Board index » delphi » VCL's Future

VCL's Future


2004-06-17 01:26:32 AM
delphi237
My only interest is in applications compiled to machine language, not
.Net's CLR bytecodes, so all my questions refer to the "regular" VCL
used in compiling to machine language:
Does anyone think the VCL will ever be updated to produce applications
that directly use the .Net API? Or unicode? Or will Borland let it
wither on the vine? If these improvements were to take place about 10
years from now then that would be ok, I can live with that. But if the
.NET API transition *never* takes place then my Win32 Delphi apps will
all be running on an emulation layer and some users may not even have
the Win32 emulation layer anymore if MS makes the installation of it
optional. Logic would say that the VCL would eventually (let's say 10
years from now) be upgraded, but this is a Borland decision and I have
not been too happy with some of their decisions lately. Delphi is such
an awesomely productive tool that it pains me when its future is not
clear. Borland is not displaying a roadmap for its "regular" VCL, so I
do not expect anybody to really know its future, but can some of the
professionals weigh in with their opinion and say whether or not they
think any updates will be made? I guess I am looking for someone to
cheer me up.
 
 

Re:VCL's Future

I second that. The lack of Unicode controls is especially troubling - given
that .Net version of Delphi only runs on Windows 2000 and up, doesn't it
make sense to finally say goodbye to Win 95/98/ME for the Win32 version of
Delphi as well?
Oh, and if Borland could also fix the TLB editor problems, that would be
even better <g>
With all due respect to the Borland developers, I truly believe this is a
case of lunatics running the asylum: Win32 is not {*word*226} and fun anymore, so
it doesn't get the attention it deserves.
Dmitry Streblechenko (MVP)
www.dimastr.com/
OutlookSpy - Outlook, CDO
and MAPI Developer Tool
"Mike Vance" <XXXX@XXXXX.COM>writes
Quote
My only interest is in applications compiled to machine language, not
.Net's CLR bytecodes, so all my questions refer to the "regular" VCL
used in compiling to machine language:

Does anyone think the VCL will ever be updated to produce applications
that directly use the .Net API? Or unicode? Or will Borland let it
wither on the vine? If these improvements were to take place about 10
years from now then that would be ok, I can live with that. But if the
.NET API transition *never* takes place then my Win32 Delphi apps will
all be running on an emulation layer and some users may not even have
the Win32 emulation layer anymore if MS makes the installation of it
optional. Logic would say that the VCL would eventually (let's say 10
years from now) be upgraded, but this is a Borland decision and I have
not been too happy with some of their decisions lately. Delphi is such
an awesomely productive tool that it pains me when its future is not
clear. Borland is not displaying a roadmap for its "regular" VCL, so I
do not expect anybody to really know its future, but can some of the
professionals weigh in with their opinion and say whether or not they
think any updates will be made? I guess I am looking for someone to
cheer me up.
 

Re:VCL's Future

Dmitry Streblechenko writes:
Quote
I second that. The lack of Unicode controls is especially troubling -
given that .Net version of Delphi only runs on Windows 2000 and up,
doesn't it make sense to finally say goodbye to Win 95/98/ME for the
Win32 version of Delphi as well?
No. A lot of people are still running Win98. MS tried to desupport
Windows 98, and decided not to after the ensuing hue and cry. I think
that if Borland went with a Unicode only version of Delphi32, a lot of
people wouldn't upgrade, because they couldn't distribute the
applications that it created to a significant percentage of their user
base.
--
Mike Swaim XXXX@XXXXX.COM at home | Quote: "Boingie"^4 Y,W & D
MD Anderson Dept. of Biostatistics & Applied Mathematics
XXXX@XXXXX.COM or XXXX@XXXXX.COM at work
ICBM: 29.763N -95.363W|Disclaimer: Yeah, like I speak for MD Anderson.
 

Re:VCL's Future

Well, neither MS nor Borland support Windows 3.1. You have to stop
supporting old OSes at some point, there is no way around it, so it is just
a question of when.
If you want to develop for Win 9x, you can use any of the previous versions
of Delphi. Remember that 16 bit Delphi 1 was included with the new 32 bit
Delphi 2? Why not do the same?
Again, at some point you need to weigh the cost of dropping Win 9x support
vs. the benefit of adding new Win 2000 specific features. For me that point
is now.
And just for the record, you *can* have Unicode controls on Win 9x.
Dmitry Streblechenko (MVP)
www.dimastr.com/
OutlookSpy - Outlook, CDO
and MAPI Developer Tool
"Mike Swaim" <XXXX@XXXXX.COM>writes
Quote
Dmitry Streblechenko writes:

>I second that. The lack of Unicode controls is especially troubling -
>given that .Net version of Delphi only runs on Windows 2000 and up,
>doesn't it make sense to finally say goodbye to Win 95/98/ME for the
>Win32 version of Delphi as well?

No. A lot of people are still running Win98. MS tried to desupport
Windows 98, and decided not to after the ensuing hue and cry. I think
that if Borland went with a Unicode only version of Delphi32, a lot of
people wouldn't upgrade, because they couldn't distribute the
applications that it created to a significant percentage of their user
base.

--
Mike Swaim XXXX@XXXXX.COM at home | Quote: "Boingie"^4 Y,W & D
MD Anderson Dept. of Biostatistics & Applied Mathematics
XXXX@XXXXX.COM or XXXX@XXXXX.COM at work
ICBM: 29.763N -95.363W|Disclaimer: Yeah, like I speak for MD Anderson.
 

Re:VCL's Future

Mike Vance writes:
Quote
Does anyone think the VCL will ever be updated to produce applications
that directly use the .Net API? Or unicode?
Do you know about the TNT controls for Unicode? They work very nicely
and they're free and come with source code. So... don't worry too much
about Borland not providing a Unicode VCL.
Cheers,
Kevin.
 

Re:VCL's Future

Quote
benefit of adding new Win 2000 specific features
unicode is SO benefit! - to whom?
 

Re:VCL's Future

On Wed, 16 Jun 2004 10:26:32 -0700, Mike Vance writes:
Quote
My only interest is in applications compiled to machine language, not
.Net's CLR bytecodes, ...
Does anyone think the VCL will ever be updated to produce applications
that directly use the .Net API?
Isn't this a contradiction? You want apps to work with .NET but not be
.NET?
--
Marc Rohloff
marc rohloff at myrealbox dot com
 

Re:VCL's Future

Andrew Rybenkov writes:
Quote
>benefit of adding new Win 2000 specific features

unicode is SO benefit! - to whom?
To me. I have a database that displays text in multiple languages within
the same application. It needs to run on all 32 bit versions of Windows,
even if it only displays all the character sets on Windows 2000 and above. I
love the TNT controls, but it would be simpler if this was part of the VCL.
--Rob McDonell
 

Re:VCL's Future

Eh? To you as a developer who can sell the software outside of North and
South America and Europe?
I don't know about you, but I have no problem with currencies other than US
dollars or euros...
Dmitry Streblechenko (MVP)
www.dimastr.com/
OutlookSpy - Outlook, CDO
and MAPI Developer Tool
"Andrew Rybenkov" <XXXX@XXXXX.COM>writes
Quote
>benefit of adding new Win 2000 specific features

unicode is SO benefit! - to whom?

--
Andrew Rybenkov.



 

Re:VCL's Future

Quote
You want apps to work with .NET but not be .NET?
I want apps that use the DotNet API without the bytecode disadvantages
and forced garbage collection. Perhaps my wording was not so correct,
but I think I wrote it down well enough to get that through.
 

Re:VCL's Future

Quote
Isn't this a contradiction? You want apps to work with .NET but not be
.NET?
He wants what the new C++ compiler in Whidbey will do.
Z
 

Re:VCL's Future

"Mike Vance" <XXXX@XXXXX.COM>writes
Quote
>You want apps to work with .NET but not be .NET?

I want apps that use the DotNet API without the bytecode disadvantages
and forced garbage collection. Perhaps my wording was not so correct,
but I think I wrote it down well enough to get that through.
Sounds like you're asking for a native compiler that will understand the
.NET Framework and generate 32-bit or 64-bit machine code for it.
Just like Borland uses the VCL and generates 32-bit machine code.
Fine. Lets say such a compiler exists. Delphi, C# whatever.
At the end of the day, the unmanaged apps produced will fly in the face
of the biggest selling point of moving to Longhorn and .NET, and that is
building and running pure "managed" applications.
Microsoft will spend a billion dollars over the next few years telling, no,
more like, convincing the world we need to upgrade to Longhorn.
How exactly will this multi-billion dollar company will do this? Will they
sell the world on: new features? faster processing? cooler looking UI?
more robust? mmm, getting warmer. More secure??? Bingo.
Toss in some spin offs with "no spam" and "no viruses" and the
marketing plan for Longhorn pretty much writes itself.
How Longhorn will become "more secure" is by running managed apps.
Not by running unmanaged, native apps. In fact, your unmanaged apps
will trigger a slew of warning messages by Longhorn -- alerting the user
that your app cannot be trusted. Compare that to your competitor who
built a 100% managed app. Users will install the managed app over
yours because it is a safer choice. How many friends do you know
that have had their computers taken over by spyware and have called
you to help fix their system? Its crazy.
Now do you see why Microsoft is re-writing the entire Office Suite
in managed code?
It's such a simple vision but so many people can not see it.
Back to your point of wanting this .NET compiler. I suggest to reconsider.
Forget optimization. That was 1990's selling point. Security will be the
key selling point from now on. And building managed apps using the
.NET framework is the one way ticket to this new frontier.
--
Anthony Carrabino
www.vistadb.net
 

Re:VCL's Future

Quote
It's such a simple vision but so many people can not see it.
And somehow you got it wrong, this was what they told 2 years ago,
and now things have changed (they always do, this is IT).
Look at what got enhanced and promoted Whidbey: C++ and native
compilation are back at "the top of the food chain". Shocker? No.
The reason? Security is overrated, and XP SP2 should get them there.
More importantly, you don't _need_ Microsoft products to achieve
security, they know it, you know it, they only pushed this issue
to shake off their image of the most insecure OS on earth,
but security isn't what sells, applications are what sell,
the mere fact they thrived and still thrive on insecure software
is proof enough.
Security _was_ the next big thing, we're almost there, time to move on.
I'd rather have them focus on eliminating spam now, which is much more
annoying than the odd worm alert that never materializes if your software
is properly setup <g>
Slightly related article for you
www.joelonsoftware.com/articles/APIWar.html
Z
 

Re:VCL's Future

Quote
To me. I have a database that displays text in multiple languages within
the same application.
If it only displays (not prints), than one of solutions - do not use True Types, but
rather draw text yourself keeping char glyphs in some "imagelists". Then it will work on
any OSes, and you can forget about that monstrous ArialUnicode.
 

Re:VCL's Future

Quote
Eh? To you as a developer who can sell the software outside of North and
South America and Europe?
You do not need Unicode for it - just successful software. As byside question -
how big that other market? Do you think all the China will start cash out for your software,
or you will penetrate rather closed Japanese market?
Quote
I don't know about you, but I have no problem with currencies other than US
dollars or euros...
I have not problem with other currencies too - I just don't need them.