Board index » cppbuilder » Re: It's Wednesday, and...

2003-11-07 04:40:17 AM
cppbuilder77
John B. Breckenridge wrote:
the future of CBX...
OTOH there have been sounds to the effect that managed C++ is an awkward
something, that is covered under the existing VS.Net anyway...
--
Kristofer
 

Remy Lebeau (TeamB) < XXXX@XXXXX.COM >wrote:
been around doing applications, Remy?
If something is promissed to allow me
to maintain existing projects, that
seems a pretty long-term promiss to me.
interfaces. What these are implemented
with, users shouldn't care. And these
interfaces aren't affected by MS at all.
--
XXXX@XXXXX.COM is never read
I'm Schobi at suespammers org
"And why should I know better by now/When I'm old enough not to?"
Beth Orton
 

Tamas Demjen wrote:
the shop wants used and still be comfortable within an IDE and de{*word*81}
that you are familiar/comfortable with.
Historically, my problem with Microsoft was never the compiler for my
apps, it was the IDE and de{*word*81}. Microsofts tools were not as
intuitive for me to use as Borlands have been.
-Michael
 

{smallsort}

Remy Lebeau (TeamB) wrote:
is going to happen. All this "that is the least they can do for us" doesn't
mean anything other than your hopes, and my hopes also, since Borland does
whatever they want. Yes, it would be nice to get an updated VCL with Windows
XP and any VCL bugs fixed, but seeing how Borland didn't fix any bugs
related to the VCL in BCB6, it seems doubtful they will do it for CBX. What
about the new C++ compiler, and the bugs fixed that are part of that. Will
that make it into CBX for compiling VCL code ?
What is really hard to understand: for two years Borland ignored nearly all
bugs and problems with BCB6.
OK, they were working on CBX, whose IDE is different and that should fix the
IDE bugs.
OK, they were working on an updated compiler for CBX which is much more Ansi
C++ compliant, which fixes compiler bugs.
OK they were updating their VCL for Delphi to fix VCL bugs.
All are commendable end results. But then release 1 of CBX is put out, at
ridiculous prices, and neither the updated compiler nor support for the VCL
is part of the release. Didn't Borland realize that their BCB customers
would be greatly disappointed in this new "version" ? What is currently CBX
1.0 can't possibly be the work of a programming staff over a period of two
years. It is way too crude.
What Borland needs to do, since CBX is it:
1) Put out the latest VCL for it, along with RAD VCL tools to do RAD
programming like we have in BCB.
2) Put out the Ansi C++ standard compiler with it with hooks to compile VCL
extensions.
3) Sell CBX 2.0 as a reasonable upgrade price for BCB customers ( $600 is
not reasonable ).
If these conditions are met in CBX 2.0, then BCB customers can switch to CBX
and plan on going the Managed C++ route for Windows and/or the wxWindows
route for native cross-platform programming, while maintaining their
investment in VCL components and programming.
If these conditions are not met, then I think there will be a large exodus
of C++ programmers away from Borland and to either VC++ .NET 2003 or
wxWindows/Qt/Whatever for cross-platform support.
I hope Borland does the right thing !
 

"David Ray" < XXXX@XXXXX.COM >wrote in message
The Delphi language itself *is* the VCL, so they have to maintain it
otherwise Delphi is dead. That is not the case with C++.
.NET platform. Its incorporated directly. And since the VCL is already
written in Delphi to begin with, migration is much easier on the Delphi side
than on the C++ side. 90+% of the existing Delphi VCL has already been
ported to the .NET platform with little change to the source code at all,
except at the core levels to use .NET primitives/functionalities now.
On the other hand, where C++ is concerned, the VCL is just an add-on
library, it has nothing to do with the C++ language itself at all. However,
Borland as always had to go through a whole bunch of issues to make a
Delphi-based library work under a C++ environment (and now, with the release
of CBX, under a Java environment as well). Compiler extensions, IDE
extensions, non-standard behaviors that don't even follow the rules of C++,
etc. C++ shouldn't need to be used that way.
Gambit
 

"David Erbas-White" < XXXX@XXXXX.COM >wrote in message
information they wanted to release, but management sat on it. That is
management's fault, not the developers.
Gambit
 

Remy Lebeau (TeamB) wrote:
David Erbas-White
 

number.
impression that these emails are being used as proof to higher ups that
VCL support is needed.
H^2
 

"Edward Diener" < XXXX@XXXXX.COM >wrote in message
which version they ship. For several versions, BCB and Delphi shared the
exact same VCL, and were kept in sync with each other. The fact that Delphi
moved on to v7 (and soon v8) and BCB did not make it past v6 is irrelevant.
The VCL is the VCL.
they have been telling people for several weeks now. They set up email
addresses specifically for this purpose.
front end, new back end. You can't compare the two.
will include the new compiler. Whether that will support compiling VCL
code, I do not know, it was not mentioned. However, if they do include the
VCL, then obviously they're going to have to be able to compile it as well,
don't you think?
in the making.
users. Period. It was not meant to be. That has been stated over and
over. They always intended it that way. That was meant for trying to draw
in the larger non-VCL C++ userbase - the other platforms, the dispelled
Microsoft users, etc. VCL users are targatted for CBX v2.0. They have been
that way for awhile now. Wait for CBX v2.0. You will not have to wait
long.
you do not see up front. So yes, I can image this kind of product taking
that long, there is an entirely new infrastructure that takes time to design
and implement. This is this just the beginning.
CBX, that is what gives Borland the competitive edge in all of its products.
VCL was never planned for v1.0. However, it was not ignored, either.
the back-end. If they put support for parsing extensions into their
front-end (which they would have had to in order to be 100% compliant), then
there is nothing preventing Borland from adding support for the extensions
into their backend.
Gambit
 

"John B. Breckenridge" < XXXX@XXXXX.COM >wrote in message
yet.
Gambit
 

"Remy Lebeau (TeamB)" < XXXX@XXXXX.COM >wrote in message
their plans for the future until you've read it.
(Sorry about that; it's just that the irony tickles me so!)
 

Remy Lebeau (TeamB) wrote:
note the latest bug I reported in vcl.components.writing for which Craig
Farrell gave me the solution when I repeated it in a different form in the
ide NG. If Borland has gotten this wrong in the past, they can mess up in
the future also unless they do a good job.
released. My suggestion was actually to update BCB6 with new VCL, compiler,
and IDE fixes, and not bother with supporting VCL in CBX. But as CBX is it
as the C++ IDE of the future for Borland, they need to do the first two
items under CBX.
code also as well as my correct C++ code.
compiler, with its bugs and all, to compile VCL code. They will give as a
reason that the new compiler is Ansi C++ standard and that adding support
for it for VCL-isms will break standardization. I can already see this
happening.
supporting VCL until .NET takes hold. I think that is still a few years down
the line even if I am aware of many programmers are using it now. Pulling
the plug from C++ VCL programming immediately is very unpleasant for
developers.
It is still disappointing that Borland sees wxWindows as a way to go for
cross-platform support. Even they must blanch when they look at wxWindows
code and realize that they had left all this behind ten years ago and now
they are returning to it, and embracing it ! It is remarkably primitive
compared to the VCL and a great step backwards. They would have done much
better adapting the RAD VCL environment to C++, using boost::function,
boost::bind, and boost::signals as their callback/event paradigm and doing
their own C++ properties using templates. Nearly all modern C++ compilers
support these paradigms ( check Boost web site ) and it would have been
easier to go from modern RAD ideas to modern C++ ideas than to go backward
on both by choosing wxWindows. But I am guessing that Borland doesn't have
the C++ talent anymore for such an undertaking.
 

 

"Harold Howe [TeamB]" < XXXX@XXXXX.COM >wrote in message
much into it. They did say "second release" more so then "v2.0"
specifically, although I'm pretty sure I heard "v2.0" said at least once for
sure. Either way, when I hear "release", I think more of an actual version
than just an upgrade/patch. The things that they want to include in CBX
that aren't implemented yet would certainly warrant an increase in version
numbers.
Gambit
 

using VCL?