| View previous topic :: View next topic |
| Author |
Message |
Randall Parker
Guest
|
 Posted: Fri May 28, 2004 6:22 am Post subject: Next BCB version: time to market and quality important |
 |
|
I am guessing that the next version of BCB/BCBX could be based on:
1) The Delphi v7 IDE.
Basically make BCB v9 really be something similar to what BCB v7 would have been.
2) The Delphi v8 IDE.
I don't know if this really is feasible. Are there basic IDE improvements in it
that make it a better starting point than the v7 source code base without being any
harder to fit in the BCB C++ specific code to it?
2) The Delphi v9 IDE.
But of course it is not done yet.
3) The JBX IDE.
Basically come out with another rev of BCBX with a lot of IDE fixes but with a
VCL designer and other Win32 support such as Active X wizards and CodeGuard.
My take on these three options is that most of all I want fixes. I don't like the
project configuration limitations of BCB v6 and in terms of features would really
like to see something done about that. But I really just want a large number of bugs
fixed. The more fixes the better.
If going with one of these options requires so much work to make it capable of
matching everything BCB v6 can already do without making much headway of bugs then
I'm unenthusiastic about that choice.
Also, if basing BCB v6 on the next Delphi v9 IDE requires waiting until after Delphi
v9 is released (which is 4Q 2004 or perhap some time in 2005) then I'm unenthusiastic
about that choice too.
I want the next BCB to be totally compatible with Delphi 3rd party libraries. I want
the linker to be way better. I don't want IDE crashes that are specific to parts of
the IDE that C++ developers use.
I guess what I'm getting at are two points:
1) The longer it takes to get the next version of BCB released the less valuable it
will be. A tool we don't have is a tool that is not doing us any good.
BCBX enhanced to have a Win32 VCL designer and various wizards that are specific
to Win32 (Active X stuff for instance) may take too long and the first release may
have too many bugs. BCBX is already like a rough draft of an IDE without even trying
to support Win32 stuff.
2) If the next version of BCB has some new features but is just as buggy as BCB v6
then I don't think we will have gained all that much. If it is more buggy it will not
improve productivity and the result will be the same thing seen with v5 users not
upgrading to v6 due to regression bugs and bugs not fixed.
The next BCB/BCBX ought to he higher in quality than BCB v6 sp4. By many reports
BCBX is currently lower in quality.
It is the compiler internal error messages, the times when I can't build because
some dfm file is messed up, or when I have to revert back to a previous version of
the project file because it has gotten too messed up to build that I most want BCB to
be better.
There is a lot to be said for just fixing some of the bugs in BCB v6.
|
|
| Back to top |
|
 |
Gord
Guest
|
| Posted: Fri May 28, 2004 1:48 pm Post subject: Re: Next BCB version: time to market and quality important |
|
|
Randall I think your post sums up the situation very well for most of the current BCB
users AFAICT. Robert E's post a couple days ago has started to really scare me. We are
hoping for an open letter so that Borland can let us see what their road map is for C++.
Robert's post made it clear that they don't know what their road map is right now. If
they try to take on a huge new project (eg. merge VCL with CBX) we might not see anything
until 2005 which is unacceptable.
They need to fix BCB ASAP.
Randall Parker wrote:
| Quote: | I am guessing that the next version of BCB/BCBX could be based on:
1) The Delphi v7 IDE.
Basically make BCB v9 really be something similar to what BCB v7 would have been.
2) The Delphi v8 IDE.
I don't know if this really is feasible. Are there basic IDE improvements in it
that make it a better starting point than the v7 source code base without being any
harder to fit in the BCB C++ specific code to it?
2) The Delphi v9 IDE.
But of course it is not done yet.
3) The JBX IDE.
Basically come out with another rev of BCBX with a lot of IDE fixes but with a
VCL designer and other Win32 support such as Active X wizards and CodeGuard.
My take on these three options is that most of all I want fixes. I don't like the
project configuration limitations of BCB v6 and in terms of features would really
like to see something done about that. But I really just want a large number of bugs
fixed. The more fixes the better.
If going with one of these options requires so much work to make it capable of
matching everything BCB v6 can already do without making much headway of bugs then
I'm unenthusiastic about that choice.
Also, if basing BCB v6 on the next Delphi v9 IDE requires waiting until after Delphi
v9 is released (which is 4Q 2004 or perhap some time in 2005) then I'm unenthusiastic
about that choice too.
I want the next BCB to be totally compatible with Delphi 3rd party libraries. I want
the linker to be way better. I don't want IDE crashes that are specific to parts of
the IDE that C++ developers use.
I guess what I'm getting at are two points:
1) The longer it takes to get the next version of BCB released the less valuable it
will be. A tool we don't have is a tool that is not doing us any good.
BCBX enhanced to have a Win32 VCL designer and various wizards that are specific
to Win32 (Active X stuff for instance) may take too long and the first release may
have too many bugs. BCBX is already like a rough draft of an IDE without even trying
to support Win32 stuff.
2) If the next version of BCB has some new features but is just as buggy as BCB v6
then I don't think we will have gained all that much. If it is more buggy it will not
improve productivity and the result will be the same thing seen with v5 users not
upgrading to v6 due to regression bugs and bugs not fixed.
The next BCB/BCBX ought to he higher in quality than BCB v6 sp4. By many reports
BCBX is currently lower in quality.
It is the compiler internal error messages, the times when I can't build because
some dfm file is messed up, or when I have to revert back to a previous version of
the project file because it has gotten too messed up to build that I most want BCB to
be better.
There is a lot to be said for just fixing some of the bugs in BCB v6.
|
|
|
| Back to top |
|
|
Arnie Mauer
Guest
|
| Posted: Fri May 28, 2004 3:48 pm Post subject: Re: Next BCB version: time to market and quality important |
|
|
"Randall Parker" <STOPtechiepundit (AT) EVILfuturePOXpunditSPAM (DOT) com> wrote
in message news:40b6daed$1 (AT) newsgroups (DOT) borland.com...
| Quote: | There is a lot to be said for just fixing some of the bugs in BCB
v6. |
I agree. I don't see how Borland can keep releasing new versions
without having fixed known bugs in past versions.
I was the primary advocate for using BCB when I joined the company
seven years ago. It was without a doubt much better than MSVC++ for
forms layout and implementation. Maybe it still is. MFC was a thin
encapsulation with little or no added value. However, now I'm getting
some dirty looks from coworkers.
We're still using BCB 5 - there was no motivation to switch to 6.
Probably won't be for 9 either. So now we're between a rock and a
hard place. Our product has over 200 EXE's and over 300 forms. Many
of the EXE's are console apps. I'm starting to look at converting the
console apps to 64 bit later this year or early next year. All are DB
related. All I have to do is get rid of those pesky TADO components,
AnsiString and TDateTime.
Sorry for rambling,
- Arnie
|
|
| Back to top |
|
|
Remy Lebeau (TeamB)
Guest
|
| Posted: Fri May 28, 2004 6:14 pm Post subject: Re: Next BCB version: time to market and quality important |
|
|
"Randall Parker" <STOPtechiepundit (AT) EVILfuturePOXpunditSPAM (DOT) com> wrote in
message news:40b6daed$1 (AT) newsgroups (DOT) borland.com...
| Quote: | 1) The Delphi v7 IDE.
Basically make BCB v9 really be something similar to what BCB v7 would
have been. |
It has already need stated that BCB9 whould be based on Delphi 9, not 7.
| Quote: | 2) The Delphi v9 IDE.
But of course it is not done yet.
|
Its closer than you might think...
That is what CBX already does.
| Quote: | I want the next BCB to be totally compatible with
Delphi 3rd party libraries.
|
Hense why it has been stated that BCB9 will be the same Win32 VCL that BCB
has always been.
Gambit
|
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
