Board index » cppbuilder » Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX


2003-10-12 09:26:42 AM
cppbuilder36
"Randall Parker" < XXXX@XXXXX.COM >wrote in
message news:3f88a269$ XXXX@XXXXX.COM ...
Quote
Yes, they had many years in which they failed to make
it possible to write VCL components in C++.
That is an inaccurate statement. It has always been possible to write
components in C++, I do it all the time, and distribute several of them
publically, in fact. The real issue is that C++-written components are not
usable in Delphi, but the opposite is supported. Those are two separate
topics.
Quote
The other big disappointment was when they introduced CLX
rather than making the bigger effort to clone VCL onto Linux. I
think it could bave been done even if they would have had to
introduce some restrictions.
At the time, the VCL was simply too tied to Win32 to port to Linux very
well. That is why they tried to make a new cross-platform library in the
first place.
Gambit
 
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Remy,
I know you can write VCL components for only BCB. But I'm talking about
the controls market. Delphi has a bigger market and by not allowing C++
controls to be used on Delphi it pretty much dictated that anyone who
wanted to write a mass market VCL control had to code in Pascal. That
reduced the total number of VCL controls that got developed.
VCL and Win32: It could have been done. It just would have been harder.
They took the easy way out and created something that was too
unappealing for the vast bulk of Win32 app writers.
Look, Win32 is over 90% of the client OS market. A portable RAD
framework for making apps should have been made to do best on Windows,
not on Linux. Borland's choice was dumb.
So think about where Borland manuevered themselves: Their VCL framework
can be coded only in a language that is a minor league player. C++ and
Basic are used by much larger numbers of people. Heck, Java is probably
used by more people. Their CLX framework is undesireable for the largest
OS market. If you want to argue that they didn't have sufficient
resources fine. But with enough resources they could have produced much
more compelling and successful products by allowing C++ coders to write
VCL for Delphi and by allowing VCL apps to be ported easily to Windows.
Remy Lebeau (TeamB) wrote:
Quote
"Randall Parker" < XXXX@XXXXX.COM >wrote in
message news:3f88a269$ XXXX@XXXXX.COM ...


>Yes, they had many years in which they failed to make
>it possible to write VCL components in C++.


That is an inaccurate statement. It has always been possible to write
components in C++, I do it all the time, and distribute several of them
publically, in fact. The real issue is that C++-written components are not
usable in Delphi, but the opposite is supported. Those are two separate
topics.


>The other big disappointment was when they introduced CLX
>rather than making the bigger effort to clone VCL onto Linux. I
>think it could bave been done even if they would have had to
>introduce some restrictions.


At the time, the VCL was simply too tied to Win32 to port to Linux very
well. That is why they tried to make a new cross-platform library in the
first place.


Gambit


 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Randall Parker wrote:
Quote
Remy,

I know you can write VCL components for only BCB. But I'm talking
about the controls market. Delphi has a bigger market and by not
allowing C++ controls to be used on Delphi it pretty much dictated
that anyone who wanted to write a mass market VCL control had to code
in Pascal. That reduced the total number of VCL controls that got
developed.
Totally agree, and I made the exact same point in a post about a month or so
ago.
Quote

VCL and Win32: It could have been done. It just would have been
harder. They took the easy way out and created something that was too
unappealing for the vast bulk of Win32 app writers.
Borland was too afraid of OP programmers going to C++ to allow VCL C++
components to be used by Delphi.
Quote

Look, Win32 is over 90% of the client OS market. A portable RAD
framework for making apps should have been made to do best on Windows,
not on Linux. Borland's choice was dumb.
Qt actually has a decent cross-platform solution but because Borland
erroneously wanted to support Delphi on Linux, they created a VCL monster on
top of Qt which was doomed. Notice though how Borland proved by doing so
that they could create OP to use C++, even though they wouldn't do that for
VCL C++ components on Windows. Nice, huh !
The Linux market is heavily C/C++ oriented, a fact which Borland has finally
woken up to in creating CBX. That's another reason why I don't think your
proposal will really work. I really tend to doubt that a native Delphi VCL
version for Linux will fly, although it certainly is a better idea than CLX.
Quote

So think about where Borland manuevered themselves: Their VCL
framework can be coded only in a language that is a minor league
player. C++ and Basic are used by much larger numbers of people.
Heck, Java is probably used by more people. Their CLX framework is
undesireable for the largest OS market. If you want to argue that
they didn't have sufficient resources fine. But with enough resources
they could have produced much more compelling and successful products
by allowing C++ coders to write VCL for Delphi and by allowing VCL
apps to be ported easily to Windows.
VCL does not need to be ported to Windows as it is native to that platform.
Did you mean something else ? But you are right that allowing complete
interoperability between C++Builder and Delphi on Windows would have been
wonderful not only for C++ component developers but even for Borland
themselves, despite their fears of losing Delphi programmers. It doesn't
take a genius to realize that Delphi developers coming to C++ would
naturally have used C++Builder because they already know the VCL.
 

{smallsort}

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

"Randall Parker" < XXXX@XXXXX.COM >wrote in
message news:3f88b258$ XXXX@XXXXX.COM ...
Quote
I know you can write VCL components for only BCB. But
I'm talking about the controls market. Delphi has a bigger
market
Partially because Delphi is a much larger bread-winer for Borland than BCB
ever was, so Delphi always got more focus (and of course Delphi pales in
comparison to JBuilder).
Quote
by not allowing C++ controls to be used on Delphi it pretty
much dictated that anyone who wanted to write a mass market
VCL control had to code in Pascal. That reduced the total
number of VCL controls that got developed.
That didn't stop the thousands of components that actually were developed.
Even though the components are mostly written in Pascal, that does not mean
that they were all written in Delphi. BCB includes a Pascal compiler, so
they could just as easily be developed under the BCB IDE as they are under
the Delphi IDE. So, because Pascal-written VCL components are usable in
both BCB and Delphi, and both can develop Pascal-written components, then
BCB users can write Pascal-based VCL components to share with other BCB
users. You're concerning yourself with semantics between using Pascal code
vs. C++ code, but the end effect is the same, particularly for consumers who
aren't writing components themselves so they don't care what language they
are actually written in just so long as they can be used for their needs.
Quote
VCL and Win32: It could have been done.
You probably meant "VCL and Linux" didn't you?
Quote
Their VCL framework can be coded only in a language that
is a minor league player.
Not as minor as you think.
Quote
C++ and Basic are used by much larger numbers of people.
I don't disagree that C++ is big, but lets not forget that it is not the
only language out there.
Quote
Heck, Java is probably used by more people.
I hope not!
Gambit
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Remy Lebeau (TeamB) wrote:
Quote
"Randall Parker" < XXXX@XXXXX.COM >wrote in
message news:3f88b258$ XXXX@XXXXX.COM ...

>by not allowing C++ controls to be used on Delphi it pretty
>much dictated that anyone who wanted to write a mass market
>VCL control had to code in Pascal. That reduced the total
>number of VCL controls that got developed.

That didn't stop the thousands of components that actually were
developed. Even though the components are mostly written in Pascal,
that does not mean that they were all written in Delphi. BCB
includes a Pascal compiler, so they could just as easily be developed
under the BCB IDE as they are under the Delphi IDE. So, because
Pascal-written VCL components are usable in both BCB and Delphi, and
both can develop Pascal-written components, then BCB users can write
Pascal-based VCL components to share with other BCB users. You're
concerning yourself with semantics between using Pascal code vs. C++
code, but the end effect is the same, particularly for consumers who
aren't writing components themselves so they don't care what language
they are actually written in just so long as they can be used for
their needs.
Thanks for telling me that as a C++ programmer I had the chance to write OP
components for Delphi. I am in despair over the news, and couldn't be sadder
that I did not give up component writing in C++ in order to do it in OP.
Could it be, however, that as a C++ programmer I really did not want to
write components in OP, a language in which I have minimal interest. No !
After all, what is the difference between one language and another. Nothing
really, right ?
The basic fact is that Borland did not want C++ programmers writing
components for the Delphi market. Rather than seeing this as a boon for
their product line, they saw it as a threat for converting their OP
programmers to C++. Thus their shortsightedness hurt not only C++ component
developers but their own company. As long as Borland does not take C++
seriously they will not succeed in grabbing a greater share of the C++
market worldwide, which is still the largest single computer language in the
world, simply because C++ is the best and richest programming language in
the world. Someday Borland may swallow their foolish pride, which tells them
that they know better than millions of C++ programmers, and they will create
a product worthy of the C++ language which they have treated so poorly over
the last number of years.
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

"Edward Diener" < XXXX@XXXXX.COM >wrote in message
Quote
Thanks for telling me that as a C++ programmer I had the
chance to write OP components for Delphi.
The source code, anyway. But you still need Delphi to create the packages
to contain the components. BCB cannot create Delphi packages any more than
Delphi can create BCB packages. Packages are IDE/compiler specific. That's
why, if you ever download components that support multiple IDEs, they always
include multiple projects for the packages.
Quote
After all, what is the difference between one language and
another. Nothing really, right ?
Let me put it to you this way - I can current read, and to an extent write,
almost a dozen different programming languages. How can I fit so many
languages into my noggin, you might ask? There is a simple truth behind
that - programming is programming. Many of the *concepts* behind performing
given tasks from one language to another are similar, if not the same. So
really, at least for me, it comes down to just learning the particular
syntaxes of each language to do the same things from one to another.
Quote
The basic fact is that Borland did not want C++ programmers
writing components for the Delphi market.
That is an unsubstantiated assumption at best, there is no factual truth
behind it. If you want to know why Borland didn't implement it, then you
need to ask Borland directly.
Gambit
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

I meant to say "ported easily to Linux".
Edward Diener wrote:
Quote
Randall Parker wrote:
But with enough resources
>they could have produced much more compelling and successful products
>by allowing C++ coders to write VCL for Delphi and by allowing VCL
>apps to be ported easily to Windows.


VCL does not need to be ported to Windows as it is native to that platform.
Did you mean something else ? But you are right that allowing complete
interoperability between C++Builder and Delphi on Windows would have been
wonderful not only for C++ component developers but even for Borland
themselves, despite their fears of losing Delphi programmers. It doesn't
take a genius to realize that Delphi developers coming to C++ would
naturally have used C++Builder because they already know the VCL.


 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Remy Lebeau (TeamB) wrote:
Quote
"Randall Parker" < XXXX@XXXXX.COM >wrote in
You're concerning yourself with semantics between using Pascal code
vs. C++ code, but the end effect is the same, particularly for consumers who
aren't writing components themselves so they don't care what language they
are actually written in just so long as they can be used for their needs.
Lots of programmers don't want to learn Pascal. They know enogh
languages as it is. You can rationalize why they ought to be willing.
But that won't change their minds.
Quote

>VCL and Win32: It could have been done.
You probably meant "VCL and Linux" didn't you?
Yes, sorry about htat.
Quote
>Their VCL framework can be coded only in a language that
>is a minor league player.
Not as minor as you think.
Yes, really. Outside of Delphi programmers who uses Pascal any more?
My guess is that all these languages have more programmers than Pascal:
- Java
- C/C++
- Basic
Other languages that might also have more programmers than Pascal:
- C#
- Cobol
- Perl
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

I've written in a lot of lanuages too. But this is besides the point.
Programmers have strong preferences about what languages they program
in. There are all sorts of reasons for these preferences. The net effect
in this case is that many potential VCL programmers aren't going to
bother because they don't want to program in Pascal.
For various things I do I have to program in Java, C++, Basic, and
occasionally Perl. I make enough mistakes just editing because it all
blurs together. I don't want to take up Pascal even though I used to
program in it 20 years ago. I'm hardly alone in this desire to minimize
the number of languages I program in.
Remy Lebeau (TeamB) wrote:
Quote
Let me put it to you this way - I can current read, and to an extent write,
almost a dozen different programming languages. How can I fit so many
languages into my noggin, you might ask? There is a simple truth behind
that - programming is programming. Many of the *concepts* behind performing
given tasks from one language to another are similar, if not the same. So
really, at least for me, it comes down to just learning the particular
syntaxes of each language to do the same things from one to another.

 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Quote
The basic fact is that Borland did not want C++
programmers writing components for the Delphi market.
I do not agree.
Many times, starting back with BCB's first release, I have asked Borland
people in person as well as via email about why BCB cannot create components
usable with Delphi. The answers vary with the person and the times but all
boiled down to a core issue of the C/C++ RTL not being available under
Delphi. Some of these discussions were private and very frank, and NOT ONE
of the replies referenced limiting migration of Delphi users to C++.
. Ed
Quote
Edward Diener wrote in message
news:3f88d8f9$ XXXX@XXXXX.COM ...
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

"Edward Diener" < XXXX@XXXXX.COM >wrote in message
Quote
The basic fact is that Borland did not want C++ programmers writing
components for the Delphi market. Rather than seeing this as a boon for
their product line, they saw it as a threat for converting their OP
programmers to C++. Thus their shortsightedness hurt not only C++
component
As long as the developer remained an Borland customer, what difference would
it make?
There is a far easier and elegant explanation: Delphi developers are as
passionate and loyal to their pascal syntax as you are about your's.
If you do not believe me, just post your statement (above) to the
Delphi.NonTechnical group for instant confirmation.
If they kill delphi, however, it is by no means a given that delphi
developers will switch to c++. In fact, now, the danger is that they will
jump to C#. It isn't an accident that Borland is offering C#.NET and
Delphi.NET.
-dennis
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Dennis Landi wrote:
Quote
"Edward Diener" < XXXX@XXXXX.COM >wrote in message
news:3f88d8f9$ XXXX@XXXXX.COM ...

>The basic fact is that Borland did not want C++ programmers writing
>components for the Delphi market. Rather than seeing this as a boon
>for their product line, they saw it as a threat for converting their
>OP programmers to C++. Thus their shortsightedness hurt not only C++
>component

As long as the developer remained an Borland customer, what
difference would it make?

There is a far easier and elegant explanation: Delphi developers are
as passionate and loyal to their pascal syntax as you are about
your's.
That is not an explanation why Borland did not want to allow C++ programmers
to write VCL components for Delphi, nor did I ever say that Delphi
developers are not passionate about their language. My belief is that
Borland didn't allow it because they ( Borland ) believed that they would
lose Delphi programmers to C++, and eventually to VC++.
As for OP programmers actually switching to C++, which language is more
popular in the world, has more programming jobs, and richer functionality ?
Not too hard to answer. Thus Borland's fears. Actually even if Delphi
programmers switched to C++, as I previously pointed out they would almost
certainly have stayed with Borland, at least in the short run, rather than
migrating to VC++, because they know the VCL.
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Ed Mulroy [TeamB] wrote:
Quote
>The basic fact is that Borland did not want C++
>programmers writing components for the Delphi market.

I do not agree.

Many times, starting back with BCB's first release, I have asked
Borland people in person as well as via email about why BCB cannot
create components usable with Delphi. The answers vary with the
person and the times but all boiled down to a core issue of the C/C++
RTL not being available under Delphi. Some of these discussions were
private and very frank, and NOT ONE of the replies referenced
limiting migration of Delphi users to C++.
Do you really think, Ed, that if Borland's reason for not allowing VCL
C++Builder components to be used by Delphi were as I surmise, the people to
whom you talked would have admitted that ?
As I have mentioned before, why could not the C/C++ RTL libraries have been
distributed as part of Delphi ? It's just another LIB/DLL, and all Windows
programs, whether in C++, OP, or whatever, use underlying LIB/DLLs. I
totally believe you in what you say as to what you were told of course, but
the explanation from the people with whom you talked does not appear logical
to me.
I am not saying that having total interoperability between C++ and OP at the
VCL level would not have been difficult to do, and that Borland may have
decided that the difficulty was not worth it, but I continue to believe that
surmounting that difficulty was predicated on the belief that total
interoperability would have lost Delphi programmers to C++ and the VC++
programming market, and that Borland's fears in that area kept them from
doing something that would have greatly benefited both themselves and their
programmers.
I know it would have benefited C++Builder component developers, giving them
a much bigger market for their development efforts. It almost certainly
would have benefited Delphi programmers, giving them many more components
from which to build their modules and applications. Despite Borland's qualms
it would have benefited themselves also by providing a much bigger
programming market for their product, something which could have competed
much more successfully against VC++ and VB.
But Borland lacked the vision for it, pure and simple. Their fear of
competing with MS in MS's own domain on Windows, C++, kept them from doing
something which was logical and beneficial to everyone involved.
I know people will tell me that my opinion on this matter is only a guess
ands that I can't possibly know what Borland's real motives are or were.
They are right. But that will not keep me from believing as I do on this
matter until I hear a much better explanation otherwise than what Borland
personnel told you.
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Remy Lebeau (TeamB) wrote:
Quote
"Edward Diener" < XXXX@XXXXX.COM >wrote in message
news:3f88d8f9$ XXXX@XXXXX.COM ...

>The basic fact is that Borland did not want C++ programmers
>writing components for the Delphi market.

That is an unsubstantiated assumption at best, there is no factual
truth behind it. If you want to know why Borland didn't implement
it, then you need to ask Borland directly.
I will modify my statement. "My belief is that Borland did not want C++
programmers writing components for the Delphi market." I claim no factual
truth behind it other than the fact that Borland did not create the ability
for Delphi programmers to use VCL components written in C++. It is my
assumption and I can't possibly know Borland's real reason. However I don't
believe that I would get the real reason from Borland even if I personally
knew someone there. I am all too aware that company policies often override
personal truth from my own experience as an employee and consultant for US
corporations, and I wouldn't expect Borland to be any different. To me that
is a normal fact of US corporate life and not a put down of Borland or any
other US corporation.
 

Re:Re: Modest Proposal For Borland Regards Delphi, BCB, BCBX

Are these $4 Australian dollars or US dollars? Please elaborate.
Rodolfo
"Ed Mulroy [TeamB]" < XXXX@XXXXX.COM >wrote in message
Quote
>I'm curious about what you would qualify
>as "challenging proposals".

Solving the Middle East crises in 3 days with $4, a bottle of aspirin and
a
metal coat hanger?