Borland Backgrounder

553 views
Skip to first unread message

John Topley

unread,
Feb 6, 1999, 3:00:00 AM2/6/99
to
Could someone please bring me up to speed with Borland history, in
particular where things started to go wrong. I know that in the mid to late
eighties, Borland were hurting Microsoft ("Kill Philippe!") with products
like Turbo Pascal and Quattro Pro. Is it true that Delphi was a make or
break product? What's the background on this Del character? All I know is
that as well as running the show, he's grossly overpaid and has a fondness
for pizzas. What happened to Philippe Kahn anyway? And just who was Frank
Borland? Sorry for my ignorance.

Thanks,

--
John

(JTSoftware - http://ds.dial.pipex.com/john.topley)

Philippe Ranger

unread,
Feb 6, 1999, 3:00:00 AM2/6/99
to
John: >>

Could someone please bring me up to speed with Borland history, in
particular where things started to go wrong. I know that in the mid to late
eighties, Borland were hurting Microsoft ("Kill Philippe!") with products
like Turbo Pascal and Quattro Pro. Is it true that Delphi was a make or
break product? What's the background on this Del character? All I know is
that as well as running the show, he's grossly overpaid and has a fondness
for pizzas. What happened to Philippe Kahn anyway? And just who was Frank
Borland? Sorry for my ignorance.
<<

I have never worked at or for Borland, I have no inside story here, and I'm
working from memory. So —

I don't know when "Frank" Borland popped up, but the rumor I tend to
subscribe to is that Borland was taken as an American-sounding name for a
company, because of Frank Bormann, the astronaut.

Borland came to attention in late 1983, with an ad in Byte for Turbo Pascal,
at $50 by mail. That product was so innovative in so many ways (marketing
being not the least of them), someone should do a book. Anyhow, TP took "the
world" by storm. Version 2 came out in June 84 and version 3 in April 85.

At about that time, Borland, now an established company, started coming out
with a slew of utility-type products. One of the very first, Sidekick (not
like the current thing) was again a major innovation that marked the
industry. Especially, Sidekick's user interface slowly became the standard
for all PC apps until Windows 3 came along 5 years later.

Now, from all appearances, this stuff was bought, and seldom did the authors
stick around for recasts. Borland was unable to follow up on any of its
utilities, even SK, which remained stuck at version 1.53. Likewise, TP's
very innovative "IDE" used an editor built on a binary-code machine Borland
did not own the code for. That same machine was offered separately in an
"Editor Toolbox" for TP3, and, paf! editors sprouted by the dozen
(literally). Almost ALL Dos-based editors were born from this, unless they
were ported from VMS, Unix, etc. But Borland could not improve in any way on
the original binary editor (bined).

Likewise, in the summer of 86, if memory serves, a summer intern was hired
to "port" to TP to Basic, and Turbo Basic was born and launched. Again, it
could not follow on itself, and eventually ended up resold to its author. I
believe the first Turbo C was also a renamed third-party product. But in
this case Borland did manage to update it and, through several versions,
make it "theirs". Borland C++ v. 1 appeared in summer 89 (if memory serves),
and it was truly Borland. Also, again, earth-shaking.

Though there was a long lull before TP4 (last days of 87), the fact is that
at least TP was a purely-Borland enterprise. The author was Anders
Hejlsberg. Anders more or less moved to other things, I understand, while
Delphi 2 was a-building, and soon after was hired by Microsoft by paving his
path with gold. At the same time (again, if memory serves) the Delphi
business manager was similarly, er, seduced. But I'm getting ahead of
myself.

Anyhow, Anders was little-known except among TP afficionados (the kind of
people who hang around here), and Borland was identified with its president,
Philippe Kahn, a Frenchman. The more Borland grew, the more Kahn became a
public figure, and by the late 80's he was sounding a bit like the Divine
Ellison. As long as Borland found NEW products to launch, in the brouhaha it
went rather unnoticed that it couldn't follow-up on any of its past
successes except TP and TC.

One exception. Rather early, perhaps 85, Borland bought a very promising
database **maker**, Paradox, and simply left it to its own devices. But the
product itself was never more than an also-ran on the market, compared to
the remarkably inferior dBase, which willy-nilly made itself the standard
from 85 onwards.

So Borland bought the oldest WP app on Earth, which it relaunched as Sprint.
It bought one spreadsheet maker (everyone was making better spreadsheet than
Lotus, but Lotus was totally the standard), and let it disappear from view.
It bought another one, Quattro (Spanish for "beyond 1-2-3), got to work on
it (and kept at it), and quickly got a massive suit from friendly Lotus.
That really hurt prospects. A bit later, it bought the best-known
programmer's editor, Brief, and let that disappear too. Btw, at the time
Excel was a perpetual money-loser for MS and Quattro wasn't really a threat.
The war was with Lotus.

And in the summer of 91 it bought not dBase, but the whole of Ashton-Tate
(twice the employees but not twice the revenues). At a price I don't
remember but found unbelievably inflated (this was pre-Web times, people
actually figured sofware companies the way you'd figure tire companies).

Then things really, really went wrong. There was no way Borland could digest
A-T fast enough to get all the nourishment, and it really needed it for the
price it paid. Also, by that time every one's major concern was, not just
getting out a Windows version, but getting out a superior Windows version.
Everything had to be redone. If you notice, WordPerfect Corp., which totally
dominated its (larger) market and bought nothing also lost it at that point.

Concerning dBase, there was a rumor Borland had a dBase for Win in its
files, which it could not launch because of the legal threat from A-T. It
seems it wasn't true, dBase for Win was quite late. And Paradox was late
too. Quattro held on by the skin of its teeth, but in the meanwhile MS
knocked the teeth out of the true enemy, 1-2-3, which was also using the
Kahn policy of buying stuff right, left and center. Lotus wound up sold to
IBM mostly for one of those purchases, Notes. I cite WPC and Lotus to show
that, with very different management styles, other MS competitors also went
down the drain during the early nineties.

In 92 BP7 (Borland Pascal 7) came out, and it was the second version for
Windows. I may think the world of it, but the world thought otherwise, and
it was a marketing dud. I truly don't understand why, but the common
explanation is that by that time VB was gaining a lot of attention. Anyhow,
at the time Borland desperately needed at least to hold on to its old
revenue sources.

Note. Del has clearly operated an developer/enterprise switch. "Developer"
is practically synonymous with the old (eighties) market for TP. The people
who really saw what the enterprise end was, in marketing, were people at MS.
They brought in the notion of the Suite, early 91, and this is really how
they killed off both giants so quickly, Lotus and WPC. Well, enterprise
appeal was what was missing from BP7 marketing, and what quickly grew
regarding VB.

Anyhow, at that time, 91-94, Borland still had by far the best C++ compiler
for Windows, and that part of the business did well. But MS could just pour
money into compiler improvements (starting at least three full years late),
and finally managed to make its VC++ the standard, before Win95 showed up. I
say this through gnashed teeth, the Borland solutions at the time were
almost always superior.

Well, besides BC++, Borland had mostly trouble. Truly, the remarkable thing
is that it did NOT go down the drain, the way WPC and Lotus did. But both
its databases were disappearing from view, all other A-T products had been
dropped, Quattro was a poor second to Excel, and BP was... well was having
no new version.

There was a rumor about a "VBK" (VB Killer) in the works as the successor,
but clearly Borland did not have the money, if not to develop it, to market
it. In any case, Delphi 1 came out in early 95, at a point where any notion
of a VBK was ridiculous. It did meet with remarkable success, considering
conditions. Stock watchers, who had nothing else to hang on, said it was
"make or break". This was obviously true from the balance sheet. If money
didn't come in from Delphi, pretty soon there wouldn't be enough to print a
balance sheet on. I do think BCB and Jbuilder were financed on the returns
(and prospects) of Delphi. But I may be wrong. Anyway, Delphi gave Borland a
new life in (rather limited) public notice, for a time.

Among the expenses Borland could do without was the salary paid Philippe,
especially as his style wasn't what was needed by that time. So he went out
the door (can't remember exactly when, 94 probably), taking with him a nice
severance package and one product, the new Sidekick. On this he built
Starfish software, currently noted mostly for the Rex credit-card-sized PDA,
for which Starfish has rights on the sw.

Replacing him, at much lower salary, was not easy. Especially as Delphi
didn't keep climbing on the charts. The CEO's office had something of a very
unthankful revolving door, and by summer 96 was occupied by a member of the
board, because someone had to be there. A search for a "real" CEO was
seriously on, and seriously quite tough. In the end (Nov. 97, I guess) the
job went to an ex-Apple honcho who had left during one of miracle-worker
Amelio's cleanups, Delbert Yocam. Someone caved in, there, because Del is
more expensive than Philippe ever was.

Since this post is more than long enough, and has stuck to trying to be
factual, I'll leave off discussing the next two years.

PhR

Joe C. Hecht

unread,
Feb 6, 1999, 3:00:00 AM2/6/99
to
Wow Philippe! That was a pretty darn good write up!!!!


Mind if I add a couple of things???

I heard that Phillipe started Borland because no
one would hire him.

Borland rose to be the second (or debated third)
largest software institution.

Phillipe was always rooting for the little guy.
He never forgot what it was like to be a small
and struggeling kitchentop programmer, and commonly
loved to take swings at the knee's of the big boys.

Phillipe revelotionized software's legal side with
the famous "Borland No Nonsense License Agreement".

As I recall, at one time, Borlands legal department
produced a series of legal documents for the small
programming guy, and offered them via fax and BBS.

I believe the Borland forum on Compuserve was
(for awile) the largest forum base that Compuserve
had.

Turbo Pascal was actually a product called PolyPascal,
produced by a kid Phillipe had gone to school with
(Anders) under Professor Niklaus Wirth (the father
of Pascal)

Phillipe plays the Sax, and loves Turbo Porche 911's.

StarFish has been pretty darn successful for a small
quiet private company that resides in Borland's old
offices.

For good or bad, Phillipe got into a pricing war
with Microsoft(?). I think it was a Quattro vs Excel
war, dropping the price of the software down to $49
(if I recall correctly). This forever changed the value
of business software in the open market. (remember
when PeachTree was $5000???)....

The Borland campus (Borlands only tangable assett)
is valued at 100 million, and is a major thorn for
the company. It seems no one wants the building.
This is probably one of the biggest reasons Borland
has never been bought out <g>.


Borland has since sold or licensed out most of the
non-hard-core developer and end user products, mostly
to the Corel folk...

For awhile, Delphi's massive revinues have helped
to float the company and development of of other
products. Perhaps it still does ;)


Joe
--
Joe C. Hecht
http://home1.gte.net/joehecht/index.htm

Dave11

unread,
Feb 6, 1999, 3:00:00 AM2/6/99
to
Phillippe,

The one addition I'd make is that PdoxWin V1 came out shortly after Access
V1 (there was no credible windows DBMS at the time ...I don't count
SuperBase) and Borland chose to start a price war by slashing 75% or so off
the usual per seat license fee.

Unfortunately Access was the superior product (at the time) and MSFT had the
money to fight the war so the plan simply served to blow a big hole in
Borland cashflow.

Eryk

Mark Richter

unread,
Feb 6, 1999, 3:00:00 AM2/6/99
to
Philippe Ranger wrote:
>
> Likewise, in the summer of 86, if memory serves, a summer intern was hired
> to "port" to TP to Basic, and Turbo Basic was born and launched. Again, it
> could not follow on itself, and eventually ended up resold to its author.

I seem to recall at some gathering that a (then) Borland representative
indicated that Turbo Basic came under fire due to some problems with
licensing or other issues, particularly regarding an integrated
debugger. Maybe these issues were close to the cause of Turbo Basic
never being pursued any further. I found it to be a remarkable product
with tremendous speed and capabilities.

> ... compared to


> the remarkably inferior dBase, which willy-nilly made itself the standard
> from 85 onwards.

I believe dBase was a port over from the CP/M world onto the "new"
PC-DOS. dBase II (at the time) was pretty much the only option, and was
widely popular with those of CP/M heritage.

> So Borland bought the oldest WP app on Earth, which it relaunched as Sprint.

So that is where it came from. I remember having a test copy (still
do). Was pretty limited.

> It bought one spreadsheet maker (everyone was making better spreadsheet than
> Lotus, but Lotus was totally the standard), and let it disappear from view.

Again, Lotus was a migration from the CP/M VisiCalc (I believe that was
the name). Kapor had apparently written the earlier product, and owned
the rights for the port.

> It bought another one, Quattro (Spanish for "beyond 1-2-3), got to work on
> it (and kept at it), and quickly got a massive suit from friendly Lotus.

The suit was over "look and feel", if I recall correctly. And today,
Macintosh and Windows promote "look and feel" consistency across
products. Interesting how the markets change.

> Btw, at the time
> Excel was a perpetual money-loser for MS and Quattro wasn't really a threat.
> The war was with Lotus.

Wasn't MS offering at the time a PC-DOS port of MultiPlan, or some such
product? I thought Excel came in with the first Windows offering...



> In 92 BP7 (Borland Pascal 7) came out, and it was the second version for
> Windows. I may think the world of it, but the world thought otherwise, and
> it was a marketing dud. I truly don't understand why, but the common
> explanation is that by that time VB was gaining a lot of attention.

I suspect VB played a significant role, but I seem to remember that
Windows programming was exceedingly complex and little understood in the
early days. There were only programming examples in C, none in Pascal.
And as I remember trying to learn TPW 1.0, it was a real nightmare.

> Anyhow, at that time, 91-94, Borland still had by far the best C++ compiler
> for Windows, and that part of the business did well.

I seem to remember the early Borland C++ compiler being capable of DOS,
Windows, and OS/2 compilations.

Thanks, Philippe, for a truly remarkable post.

I thought I remembered bits and pieces, but obviously nowhere near the
amount you do.

--
Mark Richter
eMCee Software

Glynn Owen

unread,
Feb 6, 1999, 3:00:00 AM2/6/99
to
... and you never mentioned Turbo Prolog. I think it lasted almost 1
whole year somewhere around 1988.
Glynn

Roger Arnesen

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
>In 92 BP7 (Borland Pascal 7) came out, and it was the second version for
>Windows.

Ehhh.. .. 3. version for Windows. I personally used TPW 1.0 and 1.5

>Since this post is more than long enough, and has stuck to trying to be
>factual, I'll leave off discussing the next two years.


Other than my above comment, I am amazed with the accuracy. I've been there
since TP 3, and have seen "it all"

Roger

Barry Mossman

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Philippe,

great post. Thanks. Looking forward to volume 2.

<<Since this post ......, and has stuck to trying to be factual, I'll leave
off discussing the next two years.>> Go on, give it a go. In this case the
truth is stranger than fiction, so the factual should suffice to keep your
readership tapping the scroll bar.

Barry Mossman

Dmitry Streblechenko

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
>I suspect VB played a significant role, but I seem to remember that
>Windows programming was exceedingly complex and little understood in the
>early days. There were only programming examples in C, none in Pascal.
>And as I remember trying to learn TPW 1.0, it was a real nightmare.


Yeah, I remember my first Windows program - struggled for 2 weeks to make a
button _not_ to fill up the whole window. Sounds stupid nowdays, but that's
just the way it was...

Dmitry

Richard Bayarri Bartual

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Mark Richter wrote:
>
> I seem to recall at some gathering that a (then) Borland representative
> indicated that Turbo Basic came under fire due to some problems with
> licensing or other issues, particularly regarding an integrated
> debugger. Maybe these issues were close to the cause of Turbo Basic
> never being pursued any further. I found it to be a remarkable product
> with tremendous speed and capabilities.
>

I liked Turbo-Prolog too - it even had relational database capabilities.
While criticised by Prolog purists for its inability to assert rules as
well as data, it was a true compiler that produced tight executables and
had a superb development environment.

> I believe dBase was a port over from the CP/M world onto the "new"
> PC-DOS. dBase II (at the time) was pretty much the only option, and was
> widely popular with those of CP/M heritage.
>

You are correct. There never was a dBase 1 - the "II" bit was pure
marketing.
It wasn't the only CP/M database though - Condor was another notable
(which
I don't think ever made it to the PC). The three most popular CP/M
applications
were WordStar, dBase, and SuperCalc (by Computer Associates) - MS
MultiPlan
never really caught on despite being a quite capable product.

>
> > Btw, at the time
> > Excel was a perpetual money-loser for MS and Quattro wasn't really a threat.
> > The war was with Lotus.
>

> Wasn't MS offering at the time a PC-DOS port of MultiPlan, or some such
> product? I thought Excel came in with the first Windows offering...
>

Excel was originally written for the Mackintosh, a machine that Bill
Gates
liked because he felt that GUIs were the way to go (he literally begged
Apple to license Mac ROMs to clone makers in order to establish it as a
standard; they didn't listen - if they had, then history might have been
rather different). In a strange repeat of the Apple-II/VisisCalc thing,
Excel was the application which established the Mackintosh as a business
machine because using a mouse to select and drag spreadsheet columns
was so much more natural than doing the same with a keyboard, and few
potential spreadsheet users were capable enough typists to dislike
switching between keyboard and mouse for some operations.

>
> I suspect VB played a significant role, but I seem to remember that
> Windows programming was exceedingly complex and little understood in the
> early days. There were only programming examples in C, none in Pascal.
> And as I remember trying to learn TPW 1.0, it was a real nightmare.
>

It also had an uphill struggle against C++, which rapidly established
itself as the "pro" Windows programming standard thanks to several
capable OO applications frameworks (C++/Views, zApp, Zinc, etc). This
together with the fact that most of the (then rather few) existing
Windows
programmers were accustomed to C rather than Pascal (and therefore
found C++ to be a natural route to the OO which simplified Windows
programming), and that C++ could use all the existing Windows C
libraries, counted heavily against Turbo Pascal as a Windows development
language.

>
> > Anyhow, at that time, 91-94, Borland still had by far the best C++ compiler
> > for Windows, and that part of the business did well.
>

They actually had the first Windows C++ IDE, and the first debugger that
could
run in the same Windows session as both IDE and the application being
debugged
(previously, you had to debug your apps on another machine). Borland's
system was
far ahead of the competition (primarily MS and Zortech) in this respect,
and in
those days Borland gave you everything: Windows IDE, DOS IDE, Windows
application
framework, DOS application framework (if the two had been
code-compatible, MFC would
never have stood a chance of displacing them - sadly, they weren't) and
an assembler.
That (Borland C++ 3) was IMO the last really great Borland product;
afterwards they
started trying to charge us extra for bits that had always come in the
box, and the
advent of VC++ 1 (whose wizards and other IDE-based assistants were
fairly revolutionary
in the C++ world, helping to make up for a notably flaky compiler)
sealed Borland's fate
by changing them from a standard-setter to somebody who was seemingly
always running to
catch up with what MS were doing. Even Delphi was little more than a
"Pascal VB".

>
> I seem to remember the early Borland C++ compiler being capable of DOS,
> Windows, and OS/2 compilations.
>

The OS/2 compiler was a separate product. It might have stood a chance
if
Borland had ported the (then rather skeletal) OWL to OS/2, but they
didn't,
so people had to program using the OS/2 SDK, which made OS/2 programming
a lot more difficult and less productive than Windows programming was
with
the equivalent Borland tools. BC++ for OS/2 thus offered little that
wasn't
available with the IBM and MS compilers (MS C could produce "family"
executables which ran natively in both Windows and OS/2), and Zortech's
OS/2 compiler produced much better code. Fortunately, Borland lost
little
on the deal because IBM paid them to develop the compiler in the first
place, although I wouldn't be surprised if this failed adventure had
more than a little to do with Borland's subsequent reluctance to
experiment
with platforms other than Windows.

rrk

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to

POST SCRIPT

After leaving (getting kicked out, whatever) Borland, the French-born
mathematician and founder of Borland, Philippe Kahn, founded Starfish (I
think that was the name).

After a couple of years of development, Kahn sold this company to ...
Motorola I think (not sure) for $300,000,000 USD. At least that was the
scoop in the trade journals.

Since he left the company, Borland/Inprise has pretty much foundered.
Moreover, Borland/Inprise lost many of the most talented creative
individuals who made the company what it was.

As I understand (and I was NOT there), Kahn was a major PITA to work
with and a most difficult man in the French dictatorial manner (another
comment on the French Philippe <G>). However, the results speak for
themselves.

William H. Mogk

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
I bought the original Turbo Pascal for the Apple (the Apple is long
gone, but I kept the manual for posterity). It was the first piece of
software that I ever mail ordered from the U.S. I seem to recall that I
paid $66 Canadian plus shipping for Version 1.0. In those days I
struggled with one, single-sided diskette drive, 64K and an 80 column
card. TP took only half that memory and you could write some good stuff
in the remaining 32K.

Me and my partners prototyped a taxi management system for an MIS
course, and I did the programming with the Apple and TP. We brought the
computer into class for the demo, way before there was such a thing as a
lap-top. I remember the class was crowded around the little 10"
monitor.

My prof was quite impressed, since almost everyone at our university was
programming only on VAX's, DEC's and IBM 3xx's.

I was and still am a Borland fanatic to some extent. I bought most
borland stuff, except DBase and Sprint. I was at a Borland conference,
in Toronto when they demoed a late beta of QP for Windows. I was
impressed at the demo and purchased it when it was released.

Maybe my needs are simple, but Delphi has been very good to me right
from version 1 to version 4.


Bill Mogk


Roger Arnesen wrote:
>
> >In 92 BP7 (Borland Pascal 7) came out, and it was the second version for
> >Windows.
>

> Ehhh.. .. 3. version for Windows. I personally used TPW 1.0 and 1.5
>

> >Since this post is more than long enough, and has stuck to trying to be
> >factual, I'll leave off discussing the next two years.
>

Rune Moberg

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Philippe Ranger wrote:
> In 92 BP7 (Borland Pascal 7) came out, and it was the second version for
> Windows. I may think the world of it, but the world thought otherwise, and
> it was a marketing dud. I truly don't understand why, but the common
> explanation is that by that time VB was gaining a lot of attention. Anyhow,
> at the time Borland desperately needed at least to hold on to its old
> revenue sources.

TPW 1 and BP7 were the first Borland packages I bought (I'm fairly new
at this stuff, didn't start doing Pascal until '90 with TP5 at school).
However, I seem to remember at that time that the trade press was all up
in arms about "portability". It all had to be portability-this and
portability-that. The solution offered? C/C++.

Since C++ got popular, the size of software has grown considerable, the
performance has dropped and memory requirements soared.

Although VB was becoming predominant, it wasn't until VB3 people
actually started using the blasted thing. BP7 as I recall was released
prior to VB3?

OTOH, VB probably enabled a whole new generation of "programmers" to
whom "two-by-four" actually means something related to carpentry
(instead of the size of a disk or something). So although BP7 sold like
hotcakes (I seem to recall that it sold more than its predecessors?), it
still didn't cater for the unwashed masses to come.

One good thing did emerge from all this: Had BP7 become a bigger
success, if everyone had dumped C/C++/VB in favour of BP, then Borland
would probably never have bothered doing a Delphi. Borland made a lot of
mistakes during this period, e.g. they didn't release a NT compiler, but
was silent for many years. Until '95 that is (we still had to wait
another year for something that would cater for us NT users!).

The current predicament however baffles me. Delphi has sold more than 1
million licenses (that's last years numbers?) and yet there are parts of
our industry that hasn't heard of it. (and now another round of
lay-offs?)

I guess all the technical acronyms finally got to us. There are dark
holes in Delphi that I have never approached. Entera? VisiBroker? I even
mistook MIDAS for something else entirely, and thought three-tier could
be done without involving the complex MIDAS licensing stuff. (gee, why
is everyone looking at me like I'm coming out of the closet or
something?)

There are a lot of needs out there, and I'm not sure one tool can cater
to them all. Delphi is probably very-very close to achiving this
(atleast closer than VB which is just a gluegun that caters to MS
specific technology). Meanwhile every new release looks more like a
monster than anything else. I've ended up just installing the new
version, and never dwell on things I haven't immediately got any use
for.

And perhaps that's a clue? If I ever need something more than what I'm
currently doing, Delphi is more than likely to provide this. Had I been
using VB or PB, that option would be closed off from me.

The trick here is to convey this to today's VB users: Ask not what your
tool does today, but what _you'll_ be doing tomorrow.

> Anyhow, at that time, 91-94, Borland still had by far the best C++ compiler
> for Windows, and that part of the business did well. But MS could just pour
> money into compiler improvements (starting at least three full years late),
> and finally managed to make its VC++ the standard, before Win95 showed up. I
> say this through gnashed teeth, the Borland solutions at the time were
> almost always superior.

Ah, I still remember the sweet taste of victory when PCWeek's TAK
benchmark showed a better integer performance with D2 compared to VC++4.
Unfortunately by taking a couple of shortcuts, VC++ aced the floating
point test.

--
Rune, http://runesbike.com

John Elrick

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Philippe,

Started with TP3 in the 80's. I also used TB and <gasp> Turbo Prolog. And
of course, you missed Reflex. <g> I built a full inventory system with
Reflex.

Anyhow, thanks for the great - and factual - recounting of Borland's
history. You should (seriously) write a book about it.

John


Philippe Ranger <.> wrote in message <79i40t$qc...@forums.borland.com>...

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Barry: >>Go on, give it a go. In this case the truth is stranger than

fiction, so the factual should suffice to keep your
readership tapping the scroll bar.
<<

Too close to the bone. This is Inprise's forum, it's no place to start
picking them apart just from curiosity. Go see Motley Fool, or some such.

PhR

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Mark: >>I believe dBase was a port over from the CP/M world onto the "new"

PC-DOS. dBase II (at the time) was pretty much the only option, and was
widely popular with those of CP/M heritage.
<<

As Richard says, it was II from the start and, yes, it came fully-armed from
CP/M. Little major change until III.

By the way, after being hit by the combination of, one,
low-cash-high-expenses caused by that purchase and, two, the MS onslaught in
Windows apps, Borland sold Quattro to Novell, which under Noorda was into a
policy of anything's good if it attacks MS. In the same deal (very early 94,
I think), Novell bought rights to Paradox, and bought the WordPerfect Corp.
outright. About two years later, it sold the lot to Corel for about 12 cents
on the dollar, and Corel soon bought the whole of Paradox from Borland.

DBase, you might say, was left on the cutting room floor.

PhR

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Glynn: >>... and you never mentioned Turbo Prolog. I think it lasted almost

1
whole year somewhere around 1988.
<<

Sorry. Lasted more than a year. I *think* the engine was licensed from some
French developers, and that this was even mentioned when you ran it. An
interesting and justified trial balloon, even if nothing came of it.

PhR

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Richard: >>Excel was originally written for the Mackintosh, a machine that

Bill
Gates liked because he felt that GUIs were the way to go (he literally
begged
Apple to license Mac ROMs to clone makers in order to establish it as a
standard; they didn't listen - if they had, then history might have been
rather different).
<<

Yeah. Now that Gates is the richest man in the world, no matter where people
start from, the popular sport is to depict MS as the Devil. Mac Mavens have
a very strong team in that indoor sport. But the fact is that MS, and Gates
personally, were involved in the Mac from long before its launch. In the
case of Gates, he truly had faith that this, not the PC, was the way to make
personal computers. And indeed without MS the Mac would have sunk like a
rock — no apps to justify the price.

It's impossible at this point to convince people the Mac isn't the model for
Windows, but the fact is that MS was closely involved in the Mac model
itself, and that the successive version of Windows were much more an attempt
at pursuing the same faith on the PC side, than an attempt at getting the PC
to "work like" a Mac.

Gates has always had advice for Apple regarding the Mac, but I didn't know
about his early position regarding clones. Speculative business history —
The PC was a kludge, but one that inherited from an open architecture, CP/M,
while the Mac on the contrary was a veering away from any openness, even the
half-openness of the Apple II. After many years, it became clear that it is
possible to build an open Mac from an open something else, but not from a
closed Mac.

PhR

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Mark: >>Again, Lotus was a migration from the CP/M VisiCalc (I believe that

was
the name). Kapor had apparently written the earlier product, and owned
the rights for the port.
<<

Someone who worked on Lotus had worked for VisiCalc in CP/M times, but no
transfer of rights was involved. Lotus started with a clean slate. It was
noted at the time because 1-2-3 was the first app for which the launch
budget equalled the development budget.

>>
Wasn't MS offering at the time a PC-DOS port of MultiPlan, or some such
product? I thought Excel came in with the first Windows offering...
<<

Yeah, I think MS had a Dos spreadsheet like that, but saw no way to develop
it. It's first noted attempt was Excel for Win2, delivered with a Windows
runtime. Exceedingly slow and greedy for PCs at the time. Lotus all by
itself had caused the development of EMS, and with enough of it was more
powerful than Excel.

PhR

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Bob: >>After a couple of years of development, Kahn sold this company to ...

Motorola I think (not sure) for $300,000,000 USD. At least that was the
scoop in the trade journals.
<<

I'm not sure about the price either, but this happened around last November.
I remember thinking that, if Philippe could sell that one-tune orchestra for
that kind of money, in so little time, well... what couldn't he have done
with Borland? Only, in fact, he tried and failed, and several rumors have it
that deals failed because of Philippe's extravagance. Maybe he's learned
something.

PhR

Glynn Owen

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
It was an interesting idea. I bought it just to have a look, but I
couldn't get over the paradigm shift from procedural to declarative. As
I recall at the time, the Japanese were promising some wonderful stuff
using Prolog, but I don't remember hearing any more about it.

Regards, Glynn

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Rune: >>However, I seem to remember at that time that the trade press was

all up
in arms about "portability". It all had to be portability-this and
portability-that. The solution offered? C/C++.
<<

And for the last four years the same professional false-road indicators have
been drumming Java, Java, J-A-V-A Java! You can only port the functional
subset common to all platforms you're porting between. But the platforms are
different NOT because of this common subset, but because of their
non-portable capacities, that people happen to want.

>>
Although VB was becoming predominant, it wasn't until VB3 people
actually started using the blasted thing. BP7 as I recall was released
prior to VB3?
<<

Yes indeed. But, no, I think VB was in serious use well before version 3. In
fact, the users were well ahead of what the tool was designed for. (This is
like what people made of 1-2-3 macros eight years before.)

>>
So although BP7 sold like
hotcakes (I seem to recall that it sold more than its predecessors?), it
still didn't cater for the unwashed masses to come.
<<

We have memory mismatch here. I thought BP7 should sell like hotcakes, but
my various indicators (mentions in articles, books in bookstores, clientele
on Compuserve forum, etc.) remained stuck in the low position. Later I read
that it had sold nowhere near what TP got in its heyday.

>>
The current predicament however baffles me. Delphi has sold more than 1
million licenses (that's last years numbers?) and yet there are parts of
our industry that hasn't heard of it. (and now another round of
lay-offs?)
<<

I have no idea where those "sold" figures come from, that Borland people
quote from time to time. They're not part of my indicators either way. One
indicator, though, is the quarterlies.

PhR

Philippe Ranger

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
John: >>And of course, you missed Reflex.
<<

Not trying to list everything! What about Object Vision? But actually I
didn't think of Reflex, you're right.

PhR

Bob Arnson

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
rrk <ezco...@earthlink.net> wrote in message
news:36BDBB2A...@earthlink.net...

> Since he left the company, Borland/Inprise has pretty much foundered.

And long before he "left." Borland's biggest problems -- the delays in
getting Quattro and Paradox out for Windows -- were under Philippe's
leadership.

--
sig://boB/TeamB

John Elrick

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
So easy to miss, my friend :-)

John


Philippe Ranger <.> wrote in message <79l6fk$9...@forums.borland.com>...

John Elrick

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
As I mentioned elsewhere, I used Turbo Prolog.

I bring it up because I actually wrote several programs with it. It was
slow and bloated, but very interesting and fun to program in.

I actually kinda miss it.

John


Philippe Ranger <.> wrote in message <79l2d0$ss...@forums.borland.com>...
>Glynn: >>... and you never mentioned Turbo Prolog. I think it lasted almost


>1
>whole year somewhere around 1988.
><<
>

Julian Bond

unread,
Feb 7, 1999, 3:00:00 AM2/7/99
to
Damn this thread makes me feel old. I remember running these at one time
or another and installing many of them onto people's machines in a
famous US trading house.[1]

- Ansa Paradox
- Sidekick 1.51, aka the most heavily pirated software in the world.
- Superkey (sp? a keyboard macro TSR)[2]

Remember when you had 5 essential TSRs and used a utility to swap them
in and out because they all had to be loaded last?

Also wrote several install and machine checking utilities along with a
DOS menuing system in Turbo Pascal 3.

And the best bit was that every single product was <$50 (apart from Ansa
Paradox which was more like $1000)

And how about the Florida Devcon where they were showing the first pre-
release copies of Delphi?

[1]While we're on the nostalgia kick, how about IBM Topview, Desqview
and those people who did a complete MS Windows API for DOS

[2]Either this or Sidekick also did auto-dialing from any telephone
number on a DOS screen using a Hayes 1200. Something, I haven't been
able to do effectively since Windows appeared and the corporate phone
systems all went to digital phones.

--
Julian Bond <mailto:jb...@palmstech.com>
Palms Technology U.S. Inc. <http://www.palmstech.com>
+44 (0)1920-460297
"So many words, so little time"

Marcel Popescu

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Rune Moberg wrote in message <36BDD692...@runesbike.com>...

>I even
>mistook MIDAS for something else entirely, and thought three-tier could
>be done without involving the complex MIDAS licensing stuff. (gee, why
>is everyone looking at me like I'm coming out of the closet or
>something?)


Three-tier CAN be done without MIDAS - I have 3-tier programs in D4
standard. MIDAS is just simpler to use. [But then I suffer from the
Not-Developed-Here syndrome. <g>]

Mark


Julian Bond

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
In article <79l2d6$ss...@forums.borland.com>, Philippe Ranger <?.?@?.?>
writes

>Bob: >>After a couple of years of development, Kahn sold this company to ...
>Motorola I think (not sure) for $300,000,000 USD. At least that was the
>scoop in the trade journals.
><<
>
>I'm not sure about the price either, but this happened around last November.
>I remember thinking that, if Philippe could sell that one-tune orchestra for
>that kind of money, in so little time, well... what couldn't he have done
>with Borland?

Maybe Borland should follow Apple and invite their leader back.

Let's have a rallying call which I will shorten to B4.

"Bring Back Borland's Barbarians"

Richard Grossman

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to

Philippe Ranger wrote:

> Concerning dBase, there was a rumor Borland had a dBase for Win in its

> files...

They had Turbo-base for DOS and it blew away dBase IV but was very slightly
behind and always chasing Fox.

Too bad Borland didn't buy Fox and let dBase go to Microsoft...


.....................
Richard Grossman
rgro...@techIII.com

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Richard: >>computers would have to become more or less like any domestic
electrical
appliance
<<

Either you have the wrong author, or Gates was quoting someone else (Alan
Kay?) who practically has a trademark on "the computer as appliance".

>>
That's more or less what happened. History has shown again and again
that moving from open architectures to closed ones results in a loss of
market share for the originator of that open architecture,
<<

Pointing again to the gap Delphi can occupy — component-based app-building,
where you can get the source to all components and recompile it at need, vs.
the closed components used for VB, and the lack of true RAD for VC++ (not to
speak of the language).

PhR

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Leroy: >>IIRC, VisiCalc was an Apple II product, not CP/M. In fact, VisiCalc
was
*the* product that drove many businesses into buying Apple.
<<

I could check this, but I'm too lazy. I still think Visicalc ran on the Z-80
card, therefore on CP/M. But the Apple II was just about the only thing with
diskettes and enough of a market to make their particular format (among a
score) a relative standard. So, to a business, it was the default Z-80
installation.

PhR

Marcel Popescu

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Philippe Ranger <.> wrote in message <79ml27$1m...@forums.borland.com>...

>I could check this, but I'm too lazy. I still think Visicalc ran on the
Z-80
>card, therefore on CP/M. But the Apple II was just about the only thing
with
>diskettes and enough of a market to make their particular format (among a
>score) a relative standard. So, to a business, it was the default Z-80
>installation.


VisiCalc definitely ran on Z-80. But this didn't mean only CP/M - I remember
having either VisiCalc, or a very close clone on Spectrum!

Mark


Janet De Lu

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
A few more details:

The company's original name was MIT (I forget now what that stood for). The university was making legal noises and they renamed it Borland, taking the name of another company that owed them money when it went bancrupt.

My personal take on it was that Windows was the beginning of the end. Borland bet on OS2 originally, and was late to start on the Windows platform. By the time they got their products ported to Windows, Microsoft had market dominance in everything but databases. Paradox had a brief day in the sun, but was soon wiped out by a combination of bad management decisions and a price war they couldn't win.


Joe C. Hecht wrote:

> Wow Philippe! That was a pretty darn good write up!!!!
>
> Mind if I add a couple of things???
>
> I heard that Phillipe started Borland because no
> one would hire him.
>
> Borland rose to be the second (or debated third)
> largest software institution.
>
> Phillipe was always rooting for the little guy.
> He never forgot what it was like to be a small
> and struggeling kitchentop programmer, and commonly
> loved to take swings at the knee's of the big boys.
>
> Phillipe revelotionized software's legal side with
> the famous "Borland No Nonsense License Agreement".
>
> As I recall, at one time, Borlands legal department
> produced a series of legal documents for the small
> programming guy, and offered them via fax and BBS.
>
> I believe the Borland forum on Compuserve was
> (for awile) the largest forum base that Compuserve
> had.
>
> Turbo Pascal was actually a product called PolyPascal,
> produced by a kid Phillipe had gone to school with
> (Anders) under Professor Niklaus Wirth (the father
> of Pascal)
>
> Phillipe plays the Sax, and loves Turbo Porche 911's.
>
> StarFish has been pretty darn successful for a small
> quiet private company that resides in Borland's old
> offices.
>
> For good or bad, Phillipe got into a pricing war
> with Microsoft(?). I think it was a Quattro vs Excel
> war, dropping the price of the software down to $49
> (if I recall correctly). This forever changed the value
> of business software in the open market. (remember
> when PeachTree was $5000???)....
>
> The Borland campus (Borlands only tangable assett)
> is valued at 100 million, and is a major thorn for
> the company. It seems no one wants the building.
> This is probably one of the biggest reasons Borland
> has never been bought out <g>.
>
> Borland has since sold or licensed out most of the
> non-hard-core developer and end user products, mostly
> to the Corel folk...
>
> For awhile, Delphi's massive revinues have helped
> to float the company and development of of other
> products. Perhaps it still does ;)
>
> Joe
> --
> Joe C. Hecht
> http://home1.gte.net/joehecht/index.htm


Richard Bayarri Bartual

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Philippe Ranger wrote:
>
> Yeah. Now that Gates is the richest man in the world, no matter where people
> start from, the popular sport is to depict MS as the Devil. Mac Mavens have
> a very strong team in that indoor sport. But the fact is that MS, and Gates
> personally, were involved in the Mac from long before its launch. In the
> case of Gates, he truly had faith that this, not the PC, was the way to make
> personal computers. And indeed without MS the Mac would have sunk like a
> rock — no apps to justify the price.
>

From the earliest days of MS (i.e. pre-IBM), Gates' vision was a
computer
in every home. I remember an interview years ago (although in what
magazine
I can no longer recall) in which he stated that for this to become a
reality,


computers would have to become more or less like any domestic electrical

appliance, i.e. you just plug the thing in, turn it on, and use it. The
Mac's little "all in one box", its GUI, mouse, and everything about it
fitted in with his ideas of what a "computer for ordinary people" should
be - you didn't even have to learn cryptic OS commands to use it. He was
very enthusiastic about the machine (as indeed he had been about the
earlier, experimental, and very expensive Lisa), but he was convinced
that
no single company could possibly dominate the entire market because
attempts to do so would result in competing hardware and software
standards which would fragment and confuse the market, thereby reducing
the overall world demand for microcomputers. The key to enduring success
for Apple (according to Gates) would be through licensing firmware, OS,
and various other necessary bits and pieces to third-party
manufacturers,
thereby establishing a global standard which Apple would earn money from
no matter who was actually building the computers people were buying.

Apple didn't listen, and nearly died as a result of being too greedy;
by the time they eventually decided to allow Mac clones, they'd already
lost the standards battle to Wintel (whose success was based on the very
model Gates suggested to Apple before the first Mac ever rolled off the
production line).

>
> It's impossible at this point to convince people the Mac isn't the model for
> Windows, but the fact is that MS was closely involved in the Mac model
> itself, and that the successive version of Windows were much more an attempt
> at pursuing the same faith on the PC side, than an attempt at getting the PC
> to "work like" a Mac.
>

This argument would only hold water if Apple had invented the GUI, which
they didn't. What Apple did more or less invent was the idea of having a
pull-down menu bar in a GUI (although menu bars had been used previously
in text-mode applications), and this was the crux of the "look and feel"
action against DR and MS (the earlier Xerox GUIs used pop-up context
menus). However, as you say, Gates was not so much trying to copy Apple
as making a statement (along with DR, Epson, and various other companies
who produced GUIs) that there was a new and easier way of doing things
with computers that promised to make them more generally approachable.
Unfortunately, the PC platform proved ill-suited to the concept because
of memory constraints, so GUIs were not really successful on it for
nearly a decade (although they were extremely successful on the more
powerful Atari ST and Amiga, which sold to a lot of home users, thus
proving Gates' idea that GUIs were the key to his vision of a computer
in every home).

>
> Gates has always had advice for Apple regarding the Mac, but I didn't know
> about his early position regarding clones.

The legend is that he actually went down on his knees in an effort to
get them to franchise the Mac firmware and OS, but this doesn't sound
very Gates-like to me (he was never modest, even when MS were a tiny
BASIC house). He did however try very hard to make Jobs and Wozniac
see that they could actually make more money selling operating
software to all of the people than by making the entire machine for
just some of them.


> Speculative business history —
> The PC was a kludge, but one that inherited from an open architecture, CP/M,
> while the Mac on the contrary was a veering away from any openness, even the
> half-openness of the Apple II. After many years, it became clear that it is
> possible to build an open Mac from an open something else, but not from a
> closed Mac.
>

That's more or less what happened. History has shown again and again
that
moving from open architectures to closed ones results in a loss of

market share for the originator of that open architecture, even though
the impetus for such a move is the desire to earn money by dominating
the market. In the pre-micro days for example, IBM built an open
big machine called the 360, and it was the Wintel of its day, becoming
so dominant that entire industries grew up around it (3M for example
started out making accessories for the 360). IBM were not happy with
this
because they actually made their money in those days by what we now
call "dumping", i.e. they'd sell you a basic system for less than what
it cost them to build it, and then stripe you up for the expansions
that you quickly discovered were necessary to make the thing actually
do anything. This plan didn't work anything like as well as it should
have because all manner of upstart little companies sprung up offering
360 expansions (and even replacement machines that performed faster
and were smaller and cheaper) for a lot less than IBM were charging -
the
360 was the first machine in history to be "cloned".

IBM's answer to this problem was the 370, a more powerful system that
supposedly offered all sorts of advantages (or at least would have if
the OS had been finished - unfortunately, it wasn't, so IBM's customers
ended up writing most of it for them). The 370 was also not an open
system like its predecessor: IBM changed all the sockets that connected
it to perpherals and refused to divulge the pin-outs, and it used a
different instruction set, so all existing software had to be recompiled
to run on it (and this recompilation wasn't a simple "bung it in the
front and it comes out the back working" job).

To IBM's surprise, instead of jumping on the 370 bandwagon, a lot of
their existing customers decided to stick with their 360 setups
because they could upgrade to a "clone" CPU which actually offered
better performance than the 370, and still make use of all their
existing peripherals and software without the expense of recompiling
(i.e. rewriting) it. A few decades later they repeated the same
mistake with something called PS/2, thereby proving that being a
part of history is no guarantee that one will learn from it...

As you said, Apple did the same with their move from the Apple-II
(which, like the 360 and later PCs, had an entire industry dedicated
to producing peripherals and expansions for it) to the much more
"hush hush" Mac, thereby losing on the swings what they'd gained on
the roundabouts by having a more capable machine with an interesting
and new (in the home/small business market) idea of using a GUI.

Meanwhile, those who have stuck to open standards (e.g. Compaq) have
gone from strength to strength, while Bill Gates has made a fortune
doing what he told Apple would make them a fortune all those years
ago, and the once mighty, protectionist IBM have been forced to use
somebody else's open architecture running Mr. Gates' OS. In other
markets, Ford own Jaguar and Volvo, and VW have R&R, thereby proving
once and for all that there's a lot more money to be made selling cheap
standard things to everybody than expensive specialist ones to a select
few - always assuming of course that you are offering something they
actually want to buy...

A.A.Katz (Alan)

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
<sigh>

So much for more dBASE disinformation :-)
dBASE was the database standard from late 1982 with its CPM release, DOS in
1983.
dBASE was the database standard on the PC long before Paradox existed (or at
least before it saw the light of the market).
As to "inferior" to Paradox, well... that's a matter of opinion isn't it?
I've used both for years and found dBASE to be much more robust.
Paradox observed a more SQL-Like behavior, but the ubiquitous .dbf table had all
kinds of support (especially in its indexes) and far fewer files needed to
support its capabilities. It also had a strong clone market generating products
like Fox, Clipper, Force and DBXL each of which improved upon the original,
especially in adding any variety of compilers (including a native code compiler
in Force).
The great thing about Xbase is that it was infinitely "growable" in all its
various guises.

Philippe Ranger wrote:

> John: >>
> Could someone please bring me up to speed with Borland history, in
> particular where things started to go wrong. I know that in the mid to late
> eighties, Borland were hurting Microsoft ("Kill Philippe!") with products
> like Turbo Pascal and Quattro Pro. Is it true that Delphi was a make or
> break product? What's the background on this Del character? All I know is
> that as well as running the show, he's grossly overpaid and has a fondness
> for pizzas. What happened to Philippe Kahn anyway? And just who was Frank
> Borland? Sorry for my ignorance.
> <<
>
> I have never worked at or for Borland, I have no inside story here, and I'm
> working from memory. So —
>
> I don't know when "Frank" Borland popped up, but the rumor I tend to
> subscribe to is that Borland was taken as an American-sounding name for a
> company, because of Frank Bormann, the astronaut.
>
> Borland came to attention in late 1983, with an ad in Byte for Turbo Pascal,
> at $50 by mail. That product was so innovative in so many ways (marketing
> being not the least of them), someone should do a book. Anyhow, TP took "the
> world" by storm. Version 2 came out in June 84 and version 3 in April 85.
>
> At about that time, Borland, now an established company, started coming out
> with a slew of utility-type products. One of the very first, Sidekick (not
> like the current thing) was again a major innovation that marked the
> industry. Especially, Sidekick's user interface slowly became the standard
> for all PC apps until Windows 3 came along 5 years later.
>
> Now, from all appearances, this stuff was bought, and seldom did the authors
> stick around for recasts. Borland was unable to follow up on any of its
> utilities, even SK, which remained stuck at version 1.53. Likewise, TP's
> very innovative "IDE" used an editor built on a binary-code machine Borland
> did not own the code for. That same machine was offered separately in an
> "Editor Toolbox" for TP3, and, paf! editors sprouted by the dozen
> (literally). Almost ALL Dos-based editors were born from this, unless they
> were ported from VMS, Unix, etc. But Borland could not improve in any way on
> the original binary editor (bined).
>
> Likewise, in the summer of 86, if memory serves, a summer intern was hired
> to "port" to TP to Basic, and Turbo Basic was born and launched. Again, it
> could not follow on itself, and eventually ended up resold to its author. I
> believe the first Turbo C was also a renamed third-party product. But in
> this case Borland did manage to update it and, through several versions,
> make it "theirs". Borland C++ v. 1 appeared in summer 89 (if memory serves),
> and it was truly Borland. Also, again, earth-shaking.
>
> Though there was a long lull before TP4 (last days of 87), the fact is that
> at least TP was a purely-Borland enterprise. The author was Anders
> Hejlsberg. Anders more or less moved to other things, I understand, while
> Delphi 2 was a-building, and soon after was hired by Microsoft by paving his
> path with gold. At the same time (again, if memory serves) the Delphi
> business manager was similarly, er, seduced. But I'm getting ahead of
> myself.
>
> Anyhow, Anders was little-known except among TP afficionados (the kind of
> people who hang around here), and Borland was identified with its president,
> Philippe Kahn, a Frenchman. The more Borland grew, the more Kahn became a
> public figure, and by the late 80's he was sounding a bit like the Divine
> Ellison. As long as Borland found NEW products to launch, in the brouhaha it
> went rather unnoticed that it couldn't follow-up on any of its past
> successes except TP and TC.
>
> One exception. Rather early, perhaps 85, Borland bought a very promising
> database **maker**, Paradox, and simply left it to its own devices. But the
> product itself was never more than an also-ran on the market, compared to
> the remarkably inferior dBase, which willy-nilly made itself the standard
> from 85 onwards.
>
> So Borland bought the oldest WP app on Earth, which it relaunched as Sprint.
> It bought one spreadsheet maker (everyone was making better spreadsheet than
> Lotus, but Lotus was totally the standard), and let it disappear from view.
> It bought another one, Quattro (Spanish for "beyond 1-2-3), got to work on
> it (and kept at it), and quickly got a massive suit from friendly Lotus.
> That really hurt prospects. A bit later, it bought the best-known
> programmer's editor, Brief, and let that disappear too. Btw, at the time
> Excel was a perpetual money-loser for MS and Quattro wasn't really a threat.
> The war was with Lotus.
>
> And in the summer of 91 it bought not dBase, but the whole of Ashton-Tate
> (twice the employees but not twice the revenues). At a price I don't
> remember but found unbelievably inflated (this was pre-Web times, people
> actually figured sofware companies the way you'd figure tire companies).
>
> Then things really, really went wrong. There was no way Borland could digest
> A-T fast enough to get all the nourishment, and it really needed it for the
> price it paid. Also, by that time every one's major concern was, not just
> getting out a Windows version, but getting out a superior Windows version.
> Everything had to be redone. If you notice, WordPerfect Corp., which totally
> dominated its (larger) market and bought nothing also lost it at that point.


>
> Concerning dBase, there was a rumor Borland had a dBase for Win in its

> files, which it could not launch because of the legal threat from A-T. It
> seems it wasn't true, dBase for Win was quite late. And Paradox was late
> too. Quattro held on by the skin of its teeth, but in the meanwhile MS
> knocked the teeth out of the true enemy, 1-2-3, which was also using the
> Kahn policy of buying stuff right, left and center. Lotus wound up sold to
> IBM mostly for one of those purchases, Notes. I cite WPC and Lotus to show
> that, with very different management styles, other MS competitors also went
> down the drain during the early nineties.


>
> In 92 BP7 (Borland Pascal 7) came out, and it was the second version for
> Windows. I may think the world of it, but the world thought otherwise, and
> it was a marketing dud. I truly don't understand why, but the common
> explanation is that by that time VB was gaining a lot of attention. Anyhow,
> at the time Borland desperately needed at least to hold on to its old
> revenue sources.
>

> Note. Del has clearly operated an developer/enterprise switch. "Developer"
> is practically synonymous with the old (eighties) market for TP. The people
> who really saw what the enterprise end was, in marketing, were people at MS.
> They brought in the notion of the Suite, early 91, and this is really how
> they killed off both giants so quickly, Lotus and WPC. Well, enterprise
> appeal was what was missing from BP7 marketing, and what quickly grew
> regarding VB.


>
> Anyhow, at that time, 91-94, Borland still had by far the best C++ compiler
> for Windows, and that part of the business did well. But MS could just pour
> money into compiler improvements (starting at least three full years late),
> and finally managed to make its VC++ the standard, before Win95 showed up. I
> say this through gnashed teeth, the Borland solutions at the time were
> almost always superior.
>

> Well, besides BC++, Borland had mostly trouble. Truly, the remarkable thing
> is that it did NOT go down the drain, the way WPC and Lotus did. But both
> its databases were disappearing from view, all other A-T products had been
> dropped, Quattro was a poor second to Excel, and BP was... well was having
> no new version.
>
> There was a rumor about a "VBK" (VB Killer) in the works as the successor,
> but clearly Borland did not have the money, if not to develop it, to market
> it. In any case, Delphi 1 came out in early 95, at a point where any notion
> of a VBK was ridiculous. It did meet with remarkable success, considering
> conditions. Stock watchers, who had nothing else to hang on, said it was
> "make or break". This was obviously true from the balance sheet. If money
> didn't come in from Delphi, pretty soon there wouldn't be enough to print a
> balance sheet on. I do think BCB and Jbuilder were financed on the returns
> (and prospects) of Delphi. But I may be wrong. Anyway, Delphi gave Borland a
> new life in (rather limited) public notice, for a time.
>
> Among the expenses Borland could do without was the salary paid Philippe,
> especially as his style wasn't what was needed by that time. So he went out
> the door (can't remember exactly when, 94 probably), taking with him a nice
> severance package and one product, the new Sidekick. On this he built
> Starfish software, currently noted mostly for the Rex credit-card-sized PDA,
> for which Starfish has rights on the sw.
>
> Replacing him, at much lower salary, was not easy. Especially as Delphi
> didn't keep climbing on the charts. The CEO's office had something of a very
> unthankful revolving door, and by summer 96 was occupied by a member of the
> board, because someone had to be there. A search for a "real" CEO was
> seriously on, and seriously quite tough. In the end (Nov. 97, I guess) the
> job went to an ex-Apple honcho who had left during one of miracle-worker
> Amelio's cleanups, Delbert Yocam. Someone caved in, there, because Del is
> more expensive than Philippe ever was.


>
> Since this post is more than long enough, and has stuck to trying to be
> factual, I'll leave off discussing the next two years.
>

> PhR

--
--- A. A. Katz (Alan)-----------Borland TeamB----
--- Ksoft Inc
--- Johnson City, NY
--- http://www.ksoftinc.com
---
--- Editor, Visual dBase Web Magazine
--- http://www.ksoftinc.com/vdb
-------------------------------------------------

A.A.Katz (Alan)

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
A couple more corrections:

1. As part of Phillippe's severance, he got two products, not one: Sidekick and
Dashboard. Both were included in Corel's suite.

2. Starfish was sold last year to Motorola, which put more than $15 million on
Inprise's balance sheet as a result of its 10% stake in Starfish.

3. I differ with much of your analysis as to what went wrong. What went wrong,
to a great extent, was Microsoft - and Phillippe's inability to walk away from a
head-to-head battle with Gates. If you remember, much of the resources of the
early '90s went into the Borland/Word Perfect Suite (later the Borland/Novell
suite), which was a direct competitor to the MS Office Suite. Obviously, it lost
the battle and left Borland with an unfocused collection of products and no
office suite to sustain them.

The same for dBASE. Contrary to what you indicate, and despite Borland's
problems in getting a clean compiler and Windows version out, dBASE produced
very significant revenue for Borland for a number of years. What happened to
dBASE and Paradox both was -Access-. When Borland bought dBASE, it sold for $895
a copy with "Lan Packs" sold on a per-seat basis. Once Gates launched Access as
a $99 stand-alone product and a member of the "Pro" suite (specifically targeted
as a "Kahn Killer",by the way), the handwriting was on the wall for both Paradox
and dBASE as "end-user desktop databases".

A.A.Katz (Alan)

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Once again, I disagree as to history. :-)

The Mac was the successor the the Lisa.
Both were the successor to the graphic user-interface developed at Xerox PARC.
Microsoft -licensed- the Windows interface from Apple.
The original GUI was tiled. Microsoft changed it to "nested" and overlapping
windows which resulted in the famous lawsuit between MS and Apple.

Philippe Ranger wrote:

> Richard: >>Excel was originally written for the Mackintosh, a machine that
> Bill
> Gates liked because he felt that GUIs were the way to go (he literally
> begged
> Apple to license Mac ROMs to clone makers in order to establish it as a
> standard; they didn't listen - if they had, then history might have been
> rather different).
> <<
>

> Yeah. Now that Gates is the richest man in the world, no matter where people
> start from, the popular sport is to depict MS as the Devil. Mac Mavens have
> a very strong team in that indoor sport. But the fact is that MS, and Gates
> personally, were involved in the Mac from long before its launch. In the
> case of Gates, he truly had faith that this, not the PC, was the way to make
> personal computers. And indeed without MS the Mac would have sunk like a
> rock — no apps to justify the price.
>

> It's impossible at this point to convince people the Mac isn't the model for
> Windows, but the fact is that MS was closely involved in the Mac model
> itself, and that the successive version of Windows were much more an attempt
> at pursuing the same faith on the PC side, than an attempt at getting the PC
> to "work like" a Mac.
>

> Gates has always had advice for Apple regarding the Mac, but I didn't know

> about his early position regarding clones. Speculative business history —


> The PC was a kludge, but one that inherited from an open architecture, CP/M,
> while the Mac on the contrary was a veering away from any openness, even the
> half-openness of the Apple II. After many years, it became clear that it is
> possible to build an open Mac from an open something else, but not from a
> closed Mac.
>

> PhR

dintersimone

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Richard Bayarri Bartual wrote:

> I liked Turbo-Prolog too - it even had relational database capabilities.
> While criticised by Prolog purists for its inability to assert rules as
> well as data, it was a true compiler that produced tight executables and
> had a superb development environment.

Turbo Prolog still lives as Visual Prolog at http://www.pdc.dk/vip/.
The guys from the Prolog Development Center built Turbo Prolog for
Borland, were working here in Scotts Valley and licensed the rights back
when we got out of the Prolog business.

--
David Intersimone "david i"
Director, Developer Relations
Inprise Corporation, Borland and VisiBroker products
See you at the 10th Annual Inprise Conference
Philadelphia, July 17-21, 1999

dintersimone

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Mark Richter wrote:

> I seem to recall at some gathering that a (then) Borland representative
> indicated that Turbo Basic came under fire due to some problems with
> licensing or other issues, particularly regarding an integrated
> debugger. Maybe these issues were close to the cause of Turbo Basic
> never being pursued any further. I found it to be a remarkable product
> with tremendous speed and capabilities.

Turbo Basic 1.0 was a success. It frightened Microsoft so much that they
quickly came out with QuickBasic
3.0 in response while they continued to work on Quick Basic 4.0 (the
edit and continue product). Visual Basic was no where in sight at that
point. There were never any licensing issues or integrated debugger
issues.

dintersimone

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
my visicalc version ran on the apple II. it did not require a Z80 card.

Philippe Ranger wrote:
>
> Leroy: >>IIRC, VisiCalc was an Apple II product, not CP/M. In fact, VisiCalc
> was
> *the* product that drove many businesses into buying Apple.
> <<
>

> I could check this, but I'm too lazy. I still think Visicalc ran on the Z-80
> card, therefore on CP/M. But the Apple II was just about the only thing with
> diskettes and enough of a market to make their particular format (among a
> score) a relative standard. So, to a business, it was the default Z-80
> installation.
>

> PhR

dintersimone

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
I love reading all this. it is very interesting to see what people
think from the outside and to understand how much mythology and
misinformation is out there. I did write a column about the history of
the borland name - you can find it at
http://www.borland.com/firehose/1998/04-29-98.html

Philippe Ranger wrote:
> I don't know when "Frank" Borland popped up, but the rumor I tend to
> subscribe to is that Borland was taken as an American-sounding name for a
> company, because of Frank Bormann, the astronaut.

the company name heritage is in the column. the name Frank Borland
first appeared (as far as I know) in the opening paragraph of the Turbo
Tutor version 1 manual. I have the text at home and will post it later
along with some great line art.

> Now, from all appearances, this stuff was bought, and seldom did the authors
> stick around for recasts. Borland was unable to follow up on any of its
> utilities, even SK, which remained stuck at version 1.53. Likewise, TP's
> very innovative "IDE" used an editor built on a binary-code machine Borland
> did not own the code for. That same machine was offered separately in an
> "Editor Toolbox" for TP3, and, paf! editors sprouted by the dozen
> (literally). Almost ALL Dos-based editors were born from this, unless they
> were ported from VMS, Unix, etc. But Borland could not improve in any way on
> the original binary editor (bined).

the above paragraph is littered with mistakes and untruths. Many of the
original developers of products stayed with the company and worked on
other products. yes, some of them eventually left (doesn't everyone
except me?). The sidekick editor appeared in Turbo Pascal 1 and on.
two editors were shipped with the Turbo Editor Toolbox - the binary
editor and a pascal source code version. Sidekick went on to be revised
many times and is still being worked on today at Starfish. There was
Sidekick 2, Sidekick 2 Plus, Sidekick for Windows, etc.

> Likewise, in the summer of 86, if memory serves, a summer intern was hired
> to "port" to TP to Basic, and Turbo Basic was born and launched. Again, it
> could not follow on itself, and eventually ended up resold to its author.

This is the funniest one of all. I don't know who that summer intern
was, must have been the one who fetched coffee and donuts for the real
developer of Turbo Basic - Bob Zale. Turbo Basic is still alive in
Bob's Power Basic. I think Philippe met Bob at a user group meeting in
Chicago and Bob joined Borland to work on Turbo Basic. I was running
language R&D at the time. We shipped TB 1.0 and 1.1. We licensed back
TB to Bob when we decided to get out of the Basic compiler market. you
can find power basic at http://www.powerbasic.com/.

> I believe the first Turbo C was also a renamed third-party product. But in
> this case Borland did manage to update it and, through several versions,
> make it "theirs". Borland C++ v. 1 appeared in summer 89 (if memory serves),
> and it was truly Borland. Also, again, earth-shaking.

The compiler for Turbo C came from the Wizard C compiler by Bob Jervis
who joined the company. The linker came from work being done at
Borland, the run-time library came from Wizard C and work done at
Borland. Future versions were all internal work by Borland employees.

> Though there was a long lull before TP4 (last days of 87), the fact is that
> at least TP was a purely-Borland enterprise. The author was Anders
> Hejlsberg. Anders more or less moved to other things, I understand, while
> Delphi 2 was a-building, and soon after was hired by Microsoft by paving his
> path with gold. At the same time (again, if memory serves) the Delphi
> business manager was similarly, er, seduced. But I'm getting ahead of
> myself.

Actually the first pascal compiler by Anders was called Poly Pascal for
the Z80. Borland acquired the rights to that compiler, added the
sidekick editor and run in memory technology to create Turbo Pascal
1.0. Who is this Delphi business manager that was seduced? are you
talking about Gary Whizin the R&D director - he retired from computers.

> Anyhow, Anders was little-known except among TP afficionados (the kind of
> people who hang around here), and Borland was identified with its president,
> Philippe Kahn, a Frenchman. The more Borland grew, the more Kahn became a
> public figure, and by the late 80's he was sounding a bit like the Divine
> Ellison. As long as Borland found NEW products to launch, in the brouhaha it
> went rather unnoticed that it couldn't follow-up on any of its past
> successes except TP and TC.

Anders is known to millions of developers around the world. there were
articles in european magazines and also Computer Language here in the US
with pictures of Anders, Chuck J and Gary.

> One exception. Rather early, perhaps 85, Borland bought a very promising
> database **maker**, Paradox, and simply left it to its own devices. But the
> product itself was never more than an also-ran on the market, compared to
> the remarkably inferior dBase, which willy-nilly made itself the standard
> from 85 onwards.

dBase and Paradox were the leaders in the database market. Paradox,
Quattro, Sidekick, Reflex, along with our developer tools all fueled the
growth of Borland. The database market changed dramatically when Access
1.0 was launched with a $99 price.

> So Borland bought the oldest WP app on Earth, which it relaunched as Sprint.
> It bought one spreadsheet maker (everyone was making better spreadsheet than
> Lotus, but Lotus was totally the standard), and let it disappear from view.
> It bought another one, Quattro (Spanish for "beyond 1-2-3), got to work on
> it (and kept at it), and quickly got a massive suit from friendly Lotus.
> That really hurt prospects. A bit later, it bought the best-known
> programmer's editor, Brief, and let that disappear too. Btw, at the time
> Excel was a perpetual money-loser for MS and Quattro wasn't really a threat.
> The war was with Lotus.

I don't ever remember buying a spreadsheet maker. we did find some
assembly language programmers in Hungary who had built a better than
1-2-3 spreadsheet that became Quattro. We did buy Surpass to add
spreadsheet technology and engineering to the Quattro Pro team. I think
I remember that Microsoft Excel was purchased from another company.


> In 92 BP7 (Borland Pascal 7) came out, and it was the second version for
> Windows. I may think the world of it, but the world thought otherwise, and
> it was a marketing dud. I truly don't understand why, but the common
> explanation is that by that time VB was gaining a lot of attention. Anyhow,
> at the time Borland desperately needed at least to hold on to its old
> revenue sources.

There are many sociological reasons for the decline of Pascal
compilers. The schools and universities were switching from Pascal to
Modula to Ada. The Educational Testing Service (ETS) switched the
advanced placement test from Pascal to C++. C/C++ had become the choice
for the majority of software engineers.


Don't let my message be a deterrent to all those who can add to the
collective memory of Borland history. I'm sure that my 13.5 years here
have caused my recollections to be fuzzy. I stand ready to be corrected
at all times.

ps: if anyone has a copy of the Turbo Tutor v1.0 floppy, and the Turbo
Prolog v1.1 system disk - can they send me a zip of the files? thanks.

dintersimone

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to

Philippe Ranger wrote:
>
> Glynn: >>... and you never mentioned Turbo Prolog. I think it lasted almost
> 1 whole year somewhere around 1988.
> <<
>
> Sorry. Lasted more than a year. I *think* the engine was licensed from some
> French developers, and that this was even mentioned when you ran it. An
> interesting and justified trial balloon, even if nothing came of it.

the developers are Danish. you'll find them at http://www.pdc.dk/vip/.

Borland shipped Turbo Prolog 1.0 and 1.1. we were finishing version 2.0
when we stopped in that market.

Leroy Casterline

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Mark Richter <mric...@emceesoftware.com> wrote:

>I believe dBase was a port over from the CP/M world onto the "new"
>PC-DOS. dBase II (at the time) was pretty much the only option, and was
>widely popular with those of CP/M heritage.

Yes, as I recall it was originally called Vulcan, one of the 1st
relational databases for PC's.

>Again, Lotus was a migration from the CP/M VisiCalc (I believe that was
>the name). Kapor had apparently written the earlier product, and owned
>the rights for the port.

IIRC, VisiCalc was an Apple II product, not CP/M. In fact, VisiCalc was
*the* product that drove many businesses into buying Apple. Also, I
don't think Kapor owned the rights to VisiCalc, but built 1-2-3 as an
improvement after leaving Personal Software(?), which it definitely was.
I think I still have the original pre-release dealer version of 1-2-3
somewhere. As I remember, the executable was just over 90K, an
incredibility large program for the times. I also think Kapor was the
1st to offer a keyboard template for the function keys on the PC
keyboard.

>Wasn't MS offering at the time a PC-DOS port of MultiPlan, or some such


>product? I thought Excel came in with the first Windows offering...

Yes, I'd forgotten about MultiPlan! Compared to Mitch Kapor's 1-2-3,
MultiPlan was terrible. In fact, compared to VisiCalc MultiPlan was
terrible...

This thread is waking a bunch of old memories...
--------------------------------------------------------------
- Leroy Casterline Cahill Casterline Ltd 970/484-2212 -
- Electronics/ASM/C/Delphi/CBuilder/Telephony/Instrumentation-
--------------------------------------------------------------

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Janet: >>

The company's original name was MIT (I forget now what that stood for). The
university was making legal noises and they renamed it Borland, taking the
name of another company that owed them money when it went bancrupt.
<<

Not the first time I've read this. Only, it didn't "catch". Thanks for the
reminder, I'll try to remember henceforth.

>>
My personal take on it was that Windows was the beginning of the end.
Borland bet on OS2 originally, and was late to start on the Windows
platform.
<<

That's interesting. Prior to the fall 1990, perhaps later, I think MS was
pushing everyone "serious" towards OS/2.

PhR

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Alan: >>So much for more dBASE disinformation :-)

dBASE was the database standard from late 1982 with its CPM release, DOS in
1983.
<<

Well, thank you, Alan, for the courtesy of calling my post disinformation,
and quoting the whole of it (in case someone doesn't have a newsreader), but
not pointing out the disinformation. :-) ;-)

I suppose the guilty part is actually this sentence about Paradox —

PhR: >>>


But the product itself was never more than an also-ran on the market,
compared to the remarkably inferior dBase, which willy-nilly made itself the
standard from 85 onwards.
<<<

This is disinformation? By the way, just checked Programmers At Work (MS
Press, 86) —

--------------
[C. Wayne Ratliff] in 1978 began writing the Vulcan program, which he
marketed by himself from 1979 to 1980. In late 1980 he entered into a
marketing agreement with Ashton-Tate and renamed the Vulcan product dBase
II.
--------------

>>
As to "inferior" to Paradox, well... that's a matter of opinion isn't it?
<<

I am not talking about all the improved non-clones — who possibly could
clone dBase II without improving on it? Foxbase, Clipper, etc. came by
because the bloody thing WAS the standard — that's the point of the indicted
sentence. As it happens, I worked more than I'd have wished to with
authentic, genuine dBase (never set up a system on my initiative, though,
just came in aftet idiots did). Hated every second line of it. Then I didn't
work with Paradox, unfortunately, just used it. What a revelation!

PhR

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Alan: >>1. As part of Phillippe's severance, he got two products, not one:

Sidekick and
Dashboard. Both were included in Corel's suite.
<<

Yeah, I remembered that later.

>>
3. I differ with much of your analysis as to what went wrong. What went
wrong,
to a great extent, was Microsoft - and Phillippe's inability to walk away
from a
head-to-head battle with Gates. If you remember, much of the resources of
the
early '90s went into the Borland/Word Perfect Suite (later the
Borland/Novell
suite), which was a direct competitor to the MS Office Suite. Obviously, it
lost
the battle and left Borland with an unfocused collection of products and no
office suite to sustain them.
<<

Well, we differ. I think the goose was cooked while the ink was still wet on
the check for Ashton-Tate, in 91. AND that this was just the last episode in
an ongoing Kahnian adventure that was masked by the financial success of
three products, and image success of some others (which, along with
Sidekick, couldn't get into version 2).

The made-up suites appeared to be the only way to market Borland's and WPC's
Windows versions once they were finally out. Many millions had been spent in
the effort, it seemed reasonable to not just open the drain and flush them
as too late. Especially for WPC, which would have thereby closed its doors.

>>
Contrary to what you indicate, and despite Borland's
problems in getting a clean compiler and Windows version out, dBASE produced
very significant revenue for Borland for a number of years.
<<

Nothing to pay for what A-T had cost. Enough perhaps to cover Borland's own
development costs. What DID I indicate?

PhR

Richard Grossman

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
"A.A.Katz (Alan)" wrote:

> What happened to
> dBASE and Paradox both was -Access-. When Borland bought dBASE, it sold for $895
> a copy with "Lan Packs" sold on a per-seat basis. Once Gates launched Access as
> a $99 stand-alone product and a member of the "Pro" suite (specifically targeted
> as a "Kahn Killer",by the way), the handwriting was on the wall for both Paradox
> and dBASE as "end-user desktop databases".

More illegal monopoly-behavior and bundling by Microsoft. Too bad the DOJ waited so
many years before breaking up Microsoft.


.....................
Richard Grossman
rgro...@techIII.com

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
Dave: >>

Many of the
original developers of products stayed with the company and worked on
other products. yes, some of them eventually left (doesn't everyone
except me?). The sidekick editor appeared in Turbo Pascal 1 and on.
two editors were shipped with the Turbo Editor Toolbox - the binary
editor and a pascal source code version. Sidekick went on to be revised
many times and is still being worked on today at Starfish. There was
Sidekick 2, Sidekick 2 Plus, Sidekick for Windows, etc.
<<

OK. Where was version 2 of Lightning, Reflex, SuperKey? The Bined? These
were successful products. Was there any shared code in the kernel of
Sidekick between versions 1 and 2? (From version 2, it was indeed a Borland
development, but the parentage from 1 was rather non-obvious.) Why was there
such a lapse between 1.53 and 2, that 2 was met with "Auld lang syne"?

>>
This is the funniest one of all. I don't know who that summer intern
was, must have been the one who fetched coffee and donuts for the real
developer of Turbo Basic - Bob Zale. Turbo Basic is still alive in
Bob's Power Basic. I think Philippe met Bob at a user group meeting in
Chicago and Bob joined Borland to work on Turbo Basic.
<<

The "summer intern" may have gotten tacked on somewhere, I'm working from
memory. I think it's Bob Zale who wrote that he was "hired to do TB".

>>
Who is this Delphi business manager that was seduced? are you
talking about Gary Whizin the R&D director - he retired from computers.
<<

No. Chinese name, I think. Anyhow, someone who else left for Redmond with
Anders, at a similarly high raider's price.

>>
The database market changed dramatically when Access 1.0 was launched with a
$99 price.
<<

This is where I'm totally fuzzy. I missed it at the time, thought Access was
a lightweight joke. Was I ever wrong! So, you're saying that right there, in
version 1, that thing was a winner? All I heard about was people staying
with Dos databases. How did it really look?

>>
I don't ever remember buying a spreadsheet maker. we did find some
assembly language programmers in Hungary who had built a better than
1-2-3 spreadsheet that became Quattro. We did buy Surpass to add
spreadsheet technology and engineering to the Quattro Pro team.
<<

I didn't remember Quattro as a Borland launch — otoh I've never known it as
anything but pure Borland. Yeah, memory tickled now. Didn't Philippe speak
of Vroom, if you remember that, as Quattro 1 technology? By the way, do you
know what happened to Lucid 3D?

>>
Don't let my message be a deterrent to all those who can add to the
collective memory of Borland history.
<<

Well, it sure began with a rumble like "this is my turf, don't tread on me".

PhR

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
David: >>my visicalc version ran on the apple II. it did not require a Z80
card.
<<

Well, then I stand corrected. But I had another reason — normally the Z80
came with 80-col support, and a spreadsheet on 40 columns... Was that
Visicalc on 40 columns, or was there a separate 80-col card on the machine?
Native-Apple Visicalc with 80 columns would make sense as a business
machine.

PhR

Philippe Ranger

unread,
Feb 8, 1999, 3:00:00 AM2/8/99
to
David: >>Turbo Basic 1.0 was a success. It frightened Microsoft so much that

they
quickly came out with QuickBasic 3.0 in response while they continued to
work on Quick Basic 4.0 (the
edit and continue product).
<<

And this prevented Borland from doing TB 2?

PhR

Andreas Prucha

unread,
Feb 9, 1999, 3:00:00 AM2/9/99
to
In article <36BDD692...@runesbike.com>, Rune Moberg <r...@runesbike.com> wrote:
>TPW 1 and BP7 were the first Borland packages I bought (I'm fairly new
>at this stuff, didn't start doing Pascal until '90 with TP5 at school).
>However, I seem to remember at that time that the trade press was all up
>in arms about "portability". It all had to be portability-this and
>portability-that. The solution offered? C/C++.

Well, I think this was a little bit earlier, esp in the text-mode time.

>One good thing did emerge from all this: Had BP7 become a bigger
>success, if everyone had dumped C/C++/VB in favour of BP, then Borland
>would probably never have bothered doing a Delphi.

Well, I think it would have not been a good idea to dump the C-compilers, nor
would it be a good idea to dump BCB now. Many people bought both, BP and BC,
and so do some Delphi user own BCB now.
IMO, the bad idea was to focus on the Office-stuff and to ignore BP. I think
Borland lost many BP users to MS because they did'nt longer trust Borland. Why
should anyone buy a C-compiler from a company he does not longer trust, if a
C-compiler is available from MS.

If Inprise neglected Delphi now, I think they would loose most users to MS or
an other vendor, but not to BCB.

>
>I guess all the technical acronyms finally got to us. There are dark
>holes in Delphi that I have never approached. Entera? VisiBroker? I even


>mistook MIDAS for something else entirely, and thought three-tier could
>be done without involving the complex MIDAS licensing stuff. (gee, why
>is everyone looking at me like I'm coming out of the closet or
>something?)

Well, I think it is a good idea to expand the product line to
enterprise-stuff, but it should not go at the cost of the mid-level stuff.

Andreas

William H. Mogk

unread,
Feb 9, 1999, 3:00:00 AM2/9/99
to
My version of VisiCalc was natively 40 col. I seem to recall that I got
a hold of a utility that would enable VisiCalc with the 80 Col card.
SuperCalc was a CP/M spreadsheet by Computer Associates(?) that
displayed in 80 Columns.

Bill

William H. Mogk

unread,
Feb 9, 1999, 3:00:00 AM2/9/99
to
The college library where I used to work occasionally gets rid of old
books. I was browsing through the stack one day, and I found (and
bought) the original Turbo Tutor book, complete with a Frank Borland bio
in the introduction. This great piece of Borland history sits beside my
original Turbo Pascal manual, and my Word Perfect 4.0 manual and
diskette set. Thinking back, I could run WP on one diskette, save
documents on the second diskette, and still do the majority of functions
that WP 8 does. This is progress?


Bill

jhawklyn

unread,
Feb 9, 1999, 3:00:00 AM2/9/99
to
On Mon, 08 Feb 1999 13:58:24 -0800, dintersimone
<dinter...@inprise.com> wrote:

>
>except me?). The sidekick editor appeared in Turbo Pascal 1 and on.
>two editors were shipped with the Turbo Editor Toolbox - the binary
>editor and a pascal source code version. Sidekick went on to be revised
>many times and is still being worked on today at Starfish. There was
>Sidekick 2, Sidekick 2 Plus, Sidekick for Windows, etc.

Don't forget Sidekick for the Mac, and Sidekick for OS/2 pm ver 1.1.

>
>ps: if anyone has a copy of the Turbo Tutor v1.0 floppy, and the Turbo
>Prolog v1.1 system disk - can they send me a zip of the files? thanks.
>

I think I do, but the Turbo Tutor is for CPM....

Anders Ohlsson

unread,
Feb 9, 1999, 3:00:00 AM2/9/99
to
> No. Chinese name, I think. Anyhow, someone who else left for Redmond with
> Anders, at a similarly high raider's price.

You must be referring to Paul Gross. He left around the same time.

--
Take care, and enjoy the ride,
Anders Ohlsson
Delphi Developer Support
borland.com

dintersimone

unread,
Feb 9, 1999, 3:00:00 AM2/9/99
to

we droppped several products about the same time including TB and Turbo
Prolog. there were many reasons - none of them having to do with
microsoft. The great thing is that Turbo Basic is alive and well at
PowerBasic. Bob Zale has done a fabulous job adding more capability
over the years.

dintersimone

unread,
Feb 9, 1999, 3:00:00 AM2/9/99
to
visicalc ran on 40 column apple II. if you had an 80-column card you
got more spreadsheet columns. if you had the shift key mod then you
could get upper/lower case characters too. wow, I can't believe I
remember any of this. I gave my apple II to my dad years ago. these
days I just use an AppleII simulator on my PC to run old games I still
have - karateka is great, so is choplifter, and Robot Wars is still
ahead of its time (except for maybe the Lego Mindstorms kit).

Philippe Ranger wrote:
>
> David: >>my visicalc version ran on the apple II. it did not require a Z80
> card.
> <<
>
> Well, then I stand corrected. But I had another reason — normally the Z80
> came with 80-col support, and a spreadsheet on 40 columns... Was that
> Visicalc on 40 columns, or was there a separate 80-col card on the machine?
> Native-Apple Visicalc with 80 columns would make sense as a business
> machine.
>
> PhR

--