Perfect Writer, PerfectCalc, PerfectSpeller
> and PerfectFiler (which suite was bundled with the Kaypro).
>
> Prior to BDS there were apparently heavier-weight C compilers used for
> development.
Perfect Writer was a good word-processor. It used LaTeX internally.
The reason why Perfect Writer failed to oust WordStar was that Perfect
Writer was too slow. You could make nicer-looking documents with
Perfect Writer than with WordStar, but the actual operation of the
software was an exercise in patience. If you paid your secretary by
the hour, WordStar was the only viable option.
Perfect Writer failed *because* it was written in C and hence was
slower than WordStar that was written in assembly language.
Richard Owlett
> On Aug 13, 8:20 am, Andrew Haley<andre...@littlepinkcloud.invalid>
> wrote:
[snip]
>>
>> In 1977. We're talking about the IBM PC in 1984, the year of the
>> Macintosh. By then there was a great deal of choice of programming
>> languages.
>>
>> Andrew.
>
> How many people could afford a Macintosh in 1984? They were only used
> for desktop publishing and were very expensive. Also, all Mac software
> had to be developed on the Apple Lisa (and a Lisa was *extremely*
> expensive); there were no compilers that ran on the Mac itself. The
> Mac was used only by non-technical California-type people, most of
> whom were employed at printing out glitzy sales brochures.
>...
Not bothered by facts are you.
California people???
So intersection of I90 and NY15 is in CA now?
The Arch/Eng firm I worked for use a Mac for a large portion of
their preliminary drawings. I don't recall what was the
production CAD system at the time was. The land surveyors had the
old PDP11-45 based system. It had a very Forth-like subsystem
which I used to process field notes into check drawings. All I
knew of Forth at time was a recent "Byte" issue with some FORTH
coverage. All the "plot" routines were part of the language.
Paul Rubin
> Perfect Writer was a good word-processor. It used LaTeX internally.
That PW "used LaTeX internally" is an astonishing claim and I'd be
interested in seeing documentation. Otherwise I don't believe it.
Alex McDonald
LaTeX in 64k??? Astonishing.
Here's a review. http://www.atarimagazines.com/creative/v9n6/55_Perfect_Writer.php.
I seem to remember that it was written in Pascal?
These magazines give a real flavour of the early hobbyist market
http://www.atarimagazines.com/. There are some interesting articles
here too; here's a short Forth related review
http://www.atarimagazines.com/compute/issue25/053_1_REVIEW_QS_FORTH_FOR_ATARI.php
And, for the fan of old kit, here's Mark Wills' list in an earlier
post in chronological order http://oldcomputers.net/index.html. Just
look at those prices...
Alex McDonald
A colleague of mine had 3 phase power and an IBM 370/158 in the
basement with a pair of 2314 disk drives and a 1403 "flag" printer.
She only fired up when she was flush; it cost a lot in juice to keep
the beast running. The ultimate in personal, if not really portable.
Alex McDonald
The most charitable thing I can say of your posts is that they contain
a large number of inaccuracies, and what are becoming trademark not-so-
humble opinions masquerading as "facts".
Hugh Aguilar
> On Aug 13, 10:56 pm, Paul Rubin <no.em...@nospam.invalid> wrote:
>
> > Hugh Aguilar <hughaguila...@yahoo.com> writes:
> > > Perfect Writer was a good word-processor. It used LaTeX internally.
>
> > That PW "used LaTeX internally" is an astonishing claim and I'd be
> > interested in seeing documentation. Otherwise I don't believe it.
>
> LaTeX in 64k??? Astonishing.
I think they trimmed LaTeX down considerably. There was no support for
math equations. They were printing on dot-matrix printers after all. I
think Perfect Writer also used overlays; a hard-disk was pretty much
required for any kind of speed, as it would often freeze up and whir
the disk. I said that commercial software was written in assembly
language in 1984, and Perfect software is the exception that proves
the rule. At the time there was a lot of ballyhoo about how the
Perfect software was ushering in a new age, and that in the future CP/
M software would be written in high-level languages just like Unix
software, and would bring Unix technology (such as LaTeX) to CP/M. As
a practical matter though, Perfect software was too slow to be usable
and this just proved the point that assembly language was a must.
WordStar continued to dominate until CP/M itself eventually died out.
The WordStar people wrote an MS-DOS version in assembly language, but
it lacked the features of WordPerfect that was written in a high-level
language (C, I think), and so WordPerfect became the new standard.
I may still have my Perfect Suite software, although it would be at
the back of the storage closet if I do. I dumped a lot of CP/M
software at the thrift store, although I kept some of the good stuff.
I also dumped all of the Kaypro machines, except for the Kaypro-10
which I kept.
P.S. --- I had Borland Sprint under MS-DOS too. I used it to write the
MFX documentation. I had a PostScript cartridge on my HP laser
printer. The PostScript interpreter was written in 8051 assembly
language and it was very slow. Printing the document took hours.
Elizabeth D Rather
> Hugh Aguilar wrote:
>> On Aug 13, 8:20 am, Andrew Haley<andre...@littlepinkcloud.invalid>
>> wrote:
>
> [snip]
>>>
>>> In 1977. We're talking about the IBM PC in 1984, the year of the
>>> Macintosh. By then there was a great deal of choice of programming
>>> languages.
>>>
>>> Andrew.
The IBM PC was introduced in 1981. Several versions of Forth were
available for it by the summer of 1982, as were a number of other languages.
>> How many people could afford a Macintosh in 1984? They were only used
>> for desktop publishing and were very expensive. Also, all Mac software
>> had to be developed on the Apple Lisa (and a Lisa was *extremely*
>> expensive); there were no compilers that ran on the Mac itself. The
>> Mac was used only by non-technical California-type people, most of
>> whom were employed at printing out glitzy sales brochures.
>> ...
>
> Not bothered by facts are you.
> California people???
> So intersection of I90 and NY15 is in CA now?
> The Arch/Eng firm I worked for use a Mac for a large portion of their
> preliminary drawings. I don't recall what was the production CAD system
> at the time was. The land surveyors had the old PDP11-45 based system.
> It had a very Forth-like subsystem which I used to process field notes
> into check drawings. All I knew of Forth at time was a recent "Byte"
> issue with some FORTH coverage. All the "plot" routines were part of the
> language.
Creative Solutions worked with Apple before the Mac was introduced, and
Don Colburn exhibited MacForth at the trade show in Feb. 1984 when the
Mac was first shown (having just gone on sale). I was there exhibiting
FORTH, Inc.'s PC product. Mac sales reached 100,000 in the first 6
months, which is unimpressive today, but was astonishing at the time.
Cheers,
Elizabeth
--
==================================================
Elizabeth D. Rather (US & Canada) 800-55-FORTH
FORTH Inc. +1 310-491-3356
5155 W. Rosecrans Ave. #1018 Fax: +1 310-978-9454
Hawthorne, CA 90250
Richard Owlett
I noticed you conveniently ignored Mr. Rubin's comment which said
Bill Leary
news:98e75357-c4af-445f...@i13g2000yqd.googlegroups.com...
> In 1984, maybe 99% of all commercial software was written in
> assembly language; either 65c02 or Z80.
Not anywhere I was working. Nor anywhere the people I worked with were or
had been working. Embedded systems, yes. But most commercial software I
worked on or with was in some high level language or other. BASICs, C,
FORTRAN, COBOL, etc. Pascal came along shortly there after.
- Bill
Eduardo Cavazos
> As an aside, Pygmy runs well in Dosbox. Any other good old DOS Forths
> you'd recommend?
A related question...
It is often mentioned that Forth was commonly used as a standalone
system on mini and micro computers back in the 70s.
I'd like to try out some standalone Forths for 32-bit x86. It would
also be nice if Forth works in a vm/emulator.
So far, I've found Chuck's ColorForth (COLOR.COM). I was able to bring
this up under Qemu/DOS however the 'save' word doesn't seem to work. I
found a GLcolorforth that seems to have a working save, however that
isn't actually standalone anymore (Linux hosted).
I also got Sean Pringle's Enth to run under bochs. It doesn't seem to
work under Qemu.
Suggestions for other standalone systems welcome.
Ed
none
> > BASIC.
> > I doubt that I paid for it; it may have given away as a freebie,
> > although I don't remember the details. I do remember that I couldn't
> > figure out how to get past a certain point in the game, so I cheated
> > by looking through the source-code to find out what I was supposed
> > to type.
> >
> > The most charitable thing that can be said about those ROM-based
> > BASIC
> > interpreters was that they were "educational." They had no practical
> > value. Essentially all software was written in assembly language.
ROM based BASICs where of educational value by showing us //now// how
much could be done with a "no practical value" rom based language ;-)
As an example, I have worked for a water adduction firm in early 90's.
Their control center had been converted to PC based, C written
applications only a few years before, but their previous control
software had run for years on a bunch of Apple IIs. Each Apple II had
an expansion box attached to it, and a color, high resolution, 384 kb of
ram vector graphics card and hires color monitor. Those cards where the
as thick as 3 ordinary apple cards.
The expansion boxes were filled with interface cards which polled
automatons.
This control center controlled the water pumps, water pressure, etc for
a relatively large region, displaying levels, pressure, alerts, and all
the like, and printing the center log.
All was written in BASIC with peeks and pokes to handle the interface
cards and vector graphics.
As a side effect, they had extracted a whole Apple II with interface
cards, modem, floopy drive, keyboard, and 4 or 5 inch display into a
technical case. This was one of their automaton diagnostic tools in case
one would refuse to communicate with the center.
BTW, some BASIC dialects allowed to save in a protected manner which
prevented to read the source code. I don't have a name just now, though.
Have a nice day
jbu
Dennis
> In 1984, maybe 99% of all commercial software was written in assembly
You have gotten plenty of responses to debunk your assertion, and here
is just one more. You can reject it, or let it amuse you. With your
aversion to word processors, I suspect the former.
http://en.wikipedia.org/wiki/WordPerfect
What isn't mentioned in that article is that WordPerfect was written
in Forth. I had a copy of SSI Forth at one point and my understanding
was that they had to abandon Forth when they went to Windows (perhaps
this was the 5.1 to 5.2 transition that the article mentions).
It most certainly was not written in assembly.
Just another fact, tainted with rumor. ;)
DaR
Hugh Aguilar
> > > The most charitable thing that can be said about those ROM-based
> > > BASIC
> > > interpreters was that they were "educational." They had no practical
> > > value. Essentially all software was written in assembly language.
>
> ROM based BASICs where of educational value by showing us //now// how
> much could be done with a "no practical value" rom based language ;-)
Those ROM-based BASIC interpreters didn't do any compilation. They
"tokenized" the source-code. Words such as PRINT, FOR, NEXT, etc.,
were converted into single-byte tokens. When the program was running,
it would find the token in a look-up table and execute the
corresponding machine-language procedure. By no stretch of the
imagination could this be considered compilation. The tokenized
program could be converted back into source-code simply by expanding
the tokens into strings (PRINT, FOR, NEXT, etc.); this is exactly what
happened when you typed LIST to look at your program; you got your
source-code back exactly the way that you wrote it.
What Forth Inc. was doing in PolyFORTH with their "indirect threaded
code" (ITC) was essentially tokenizing the source-code, except that
they had 16-bit tokens rather than 8-bit tokens. You could have more
than 128 tokens. You could even write your own tokens using CODE. By
no stretch of the imagination could this be considered compilation
however. The code generated was mind-numbingly inefficient. The NEXT
code on a 65c02 or Z80 was very lengthy. As an estimate, I would say
that an ITC Forth spends over 90% of its time executing NEXT, DOCOLON,
and EXIT, and less than 10% doing actual work. Also, ITC code tends to
waste a lot of memory. For example, consider the common code sequence
of a literal number followed by a +. In ITC, the literal number takes
up 4 bytes. That is 2 bytes for the LIT token and 2 bytes for the
number itself. The + token takes up another 2 bytes, so your code
sequence consumes 6 bytes total. By comparison, if you were compiling
into machine-code, you would use much less memory. On a processor such
as the Z80 or the 8088, the top-of-stack (TOS) would be held in a
register (HL or BX), and the addition of a literal value to this
register could be accomplished with a single machine-language
instruction. On the 65c02 the machine-code would be 7 bytes (on my
compiler: LDA(2), LDY(2), JSR(3)). On the 65c02 ITC could save a
little bit of memory, but on the Z80 and 8088 ITC generally consumed
more memory. On all processors, ITC is about an order of magnitude
slower than machine-code.
What I remember about the 1980s (and the 1990s), is that Forth had an
extremely bad reputation in regard to efficiency. The term "Forth
compiler" was considered to be an oxymoron. Nobody considered
PolyFORTH to be a compiler at all. Forth was pretty much in the same
category as those ROM-based BASIC interpreters with their tokenization
of the source-code. Way back in 1984, when I was still in H.S., I read
the book: "Threaded Interpretive Languages." Even at the age of 18
with my rudimentary understanding of assembly-language, I could see
that ITC was mind-numbingly inefficient. I just assumed that ITC was
an expedient measure that would soon be discarded, and Forth would
become an actual compiled language. Everybody assumed that this was
going to happen, but it didn't happen. Forth Inc. seemed to be fixated
on ITC and continued promoting ITC year after year after year. By the
late 1980s early 1990s, Forth Inc. had convinced everybody in the
world that Forth was inherently an interpreted language, and that
compilation of Forth was impossible. By that time, C and Pascal
compilers were readily available, and they actually compiled to
machine-language. Also, PolyFORTH on the 8088 was limited to 64K. The
application code, the application data, the Forth system itself, and
the dictionary, *all* fit inside of a single 64K segment. By
comparison, the C and Pascal compilers had their Small memory-model in
which the application data went into one 64K segment and the
application code went into another 64K segment, and the compiler and
dictionary were completely external. All in all, PolyFORTH was just a
weird joke. The problem was that Forth Inc. *owned* the name "Forth."
In most people's minds, Forth Inc. defined what Forth was. When Forth
Inc. told the world that Forth compilation was impossible, and that
access to more than 64K in a Forth system was impossible, everybody
believed them. It was assumed that Forth would never be any more than
what Forth Inc. said that it could be. Forth's reputation just went
down the drain. The idea of using a Forth ITC system to actually write
an application was not taken seriously by anybody. Nobody took
Applesoft BASIC or GW-BASIC seriously either. There may have been a
very few applications written using these interpreters, but they
weren't really considered to be a viable option.
Hugh Aguilar
What I said was that in 1984, 99% of commercial software was written
in assembly language. WordPerfect didn't exist at that time. Although
16-bit systems (MS-DOS and Mac) did exist, they were very expensive.
The vast majority of people were still using 8-bit systems (CP/M on a
Z80 or the several 65c02 home computers). On these 8-bit systems,
assembly language was necessary; the 65c02 and Z80 really didn't
support high-level languages well enough to be a viable option.
I said that WordStar dominated the CP/M market because it was written
in assembly language and was reasonably fast. Perfect Writer (written
in C) never gained much traction even though it generated nicer-
looking documents --- because it was too slow. In 1984, 65c02 and Z80
assembly language was a must for any commercial program. The Perfect
software was the exception that proves the rule --- it was written in
C, but it also failed for exactly that reason.
The times were changing though! Within only a few years the 16-bit
systems displaced the 8-bit systems. On MS-DOS, WordPerfect (written
in C) became the dominate word-processor. Writing in C allowed
WordPerfect to have more features than WordStar and generate nicer-
looking documents. The C code was fast enough for everybody on the 16-
bit processor, and so WordStar quickly disappeared. I read an article
at the time that described WordStar's plummet into obscurity as "the
death of assembly-language." The article said that, in the old days of
the Z80, assembly language had been a must, but in the brave new world
of the 8088/8086, we could forget about assembly language and program
in C or Pascal just like the big boys of the Unix world. I don't
remember what magazine this was though, so I can't reference it.
> It most certainly was not written in assembly.
I already said that WordPerfect was the flagship of the new generation
of software written in C rather than assembly language. In fact,
*everything* that I said in this post, I had already said in previous
posts. Now I'm just repeating myself. That is boring for me, so I can
only imagine what effect it is having on everybody else.
As for early versions of WordPerfect being written in Forth, I never
heard of that. I thought it was always written in C. Can you
substantiate this?
Hugh Aguilar
> Hugh Aguilar wrote:
> > On Aug 13, 4:34 pm, Alex McDonald<b...@rivadpm.com> wrote:
> >> On Aug 13, 10:56 pm, Paul Rubin<no.em...@nospam.invalid> wrote:
>
> >>> Hugh Aguilar<hughaguila...@yahoo.com> writes:
> >>>> Perfect Writer was a good word-processor. It used LaTeX internally.
>
> >>> That PW "used LaTeX internally" is an astonishing claim and I'd be
> >>> interested in seeing documentation. Otherwise I don't believe it.
>
> >> LaTeX in 64k??? Astonishing.
>
> > I think they trimmed LaTeX down considerably.
>
> "That PW "used LaTeX internally" is an astonishing claim and I'd
> be interested in seeing documentation. Otherwise I don't believe
> it."
I'm not ignoring his disputation of my claim. I said that I may or may
not yet have my copy of Perfect Writer in storage. If I can find it, I
will substantiate my claim that Perfect Writer used LaTeX internally.
Diving into my storage closet looking for some old CP/M software isn't
exactly my highest priority, but I'll get around to it. I waste a lot
of time posting messages on C.L.F. already, so why not also spend an
afternoon pulling dusty cardboard boxes out of a storage closet?
Roger Ivie
> What I said was that in 1984, 99% of commercial software was written
> in assembly language. WordPerfect didn't exist at that time.
WordPerfect started out on DG minicomputers in 1979. It was ported to
PCs in 1982.
The DG bit I already knew, dates are from Wikipedia. They have a nice
table of dates and architectures; by 1984, WordPerfect was already at
version 4.0.
> On these 8-bit systems,
> assembly language was necessary;
The CP/M-80 operating system itself was written in PL/M.
--
roger ivie
ri...@ridgenet.net
MarkWills
> Malarkey!
> Define "personal".
>
OK. A machine that is in the financial reach of the average 1984
family. A PC was not. A ZX81 was.
> In mid/late 70's I personally knew 2 individual who had recycled
> PDP-8's at home (8I's IIRC) and one individual with a PDP-11/45
> in a home office. (He had a field service contract.) (My
> mail-stop at time was ML5-5 ;)
And? They didn't *purchase* them as personal computers did they? One
had a *recycled* PDP-8 (so, previously bought and paid for elsewhere)
and the other was using a PDP-11/45 as a business asset. Not a
personal computer IHMO ;-)
>
> Or are you choosing to define "personal" as what *YOU* had or
> might afford?
Yes, I am. My father was a janitor in a local shopping centre. Not
highly paid. What we could afford as a personal computer in 1984, was
very comparable to what others could afford.
PC's weren't even used as business machines in the UK in 1984. They
were still using minis with terminals. PC's started to be used in UK
offices around 1989-1990. All IBM clones, mostly made by Amstrad. They
ran Sage payroll and Sage accounts in small businesses, and that was
about it! The secretary had full use of it. The MD wasn't interested!
The rest of the country was running Amiga 500's and Atari STs in 1989.
MarkWills
>How many people could afford a Macintosh in 1984?
Exactly. Practically nobody. And exactly the same with the PC. Which
was a P.O.S anyway.
MarkWills
Forthos. http://www.forthos.org
John Passaniti
> The most charitable thing I can say of your posts is that they
> contain a large number of inaccuracies, and what are becoming
Hugh's revisionist history doesn't seem to be the result of purposeful
deception. For example, he may honestly believe PerfectWriter
embedded LaTeX and that HP PostScript cartridges interpreted
PostScript on a 8051. Of course, there is no evidence for either
claim (and indeed, for most of what Hugh writes), and he'll
undoubtedly start downgrading these claims. In fact, he already has
downgraded his claim about PerfectWriter to "I think they trimmed down
LaTeX considerably," suggesting that they tossed out features (such as
equations), and I'm sure if someone with more familiarity with LaTeX
starts enumerating every feature it had and asking if PerfectWriter
had that feature, the answer will be a long string of "no" and "not
fully" and "not really."
So where does Hugh get these ideas? My guess is that PerfectWriter's
formatter shared some syntax with LaTeX (or more likely TeX), and so
in his mind it *was* LaTeX. And a PostScript interpreter running on a
8051 and stuffed into a cartridge? Well, no. HP called those
cartridges ROMs for a reason-- they were, and the PostScript
interpreter ran on the controller board, which were all 68k-family
processors. So what about this 8051? I don't have such a cartridge
handy anymore to open it up, but if there was an 8051 in there, it was
likely had some support role-- such as a controller for banking logic
for the ROMs or othersuch.
There is a consistency whenever Hugh offers one of his blustery
claims-- when you keep pressing him on it, he'll either downgrade the
claim or come up with a distraction to divert away from the claim.
Alex McDonald
I was using PCs as a "business machine" in 1984. You're suffering from
Aguilar's Disease; the making of statements based on personal
experience extrapolated to describe everyone's experience.
Hugh Aguilar
> I wasn't happy that Kildall derailed things a bit by only asking about
> the use of assembler in the demo program.
I remember a story about a lady driving through the country-side. She
sees a hick standing on a ladder leaning up against an apple tree. He
is holding a goat cradled in his arms, and the goat is eating the
apples off the tree. She stops and asks him: "What are you doing?" He
says: "I'm feeding my goat, of course." She asks: "Isn't it a waste of
time to do it that way?" The hick looks at her as if she were daft,
and he says: "Goats don't care about time!"
That Computer Chronicles interview would have been funnier if Gary
Kildall had pointed out that PolyFORTH "compiles" to indirect-threaded-
code, and the computer spends over 90% of its time executing NEXT,
DOCOLON and EXIT. He could have asked: "Isn't it a waste of time to do
it that way?" Elizabeth Rather could then have replied: "Computers
don't care about time!" LOL
MarkWills
> You're suffering from
> Aguilar's Disease; the making of statements based on personal
> experience extrapolated to describe everyone's experience.
>
Possibly. You may indeed be correct. However, respectfully, in my
opinion I feel I was pretty well placed to make the observation that I
described.
I left school in 1985. I went directly to work for a small computer
firm in my home town of Shrewsbury, Shropshire (England) where we
dealt primarily with business clients. I mean small businesses. Pretty
much all of them were single-computer businesses, except for the
town's solicitors. We sold a few PCs (Amstrad PC1640's) but I would
say we sold 100 Atari STs for every PC we sold in 1985/6. Most were
sold as word processors.
The killer app for PCs then was Sage Accounting. In 1989, I left the
company and set up on my own as a computer repair man/child. I'd sold
most of the computers in the town and I did very well selling new hard
disks, and hard disk controller cards (yuck) for those vile XT PCs.
By *far* the most popular office machine, was the Amstrad PCW8256 and
later the 512 - CP/M based, which ran the bundled LocoScript word
processor. The daisy wheel printer shook the desk. I sold literally
hundreds of them and became very proficient at replacing the VDU
driver boards, and the hammer in the daisy wheel printer! We couldn't
sell PCs. They were just too darned expensive at the time.
I wasn't until approx '91 IIRC that PC's seemed to catch on. The 286
processor and VGA graphics was the swinger. Before that, they were
just payroll work-horses.
Ah. Those were the days!
MarkWills
Illustrates my point nicely. Thank you.
As I said, they were business machines. They certainly were NOT
"personal" computers.
In the early 90's, those individuals that bought a PC from me (I used
to sell MITAC 286 PCs with a nice Turbo button on the front) ALL used
them as business machines. Payroll, DBase, SuperCalc and the ilk.
Maybe a golf game would be found hiding away on the hard disk
somewhere. But that was it.
Hugh Aguilar
> > Or are you choosing to define "personal" as what *YOU* had or
> > might afford?
>
> Yes, I am. My father was a janitor in a local shopping centre. Not
> highly paid. What we could afford as a personal computer in 1984, was
> very comparable to what others could afford.
As I mentioned, I was working at the Santa Barbara Public Works
department in 1985 after getting out of High School. They used a multi-
user CP/M (I think it was called MP/M) system running on a Z80
computer. All the secretaries had dumb terminals hooked up to this 8-
bit computer! That was really the Dark Ages of computers!
The Public Works department also had a PDP-11 that was used to control
the traffic lights throughout the city. At the city college I took a
class in assembly language, which was PDP-11 at that time (I still
have the textbook). The college used a VAX however, and just simulated
the PDP-11 --- by that time the PDP-11 itself was largely obsolete.
There was no IBM-PC to be found anywhere at the Public Works
department. They had more of a budget that Mark's dad the janitor did,
but they yet were unable to afford the IBM-PC --- that was pretty
typical in those days.
> The rest of the country was running Amiga 500's and Atari STs in 1989.
Those were actually pretty cool computers. I wish they had achieved
that much popularity here in America. I have heard that the Amiga is
still being used in Germany yet today, and that they have upgraded to
the PowerPC. Is this true?
Alex McDonald
By 1984 (Hugh's reference point) I had been programming professionally
for more than 5 years, mainly mainframes (COBOL, BAL, SNOBOL and APL)
but also (horrors!) BASIC on the early IBM PCs to interface with
mainframes and 3270 data streams; PCs were far far cheaper than the
IBM graphical (not the 24*80 green screens, but the 3279) terminals
available at that time. IBM PCs were, quite literally, all over the
place in most large UK organisations by the mid 80s as terminals.
I also worked with the Xerox Star, a system so far ahead of its time
that its feature list would read, even today, as more than cool. All
Lisp and Smalltalk; not a squirt of assembler to be seen.
Yes, the home/hobbyist market was different. But generalisations will
get knocked down; don't take Hugh's revisionist (and frequently
bilious) viewpoint as gospel. It's not.
Alex McDonald
This started as an assertion about commercial software. Not early
hobbyist or home PCs.
Andrew Haley
I wondered who was going to be the first to point this out. :-)
Andrew.
Alex McDonald
> On Aug 14, 1:20 pm, MarkWills <markrobertwi...@yahoo.co.uk> wrote:
>
> > > Or are you choosing to define "personal" as what *YOU* had or
> > > might afford?
>
> > Yes, I am. My father was a janitor in a local shopping centre. Not
> > highly paid. What we could afford as a personal computer in 1984, was
> > very comparable to what others could afford.
>
> As I mentioned, I was working at the Santa Barbara Public Works
> department in 1985 after getting out of High School. They used a multi-
> user CP/M (I think it was called MP/M) system running on a Z80
> computer. All the secretaries had dumb terminals hooked up to this 8-
> bit computer! That was really the Dark Ages of computers!
>
> The Public Works department also had a PDP-11 that was used to control
> the traffic lights throughout the city. At the city college I took a
> class in assembly language, which was PDP-11 at that time (I still
> have the textbook). The college used a VAX however, and just simulated
> the PDP-11 --- by that time the PDP-11 itself was largely obsolete.
The early VAXes (like the VAX-11) ran the PDP-11 instruction set; no
simulation was required. The PDP-11 was not obsolete in 1985; DEC
introduced the PDP-11/94 in 1990 (http://research.microsoft.com/en-us/
um/people/gbell/digital/timeline/1990-2.htm). Check your "facts".
>
> There was no IBM-PC to be found anywhere at the Public Works
> department. They had more of a budget that Mark's dad the janitor did,
> but they yet were unable to afford the IBM-PC --- that was pretty
> typical in those days.
What position did you hold in the Public Works department, and did you
personally hunt high and low for them? Dang it, would you have even
*recognized* one, since (a) you didn't have one, (b) no-one you knew
had one and (c) where you worked didn't have one?
Alex McDonald
> On Aug 13, 7:05 pm, Alex McDonald <b...@rivadpm.com> wrote:
>
> > The most charitable thing I can say of your posts is that they
> > contain a large number of inaccuracies, and what are becoming
> > trademark not-so-humble opinions masquerading as "facts".
>
> Hugh's revisionist history doesn't seem to be the result of purposeful
> deception.
Purposeless deception. Mixed in with a venomous dislike of Elizabeth
Rather, ITC, Forth Inc., homosexuality, and learning from others.
Attaboy.
> For example, he may honestly believe PerfectWriter
> embedded LaTeX and that HP PostScript cartridges interpreted
> PostScript on a 8051. Of course, there is no evidence for either
> claim (and indeed, for most of what Hugh writes), and he'll
> undoubtedly start downgrading these claims. In fact, he already has
> downgraded his claim about PerfectWriter to "I think they trimmed down
> LaTeX considerably," suggesting that they tossed out features (such as
> equations), and I'm sure if someone with more familiarity with LaTeX
> starts enumerating every feature it had and asking if PerfectWriter
> had that feature, the answer will be a long string of "no" and "not
> fully" and "not really."
>
> So where does Hugh get these ideas? My guess is that PerfectWriter's
> formatter shared some syntax with LaTeX (or more likely TeX), and so
> in his mind it *was* LaTeX. And a PostScript interpreter running on a
> 8051 and stuffed into a cartridge? Well, no. HP called those
> cartridges ROMs for a reason-- they were, and the PostScript
> interpreter ran on the controller board, which were all 68k-family
> processors. So what about this 8051? I don't have such a cartridge
> handy anymore to open it up, but if there was an 8051 in there, it was
> likely had some support role-- such as a controller for banking logic
> for the ROMs or othersuch.
>
> There is a consistency whenever Hugh offers one of his blustery
> claims-- when you keep pressing him on it, he'll either downgrade the
> claim or come up with a distraction to divert away from the claim.
"All wind and nae watter."
Alex McDonald
Hugh, I've never seen so much concentrated rubbish in one post on clf
before.
Elizabeth D Rather
Yes, folks clearly needed to be reminded of this. The Computer
Chronicles program that started this thread was really addressed at
commercial computer programmers and users, as was the polyFORTH product
I demo'd. polyFORTH included extensive database support, graphics, and
other features, and was priced for professional programming use. There
were some hobby-level Forths (FIGforth, LMI, Glen Haydon's system)
around at that time, and some dedicated hobbyists bought polyFORTH, but
polyFORTH was not primarily aimed at that market. For that matter,
Miller Microcomputers' MMSForth and Creative Solutions' MacForth were
also initially aimed at commercial/professional programmers.
By 1984 IBM PCs were extremely common in offices in America -- more in
small businesses than large corporations, which still used minicomputers
and mainframes with terminals.
Cheers,
Elizabeth
--
==================================================
Elizabeth D. Rather (US & Canada) 800-55-FORTH
FORTH Inc. +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
Alex McDonald
By 1984, IBM + PC clone mfrs (there were almost no clone sales at that
point) had sold 3.3million PCs. The AppleII had sold 1.9million by the
same date. http://jeremyreimer.com/totalshare0.gif, from
http://jeremyreimer.com/postman/node/329. (I believe these are IDC
numbers.)
Nice article on the growth of the PC at Ars Technica in 2005 by the
same author; http://arstechnica.com/old/content/2005/12/total-share.ars
Alex McDonald
> numbers.)
>
> Nice article on the growth of the PC at Ars Technica in 2005 by the
> same author;http://arstechnica.com/old/content/2005/12/total-share.ars
Cracking graph that refutes the "non-existent IBM PCs" assertion:
http://media.arstechnica.com/articles/culture/total-share.media/graph3-1.jpg
Marcel Hendrix
[..]
> Hugh, I've never seen so much concentrated rubbish in one post on clf
> before.
Some Googling will show that PerfectWriter was based on EMACS (an
editor), not on LaTeX (a document preparation system). What good
would LaTeX do if it couldn't handle equations and graphics?
The MikTeX version I have here takes 268 MBytes on disk and generates
over 100 Kbytes of intermediary files just to process my CV (PW needed
56K RAM and dual 180K floppies).
Apparently the OP is not familiar with EMACS, LaTeX, and Google.
-marcel
John Passaniti
> Some Googling will show that PerfectWriter was based on EMACS (an
> editor), not on LaTeX (a document preparation system). What good
> would LaTeX do if it couldn't handle equations and graphics?
Not to dispute your fine Googling, but while PerfectWriter was indeed
based on EMACS, that is just the editor. It's my understanding that
one would insert formatting directives in a manner like TeX and then
printing would be from a separate formatter. Again, my guess here is
that since the formatting language *looked* like TeX and maybe even
had some overlapping functionality, Hugh's claim that it *was* LaTeX
isn't supported.
ken...@cix.compulink.co.uk
Sp...@ControlQ.com () wrote:
> Please note that your accurate and prolific list of incompatible and
> underpowered toys was one of the reasons that commercial developers
> largely stayed away.
Obviously there are far too many professional here :-). Home computers
may have been underpowered but they were still better than the
alternatives for anybody who did not actually work with computers
professionally. Having spent half an hour just trying to log onto a
university time share system having my own computer however limited was
great.
The actual limits of the computers compared with what we can get now
are true but they were still fast enough not to have to wait for the
display to catch up with your typing which is more than I can say for
some teletype terminals. It is true that assembler was used for some
programs but a bigger limitation than language was storage. I can still
remember how much slower a database program got when it could no longer
hold the data in memory and that was using floppies not tape.
The TRS80 had 1K of video memory and an upper case only display. It did
have however a better and easier to use debugger than DDT. The ST had a
wide range of languages available including Object Forth from HiSoft.
Ken Young
ken...@cix.compulink.co.uk
<849587ab-011d-4f31...@a36g2000yqc.googlegroups.com>,
hughag...@yahoo.com (Hugh Aguilar) wrote:
> Those ROM-based BASIC interpreters didn't do any compilation.
That was why they were called interpreters Hugh. By the way no one has
claimed they did compile anything. They also supplied whatever OS there
was. They were an efficient solution for the limited memory space which
was hard to extend without pricing machines out of the market. In
addition it was possible for user programs to use a vector table to
replace some ROM commands.
Forth is an efficient user of memory at least judging by the Jupiter
Ace which had Forth in ROM instead of Basic and came with 3K of user
memory in the entry level machine.
I do not understand what seems like your personal vendetta against
Forth INC. They were never the sole source of Forth. In addition to
other commercial suppliers, I used HiSoft Object Forth on my ST there
was the Forth Interest Group which supplied FIG Forth listings for home
computers. As far as I can remember they all used ITC or DTC
implementations. Trying to fit a compiler and source into a CPM machine
was difficult and usually required the compiler to use overlays and most
8 bit home computers had less user RAM than a CPM machine. The TRS80
assembler was designed to run in 16K of memory and the output went
straight to tape.
Ken Young
Dennis
> On Aug 14, 11:39 am, Dennis <daruf...@gmail.com> wrote:
> > On Aug 12, 11:31 am, Hugh Aguilar <hughaguila...@yahoo.com> wrote:
>
> > > In 1984, maybe 99% of all commercial software was written in assembly
> What I said was that in 1984, 99% of commercial software was written
> in assembly language. WordPerfect didn't exist at that time.
Did you even both to scan through the wikipedia article on
WordPerfect?
"Originally written for Data General minicomputers, in 1982 the
developers ported the program to the IBM PC as WordPerfect 2.20,
continuing the version numbering of the Data General series."
> > It most certainly was not written in assembly.
>
> I already said that WordPerfect was the flagship of the new generation
> of software written in C rather than assembly language. In fact,
> *everything* that I said in this post, I had already said in previous
> posts. Now I'm just repeating myself. That is boring for me, so I can
> only imagine what effect it is having on everybody else.
>
> As for early versions of WordPerfect being written in Forth, I never
> heard of that. I thought it was always written in C. Can you
> substantiate this?
Other than having own a copy of SSI Forth and personally experiencing
the "feel" of both, no. I have not found any written statement to
that effect.
Do you have any proof that it was ever written in C? They needed an
OO system to handle Windows, and neither C nor Forth were up to that
task back in 1992. Even today, you don't do Windows in C.
DaR
Hugh Aguilar
> On Aug 14, 11:31 am, Hugh Aguilar <hughaguila...@yahoo.com> wrote:
> > What I said was that in 1984, 99% of commercial software was written
> > in assembly language. WordPerfect didn't exist at that time.
>
> Did you even both to scan through the wikipedia article on
> WordPerfect?
>
> "Originally written for Data General minicomputers, in 1982 the
> developers ported the program to the IBM PC as WordPerfect 2.20,
> continuing the version numbering of the Data General series."
Okay, maybe it *existed* --- but I sure don't remember it. In 1984,
WordStar was *the* word-processor and CP/M was *the* OS. It wasn't
until years later that the 16-bit MS-DOS computers became popular
(because IBM lost their legal hold on the BIOS and the price dropped)
and WordPerfect became popular (because it was much more feature rich
than WordStar).
> > As for early versions of WordPerfect being written in Forth, I never
> > heard of that. I thought it was always written in C. Can you
> > substantiate this?
>
> Other than having own a copy of SSI Forth and personally experiencing
> the "feel" of both, no. I have not found any written statement to
> that effect.
I wasn't calling you a liar; I just wanted substantiation. I've never
heard of SSI Forth. Is this Forth publicly available? If it is capable
of writing an application such as WordPerfect, maybe it is something
that people ought to know about. :-)
> Do you have any proof that it was ever written in C? They needed an
> OO system to handle Windows, and neither C nor Forth were up to that
> task back in 1992. Even today, you don't do Windows in C.
Except for the programmers who have access to the source-code, nobody
knows what it was written in --- and they certainly signed non-
disclosure statements. All I know is that, at the time, WordPerfect
was hailed as the triumph of high-level languages, and the decline of
WordStar was seen as the death of assembly-language.
Forth is an extensible language. Nobody actually writes an application
in Forth, in the same way that they write an application in C.
Applications are always written in a DSL built on top of Forth. To
some extent this is true of C also, but in C the extensibility is
accomplished with a preprocessor (such as in Objective C). That is
even true of C++ (such as in QT). For the most part, Forth is the only
language that was designed from the start to be extensible.
Hugh Aguilar
> Alex McDonald <b...@rivadpm.com> writes Re: Computer Chronicles episode
TeX/LaTeX isn't a monolithic program that you drop in. The source-code
is available, and it is public-domain. It is very possible to trim it
down substantially by deleting unneeded features. TeX/LaTeX is a Unix
program requiring extremely expensive equipment (in 1984), so it is
obvious that you can't just *drop-in* software like this on a CP/M
computer.
BTW, I have used MiKTeX on Windows (with WinEdt and, later, with
Crimson Editor) and didn't like it. Linux is far superior to Windows
as a platform for text-processing. I don't foresee Windows ever
becoming a viable system for writing books.
> What good
> would LaTeX do if it couldn't handle equations and graphics?
In 1984, people used dot-matrix printers. Typically you just sent your
file to the printer as a stream of ascii code and the printer printed
it out using a built-in font (Courier). What you got was mono-spaced
and rather blocky looking, and it was limited to the symbol set in the
built-in font. Dot-matrix printers were capable of printing out
graphical images though, in which case they were given a digital image
rather than a sequence of ascii codes. This, btw, is why Perfect
Writer was notoriously difficult to configure --- because the dot-
matrix printers varied considerably in how they printed graphics, even
if they were pretty much the same in regard to printing ascii text.
LaTeX was useful because it could generate a graphical image of a
page, complete with proportional-spacing and symbols beyond the ascii
set. For example, tables could be printed, and they would get smooth
lines rather than a lot of hyphens and vertical bars. This is why
Perfect Writer produced nicer-looking documents than WordStar.
Realistically, you aren't going to print out math equations on a dot-
matrix printer, because there isn't enough resolution. That is
something that would get deleted from a CP/M version of LaTeX.
Actually, quite a lot could be deleted. Just being able to print text
with proportional-spacing and extra-ascii symbols would be a major
step up from WordStar. That was the goal of Perfect Writer. It seems
trivial nowadays, but it was a big deal in 1984 --- technology was
pretty primitive back then.
At least part of the reason why Perfect Writer was a flop, was that
they were pushing the envelope with technology (the dot-matrix
printer) that was soon to be obsoleted (by the laser printer) anyway.
In 1984, most people assumed that CP/M and dot-matrix printers would
continue to be viable options into the 1990s; nobody really grasped
the implication of exponential growth in 1984. The same thing is
likely true of us today. Right now, the Windows strait-jacket is
restricting what we do with computers. Linux is really a strait-jacket
too. I like Linux and LaTeX, but all of this was available under Unix
in the 1970s, so it is not exactly cutting edge in 2010. It is less
expensive now, but it is just the same-old same-old. Similarly,
Wikipedia is the killer-app of the internet, but encyclopedias were
also available in the 1970s. The printed versions were more expensive,
and less up-to-date, but the basic idea hasn't changed in 40 years. If
we get our minds out of the 1970s, we might start to accomplish
something new and exciting with our computers. It could happen!
Richard Owlett
Hugh Aguilar
> I do not understand what seems like your personal vendetta against
> Forth INC.
Elizabeth Rather made a enemy of me in this thread:
http://groups.google.com/group/comp.lang.forth/browse_thread/thread/c37b473ec4da66f1
She has never apologized for attacking my symtab software and, until
she does, I'm going to continue to attack Forth Inc. software. It is a
case of how people who live in glass houses shouldn't throw stones.
Telling other people that their software "sucks" is a bad idea,
considering the kind of major problems that PolyFORTH and SwiftForth
have.
E.R. stuck her nose into an issue that wasn't any of her business. I
was getting attacked by John Passaniti, but I wasn't taking the attack
very seriously because it was just Passaniti "acting out" the way that
teenagers often do. Then E.R. felt obliged to get involved --- and now
I'm taking the attack seriously --- I'm doing everything in my power
to defeat Forth Inc..
On Aug 14, 3:07 pm, Elizabeth D Rather <erat...@forth.com> wrote:
> On 8/14/10 10:11 AM, Alex McDonald wrote:
> > This started as an assertion about commercial software. Not early
> > hobbyist or home PCs.
>
> Yes, folks clearly needed to be reminded of this. The Computer
> Chronicles program that started this thread was really addressed at
> commercial computer programmers and users, as was the polyFORTH product
> I demo'd.
Yeah, I'm sure a lot of commercial computer programmers were using
Logo. LOL And every professional programmer needs to be told that
there are three kinds of software --- low-level, middle-level and (you
guessed it!) high-level.
> polyFORTH included extensive database support, graphics, and
> other features, and was priced for professional programming use. There
> were some hobby-level Forths (FIGforth, LMI, Glen Haydon's system)
> around at that time, and some dedicated hobbyists bought polyFORTH, but
> polyFORTH was not primarily aimed at that market. For that matter,
> Miller Microcomputers' MMSForth and Creative Solutions' MacForth were
> also initially aimed at commercial/professional programmers.
I wouldn't call LMI a "hobby-level Forth." It allowed for reasonably
large programs to be written. It had the machine-code in one 64K
segment (pointed to by CS), the data and Forth-code in another 64K
segment (pointed to by DS), and the dictionary in another 64K segment
(pointed to by ES). Even still, my 65c02 cross-compiler pretty much
filled up the entire system. I had another two 64K segments
representing the bank-switched Apple II memory, so I didn't run out of
data space, but my dictionary was very close to being full. There is
no way that something like this could ever be written in PolyFORTH. As
I've said before --- you had the application code, the application
data, the system code, and the dictionary, all in *one* 64K segment.
Also, UR/Forth was direct-threaded, which is faster than indirect-
threaded (I did some benchmarks in UR/Forth and Turbo-C, and found
them to be very close in speed). I examined both 16-bit PolyFORTH and
16-bit LMI UR/Forth and it was obvious to me which was the
"professional" system that actual applications could be written in,
and which was the "hobby-level" system that only toy programs could be
written in. Even if I had had the money to buy PolyFORTH, I wouldn't
have bought it.
Direct-threaded is better than indirect-threaded, but I don't like
threaded code at all. With subroutine-threading, it is possible to
write macros as immediate words that generate machine-code inside of
colon words. This is what assembly-language is all about; writing
macros. With a threaded scheme however, assembly-language can only be
used for writing functions --- this is really no different from what
is done in Pascal --- it is not taking advantage of the fact that
Forth allows the programmer to write compile-time code in the form of
immediate words.
> By 1984 IBM PCs were extremely common in offices in America -- more in
> small businesses than large corporations, which still used minicomputers
> and mainframes with terminals.
Everybody used CP/M in 1984. The IBM-PC was way too expensive. It was
being sold by IBM, which was accustomed to selling mainframes, and the
price reflected this background. Also, there wasn't much software for
the IBM-PC at that time; IBM actually provided GW-BASIC, which was
even worse than PolyFORTH --- the IBM-PC had a lot of "potential" in
1984, but very little substance. It wasn't until clones started being
sold, and (especially) when they could be sold without paying IBM a
license fee for the use of the BIOS, that the MS-DOS market began to
take off.
Alex McDonald
> On Aug 15, 3:11 am, ken...@cix.compulink.co.uk wrote:
>
> > I do not understand what seems like your personal vendetta against
> > Forth INC.
>
> She has never apologized for attacking my symtab software and, until
> she does, I'm going to continue to attack Forth Inc. software. It is a
> case of how people who live in glass houses shouldn't throw stones.
> Telling other people that their software "sucks" is a bad idea,
> considering the kind of major problems that PolyFORTH and SwiftForth
> have.
>
> E.R. stuck her nose into an issue that wasn't any of her business. I
> was getting attacked by John Passaniti, but I wasn't taking the attack
> very seriously because it was just Passaniti "acting out" the way that
> teenagers often do. Then E.R. felt obliged to get involved --- and now
> I'm taking the attack seriously --- I'm doing everything in my power
> to defeat Forth Inc..
No-one here wishes you anything but outright failure on that score.
"Ertl's most recent tests[4] show that subroutine threading is faster
than direct threading in 15 out of 25 test cases. Ertl's most recent
tests show that direct threading is the fastest threading model on
Xeon, Opteron, and Athlon processors; indirect threading is the
fastest threading model on Pentium M processors; and subroutine
threading is the fastest threading model on Pentium 4, Pentium III,
and PPC processors."
http://www.complang.tuwien.ac.at/forth/threading/
It will be different for other processor architectures, and also note
that ITC can be much more compact that DTC or STC; a factor when
memory is tight.
Your characterisation of Forth as little more than an assembler macro
processor seems to me to be wide of the mark. One of the key features
is *portability*, and processor specific code is quite specifically
*not* portable.
>
> > By 1984 IBM PCs were extremely common in offices in America -- more in
> > small businesses than large corporations, which still used minicomputers
> > and mainframes with terminals.
>
> Everybody used CP/M in 1984. The IBM-PC was way too expensive. It was
> being sold by IBM, which was accustomed to selling mainframes, and the
> price reflected this background. Also, there wasn't much software for
> the IBM-PC at that time; IBM actually provided GW-BASIC, which was
> even worse than PolyFORTH --- the IBM-PC had a lot of "potential" in
> 1984, but very little substance. It wasn't until clones started being
> sold, and (especially) when they could be sold without paying IBM a
> license fee for the use of the BIOS, that the MS-DOS market began to
> take off.
Hugh, give it a break. Every analysis you undertake suffers from
Aguilar's Disease; that tendency to elevate personal reflection to
general observation. Your understanding of what was going on in 1984
is flawed. Much like your understanding of what's going on in 2010.
Alex McDonald
ron
Wow, Hugh -- you sure do spew.
> TeX/LaTeX is a Unix program requiring extremely expensive equipment
No it is not. It wasn't even originally written for Unix, it was
written for the PDP-10 running WAITS. Notwithstanding that, it has
been ported to hundreds of platforms, including Mac OS/X and Windows
(besides, of course, Linux).
> obvious that you can't just *drop-in* software like this on a CP/M
> computer.
Correct; TeX is a system of its own, intended to take input text and
output DVI (or PDF or whatever) for a typesetting device to handle. A
TeX system tends to have lots of parts (not all of which are necessary
for everyone at all times).
That said, it is pretty common (even today) to output TeX or LaTeX and
expect it to be fed into a TeX system, so the program asking for a
printed output doesn't really need to know how to accomplish that.
> I don't foresee Windows ever becoming a viable system for writing books.
Wow. Guess what, Hugh: most manuscripts written today (at least in my
experience) are sadly being written using MS Word. On Windows. Even
though it's a crappy application and platform.
> LaTeX was useful because it could generate a graphical image of a
> page, complete with proportional-spacing and symbols beyond the ascii
> set. For example, tables could be printed, and they would get smooth
> lines rather than a lot of hyphens and vertical bars. This is why
> Perfect Writer produced nicer-looking documents than WordStar.
Nothing to do with anything, Hugh. LaTeX (just a series of macros on
top of TeX) outputs "DVI" -- DeVice Independent output, which may then
be printed on *any* output device, from an LCD screen up to a $500000
commercial printing press. LaTeX does *not* produce a graphical image
of anything. It (and TeX) produces the DVI, which has instructions on
how to place "boxes" of information in specific locations. Read up on
the technology before talking about it, please.
>
> I like Linux and LaTeX, but all of this was available under Unix
> in the 1970s, so it is not exactly cutting edge in 2010. It is less
> expensive now, but it is just the same-old same-old.
If you say so. I'm currently using "xetex", a Unicode-enabled TeX, to
produce a bilingual book. I think it's pretty exciting that I can use
tried and true technology that's 30 years old in its genesis, to
produce cutting-edge documents today. But that's just me, I guess.
John Passaniti
> Elizabeth Rather made a enemy of me in this thread:
> She has never apologized for attacking my symtab software and, until
> she does, I'm going to continue to attack Forth Inc. software. It is a
> case of how people who live in glass houses shouldn't throw stones.
> Telling other people that their software "sucks" is a bad idea,
> considering the kind of major problems that PolyFORTH and SwiftForth
> have.
Elizabeth has no need to apologize. She did not "attack" your
software. She addressed your behavior. Anyone who bothers to read
the thread you reference can see that for themselves.
*I* was the one who supposedly "attacked" your software by showing
that it was an inferior algorithm for managing symbol tables. It uses
more memory than necessary, it uses more CPU than necessary, it
doesn't exploit locality of reference, and it's more complicated than
necessary. Competitive algorithms (splay trees, hashes, etc. )
*easily* beat it in terms of performance along any metric you want to
measure (speed, size, memory consumption, code complexity, etc.). So
yes, it does indeed suck for the purpose you intend. If it has
application elsewhere, I have no idea.
Of course, you could have proved me wrong by simply doing the
measurement I suggested-- take a real-world set of symbols, instrument
your code, and then publish the results. Imagine the feeling of
vindication you would have when you show to not just me, but to the
entire newsgroup that your "symtab" algorithm actually did beat the
competition, with cold, hard numbers to back up your claim. What's
the fear? If you really believe "symtab" is superior, then why not
prove it?
We're not stupid, Hugh. We all know what is *really* going on here.
I hurt your fragile ego by not just saying "symtab" sucked, but showed
why with the kind of basic analysis a first year software engineering
student would make. And more than just hurting your ego, I reminded
you of your own failure to make software a career. I personally see
nothing wrong with someone being a taxi driver, but you apparently do,
as you've mentioned several times in posts. So the issue isn't really
me saying "symtab" sucked, is it? The issue is a bit more meta, and
it's about you personally, and presumably the set of choices you made
that leads you to where you are now. In other words, it's all a
distraction, and it's painfully obvious to everyone.
Let's say Elizabeth really did attack your software. She didn't-- as
the link to the thread you helpfully provided shows-- but let's go
with your distraction mechanism for a moment and believe it. Wouldn't
the very best attack against Elizabeth and Forth, Inc. be to publish
the comparison measurements I suggested you make? Wouldn't the
shining truth of how awesome "symtab" is be an indisputable truth you
could hold up and use against her to show the world how brilliant you
really are?
> E.R. stuck her nose into an issue that wasn't any of her business. I
> was getting attacked by John Passaniti, but I wasn't taking the attack
> very seriously because it was just Passaniti "acting out" the way that
> teenagers often do. Then E.R. felt obliged to get involved --- and now
> I'm taking the attack seriously --- I'm doing everything in my power
> to defeat Forth Inc..
See, the thing about agitprop is that it only works when people don't
understand what's really going on. The problem is that you've slipped
up too many times, and we've all see your motivations are more about
protecting your fragile ego than any kind of vindication of your
knowledge or skill. These distraction mechanisms aren't working.
And as for "defeating Forth Inc.", I really don't see how that is
possible. You see Hugh, you're a sad clown. With each passing
message, you creep closer to being considered something like an Arthur
T. Murray-- a newsgroup crackpot. Whatever worthwhile technical
points you may occasionally offer are drowned out in a sea of
projected self-hatred, bigotry, and paranoid delusion. Or put another
way, the only threat you offer to Elizabeth and Forth, Inc. are
convincing people who already have a problem with her or her company.
The rest of the newsgroup has been here for *years* prior to you
coming in here, and while some of us (including myself) disagree with
her on various points, pretty much everyone respects her wealth of
insights and practical experience with Forth.
So, good luck, Hugh. The battle lines you've drawn are ridiculous
(you should be attacking me, not her), your motivations are obvious (a
distraction mechanism to keep from having to address the various
inefficiencies and flawed design of "symtab"), and your behavior
suggests the answer to why you are where you are.
John Passaniti
> > I'm doing everything in my power
> > to defeat Forth Inc..
>
> No-one here wishes you anything but outright failure on that score.
That's a really weird response, Alex. It suggests that Hugh actually
has built enough of a reputation through his work and words that he
actually had the ability to "defeat" Forth, Inc. Or, it suggests that
the larger lie about Elizabeth somehow transitively insulting his
software by saying his bigotry wasn't appreciated is even remotely
credible. Yeah, we all know that in this day and age, if you repeat a
lie often enough it can gain traction, but pretty much every time Hugh
makes his goofy claim, he himself is disproving it with a link to the
thread in question! But beyond that, the worst possible outcome of
Hugh's "doing everything in [his] power to defeat Forth, Inc." is to
convince an individual from buying a copy of SwiftForth or a company
from using their consulting services. And do you honestly believe
that anyone that Hugh could convince to do so wouldn't already be
predisposed to not do business with Forth, Inc.?
There is no need to wish Hugh failure. That's redundant.
w_a_x_man
> by saying his bigotry wasn't appreciated
Ambrose Bierce:
bigot, n. One who is obstinately and zealously attached to an opinion
that you do not entertain.
Alex McDonald
Add rolling eyes and read it as sarcasm; that's something that's
difficult to convey in print. If I'd said it to you in person, you'd
have understood immediately.
As an aside, I was beginning to wonder why I was even bothering to
reply to Hugh on any of his confections, and it struck me last night.
I now earn my living by doing competitive analysis and intelligence
work. The claims that Hugh makes offends my sense of professionalism;
they're sloppy, unsupported by evidence, hearsay, guesswork. But the
nature of Hugh's style is to appear knowledgeable, when it's no more
than a smattering of half truths (at best!) dressed up with a flawed
analysis; all delivered on tablets of stone.
I'm thinking of using some of this for internal training purposes as a
case study in why you should never ever accept a single data point
when doing research, and that it's all too easy to be deceived if it's
your sole source of information.
Aguilar's Disease; presenting as fact an opinion based on personal
experience, limited observation and no research. Diagnosis; simple
fact checking usually reveals the underlying affliction. Prognosis;
chronic and episodic in nature, with no known cure. Treatment;
eliminate from your source of intelligence.
You heard it here first.
Travis Bickle
> Elizabeth Rather made a enemy of me in this thread:
> http://groups.google.com/group/comp.lang.forth/browse_thread/
thread/c37b473ec4da66f1
ER> I am *personally* offended by your homophobic bias and use of
ER> personal insults. I do not choose to converse with people of your
ER> intolerant mindset.
I'm standing here; you make the move. You make the move. It's your
move...
ER> Ok, Hugh, that's the last straw. You're in my "kill file" now.
Don't try it you f#@!.
ER> I have no interest in any further postings from you on any subject.
ER> John has made far more technical contributions to
ER> comp.lang.forth than you have.
You talkin' to me? You talkin' to me? You talkin' to me? Then who
the hell else are you talking... you talking to me? Well I'm the
only one here. Who the f#@! do you think you're talking to? Oh
yeah? OK.
Alex McDonald
Hugh, is that you?
Travis Bickle
> *I* was the one who supposedly "attacked" your software by showing that
> it was an inferior algorithm for managing symbol tables. It uses more
> memory than necessary, it uses more CPU than necessary, it doesn't
> exploit locality of reference, and it's more complicated than necessary.
> Competitive algorithms (splay trees, hashes, etc. ) *easily* beat it in
> terms of performance along any metric you want to measure (speed, size,
> memory consumption, code complexity, etc.). So yes, it does indeed suck
> for the purpose you intend. If it has application elsewhere, I have no
> idea.
I got some bad ideas in my head.
> We're not stupid, Hugh. We all know what is *really* going on here. I
> hurt your fragile ego by not just saying "symtab" sucked, but showed why
> with the kind of basic analysis a first year software engineering
> student would make. And more than just hurting your ego, I reminded you
> of your own failure to make software a career. I personally see nothing
> wrong with someone being a taxi driver, but you apparently do, as you've
> mentioned several times in posts. So the issue isn't really me saying
> "symtab" sucked, is it? The issue is a bit more meta, and it's about
> you personally, and presumably the set of choices you made that leads
> you to where you are now.
Now I see this clearly. My whole life is pointed in one direction. There
never has been a choice for me.
> So, good luck, Hugh. The battle lines you've drawn are ridiculous (you
> should be attacking me, not her), your motivations are obvious (a
> distraction mechanism to keep from having to address the various
> inefficiencies and flawed design of "symtab"), and your behavior
> suggests the answer to why you are where you are.
Listen, you f#@!ers, you screwheads. Here is a man who would not take it
anymore. A man who stood up against the scum, the c#nts, the dogs, the
filth, the sh!t. Here is a man who stood up.
John Passaniti
> Hugh, is that you?
Doesn't seem like it, from the message headers. Just someone who
apparently likes Robert De Niro films.
Hugh Aguilar
No --- I never use sock puppets. I don't care enough about posting
messages on public forums to do that.
I'm essentially a computer programmer:
http://www.forth.org/novice.html
I did mention earlier in this thread that I upgraded my novice
package, although nobody seemed interested.
Hugh Aguilar
> other features, and was priced for professional programming use.
Ultimately, it is all about money. Forth Inc. has a long history of
following the old salesman's maxim:
"Sell the sizzle, not the steak." Forth Inc. owns the name "Forth" and
they are basically selling that name and the image that goes along
with it. E.R. does a lot of bragging about that PDP-11 project in
Saudi Arabia. Chuck Moore was still with the company at that time
though, and it was him who wrote the software. E.R. has basically been
selling a life-size cardboard cut-out of Chuck Moore ever since he
left the company --- there is nothing behind the facade.
My own motivation in spending $500 on SwiftForth was to essentially
buy my way into professionalism. That was a big mistake. SwiftForth
isn't any good; it is really a poor choice compared to C++ or any of
the other professional systems available for Windows. Having
experience in programming SwiftForth hasn't helped me get a job. Forth
Inc.'s reputation in the real world is so bad that claiming SwiftForth
knowledge is a mark against me rather than for me.
> There
> were some hobby-level Forths (FIGforth, LMI, Glen Haydon's system)
> around at that time
I think that E.R. was enthusiastic about convincing the world that my
novice package "sucks" and/or that I'm intolerant and homophobic,
because she felt threatened by the novice package. She also felt
threatened by LMI, etc., in the list above, which is why she
enthusiastically denounces them as "hobby-level." Elizabeth Rather's
strategy is not to upgrade SwiftForth into a useful product, but
rather to drag down everybody in the Forth community far enough that
SwiftForth looks good by comparison. I doubt that she even looked at
my symtab or anything else that I've written --- from her perspective,
it was going to "suck" no matter what. I think that it is highly
likely that Passaniti is on Elizabeth Rather's payroll, and that his
job is to instigate strife in the Forth community such that all the
Forthers are throwing vulgar insults at each other. Passaniti is like
a monkey in a tree that throws poo at every person who walks past;
even when he hits his mark, nobody calls him wise (can anybody recall
him *ever* saying anything positive about anybody?). E.R. doesn't want
to dirty her own hands doing this, so she hired a surrogate to do the
job for her. I know it sounds rather Machiavellian, but that kind of
psy-op does happen. The CIA does this in foreign countries all of the
time; it is called: "divide and conquer." Now we see Alex McDonald
attacking me in a similar manner to Passaniti (by similar, I mean that
he uses scatalogical terminology). He is just jumping on the Devil's
Bandwagon; he is doing what he believes will impress Elizabeth Rather
and possibly eventually get him on the Forth Inc. payroll.
Elizabeth Rather has actually hated me since before she even knew me.
When I discovered that SwiftForth did not run under Windows Vista (it
had run under Windows XP), the Forth Inc. CSRs (customer-service reps)
told me that I would have to pay an extra $300 in order to get a
working version. The $500 that I had already spent didn't carry any
weight. Elizabeth Rather enthusiastically supported this decision. I
got the impression from her that the CSRs had already consulted her on
the matter and that it was her personal decision that I should be
charged the extra $300. This was long before I had written the novice
package and been duly attacked by Passaniti for doing so. E.R. had no
previous contact with me at all. I think she wanted to prevent me from
getting a working copy of SwiftForth because she was afraid that I
might use it to write Forth software (such as the novice package) that
would make Forth Inc. look bad by comparison --- so she put the
squeeze on me for an extra $300. I continued complaining however, and
eventually I got through to Leon Wagner who went ahead and gave me the
TTY.DLL file that I needed to make SwiftForth run under Windows Vista.
For free! I have a lot of respect for Leon Wagner because of this. He
believes in supporting his customers. I have worked as a CSR myself
(when I was an IBM370 assembly-language programmer I also did my own
CSR work some of the time), so I can relate to him. I was really
turned off by Elizabeth Rather though. When a CSR tries to squeeze a
customer for extra money, it usually means that the company is on the
ropes financially; they have no long-term plan; they are just trying
to maintain a positive cash-flow week-by-week. When I was a CSR I
would never have tried to squeeze a customer like that --- certainly
not for such a trivial upgrade. I often wrote software for free for
the customers just to keep them happy; I only charged them extra if
they wanted a big enough upgrade that I would have to actually
schedule time to write the code and/or to run it on the mainframe
(roughly speaking, anything that takes more than two hours of
anybody's time).
Also --- when I said that I was going to defeat Forth Inc., what I
meant was that I intend to upgrade my old MFX Forth compiler to a more
mainstream and modern processor and put it in the public domain. That
should do it!
Alex McDonald
[stuff snipped]
> Now we see Alex McDonald
> attacking me in a similar manner to Passaniti (by similar, I mean that
> he uses scatalogical terminology).
I do?
> He is just jumping on the Devil's
> Bandwagon; he is doing what he believes will impress Elizabeth Rather
> and possibly eventually get him on the Forth Inc. payroll.
>
I cash a large Forth Inc. cheque every week for bandwagonning.
Honest.
[snipped again]
> (when I was an IBM370 assembly-language programmer
Really? Somehow, I don't quite believe it.
[and snipped again...]
>
> Also --- when I said that I was going to defeat Forth Inc., what I
> meant was that I intend to upgrade my old MFX Forth compiler to a more
> mainstream and modern processor and put it in the public domain. That
> should do it!
Damn. No job at Forth Inc. for me then.
Elizabeth D Rather
Nice try, but I don't think he has a sense of humor.
Cheers,
Elizabeth
--
==================================================
Elizabeth D. Rather (US & Canada) 800-55-FORTH
FORTH Inc. +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
http://www.forth.com
"Forth-based products and Services for real-time
applications since 1973."
==================================================
ron
...
Hugh, you forgot to mention that though Passaniti shot Kennedy while
McDonald was the lookout, it was really Rather who planned the devlish
plot. Alas, I was busy putting poison in the water supply.
Albert van der Horst
ron <r...@ronware.org> wrote:
>On Aug 16, 11:12 pm, Hugh Aguilar <hughaguila...@yahoo.com> wrote:
>
>Wow, Hugh -- you sure do spew.
>
>> TeX/LaTeX is a Unix program requiring extremely expensive equipment
>
>No it is not. It wasn't even originally written for Unix, it was
>written for the PDP-10 running WAITS. Notwithstanding that, it has
>been ported to hundreds of platforms, including Mac OS/X and Windows
>(besides, of course, Linux).
>
>> obvious that you can't just *drop-in* software like this on a CP/M
>> computer.
>
>Correct; TeX is a system of its own, intended to take input text and
>output DVI (or PDF or whatever) for a typesetting device to handle. A
>TeX system tends to have lots of parts (not all of which are necessary
>for everyone at all times).
>
>That said, it is pretty common (even today) to output TeX or LaTeX and
>expect it to be fed into a TeX system, so the program asking for a
>printed output doesn't really need to know how to accomplish that.
>
>> I don't foresee Windows ever becoming a viable system for writing books.
>
>Wow. Guess what, Hugh: most manuscripts written today (at least in my
>experience) are sadly being written using MS Word. On Windows. Even
>though it's a crappy application and platform.
This is because writers are not type setters.
Word is todays equivalent of a 1930 typewriter.
In the context of TeX a 1930 typewriter is not a "viable
system for writing books". It is not a system for writing books
at all. It is a means to capture creative writing.
For most writers notepad would do as well.
Henry Millers master pieces would have about the same impact
in the form of a bunch of photocopies of typewriter copy.
But the encyclopedic touch and the elaborate indices of the
1000+ Coherent manual is a different story.
The last time I saw a book that was typeset using windows (and
proudly announces so) was Petzold of Windows 3.1 fame.
The Coherent manual (1990 Unix clone on 286) was done
totally in emacs plus troff running on Coherent.
Since then I've bought quite some books, typeset with
all kinds of systems (among them, prominently, TeX)
none of them Word.
Groetjes Albert
--
--
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- being exponential -- ultimately falters.
albert@spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst
ron
wrote:
> Since then I've bought quite some books, typeset with
> all kinds of systems (among them, prominently, TeX)
> none of them Word.
I know several authors (of non-technical books), each of them writes
the "manuscript" in Word. I *almost* convinced one of them to use TeX
after I showed him the very visible difference in quality, but alas!
no such luck.
yao pygmy user
> [...] I like Linux and LaTeX, but all of this
> was available under Unix in the 1970s, so it is not exactly cutting
> edge in 2010. It is less expensive now, but it is just the same-old
> same-old.
don't give up "iTex[ringeling] clearly is the new thing"(tm).
http://river-valley.tv/an-earthshaking-announcement/
presented from Don Knuth on TUG 2010.
John Passaniti
> I know several authors (of non-technical books), each of
> them writes the "manuscript" in Word. I *almost* convinced
> one of them to use TeX after I showed him the very visible
> difference in quality, but alas! no such luck.
Wouldn't these writers be submitting their work to publishers who
would apply whatever in-house typesetting tools they were comfortable
with? I can understand them being concerned about the quality of
output if they were publishing directly to the actual printers, but if
they aren't, what does the quality of output matter? So letter pairs
won't be kerned as nicely and those who keep typography books on their
coffee tables will sneer. So, who cares?
And if it did matter-- if they're generating printer-ready copy-- why
would they want to write using TeX when there is desktop publishing
software that gives the same quality visually? I know why *I* would
use a formatting language like (La)TeX or DocBook or even XHTML. But
I'm not an investigative journalist, historian, poet, or biographer.
Embedded formatting languages fit perfectly within the mindset of a
software developer. But most non-technical writers I know (and my
partner happens to be one) find embedded formatting languages
distracting and want a visual editor.
Elizabeth D Rather
The last several books I've written have been done in Adobe Framemaker,
but the current favorite among professionals in the US is now Adobe
InDesign, which I now have and use. Word is fine for *composing* text,
though, especially for shorter works. It's also ideal when a piece has
to be circulated in a team for review, because everybody has it and is
comfortable with it.
Alex McDonald
> On Aug 18, 10:01 am, ron <r...@ronware.org> wrote:
>
> > I know several authors (of non-technical books), each of
> > them writes the "manuscript" in Word. I *almost* convinced
> > one of them to use TeX after I showed him the very visible
> > difference in quality, but alas! no such luck.
>
> Wouldn't these writers be submitting their work to publishers who
> would apply whatever in-house typesetting tools they were comfortable
> with? I can understand them being concerned about the quality of
> output if they were publishing directly to the actual printers, but if
> they aren't, what does the quality of output matter? So letter pairs
> won't be kerned as nicely and those who keep typography books on their
> coffee tables will sneer. So, who cares?
I do.
As an (ex) user of Quark on a Mac, I can vouch for the quality of
typesetting affecting the perception of the quality of the product. A
recent paperback I purchased has fairly obviously been imaged from
laser output with a poor selection of fonts. Too many cheap fonts, and
the wrong type; the kerning was all over the place because it had been
photo-reduced rather than kerned at the right point size, no ligatures
(especially important for fi) and most annoying, the lower case italic
'a' was simply a sloped version of the upright font. Horrible.
Good quality typesetting and good quality fonts cost money for a
reason; they're far superior, even to the untrained eye.