Mention of C++ in a press release
Edward Diener
press release. I assume that once the next version of C++ Builder for
Windows and the new version of C++ Builder for Linux is released, there
will be no mention of C++ for another 2 years so I want to preserve this
moment for posterity.
With this momentous occasion are we actually to hope for a mention of
the new bug reporting system within the next 10 months ( making it 2
years since it was first announced ) or is that almost too much to be
hoped for ?
John Kaster (Borland)
> With this momentous occasion are we actually to hope for a mention of
> the new bug reporting system within the next 10 months
I dunno ... 10 months is surely a long time given its current status. We
might actually make it within that timeframe.
--
John Kaster, Borland Developer Relations, http://community.borland.com
$1280/$50K: Thanks to my donors!
http://homepages.borland.com/jkaster/tnt/thanks.html
Buy Kylix! http://www.borland.com/kylix * Got source?
http://codecentral.borland.com
The #1 Java IDE: http://www.borland.com/jbuilder
pnichols
"Edward Diener" <eddi...@tropicsoft.com> wrote in message
news:3C5618D4...@tropicsoft.com...
> I would like to congratulate Borland for actually mentioning C++ in a
> press release. I assume that once the next version of C++ Builder for
> Windows and the new version of C++ Builder for Linux is released, there
> will be no mention of C++ for another 2 years so I want to preserve this
> moment for posterity.
>
about BCB overall (Linux, Windows, and Symbian) this year (2002)..
Borland is practically going to own this space. I see it as becoming the
number two product (behind JBuilder) in 2002!!
Linux developers are, overall, C/C++ ones. Having a C/C++ IDE with the
quality of BCB is going to rule in Linuxworld space. I see similar things
for Symbian.
I do hope they continue the migration of BCB to support other non Intel
based systems as well.
Edward Diener
> "Edward Diener" <eddi...@tropicsoft.com> wrote in message
> news:3C5618D4...@tropicsoft.com...
>
>>I would like to congratulate Borland for actually mentioning C++ in a
>>press release. I assume that once the next version of C++ Builder for
>>Windows and the new version of C++ Builder for Linux is released, there
>>will be no mention of C++ for another 2 years so I want to preserve this
>>moment for posterity.
>>
>>
> I think you are going to be surprised about how much you are going to here
> about BCB overall (Linux, Windows, and Symbian) this year (2002)..
>
That will be a dramatic change from the last 5 years or so of BCB on
Windows, where the existence of BCB hardly even registered on Borland's
own radar screen of PR, and the effort to fix problems and provide first
class docs was limited at best. That's not to say I don't think it was
and is a good product, only that minimum effort was put into elevating
it into a great product. I am hoping, like many others, that the move to
Linux and other platforms will also cause Borland to focus on a better
Windows version.
Chris Uzdavinis (TeamB)
> That will be a dramatic change from the last 5 years or so of BCB on
> Windows, where the existence of BCB hardly even registered on
> Borland's own radar screen of PR, and the effort to fix problems and
> provide first class docs was limited at best. That's not to say I
> don't think it was and is a good product, only that minimum effort was
> put into elevating it into a great product. I am hoping, like many
> others, that the move to Linux and other platforms will also cause
> Borland to focus on a better Windows version.
Maybe Microsoft's push for C# is leaving C++ market open a little bit
wider, creating more demand for Borland's C++ compiler.
This is just a guess, though.
--
Chris(TeamB);
Edward Diener
story you told of that conference ), I have a feeling that MS will be
pushing their C++ implementation again also. Anything to prod Borland to
take C++ more seriously as a language implementation is alright with me.
Chris Uzdavinis (TeamB)
> With Lippman at the helm of VC++ ( I know you don't like him from the
> story you told of that conference ), I have a feeling that MS will be
> pushing their C++ implementation again also. Anything to prod Borland
> to take C++ more seriously as a language implementation is alright
> with me.
With Stan giving talks not on C++ but on C# at conferences, I do
sometimes wonder if he was hired more for his name than for real
improvement to VC++. I hope it's more than just public relations. MS
has a history of dangling people on a string just long enough to get
them hooked, then pull the rug out from under them. (Thereby forcing
them to move to MS's newest change-the-world neat-o technology.
OCX/ActiveX/COM/DCOM/COM+/Whatever)
Call be skeptical, but we'll see the truth in the end. (When is the end?)
--
Chris(TeamB);
Oscar Fuentes
> I think you are going to be surprised about how much you are going to here
> about BCB overall (Linux, Windows, and Symbian) this year (2002)..
This claim is based on some official statement by Borland or is just
gut feeling / wishful thinking?
> Borland is practically going to own this space. I see it as becoming the
> number two product (behind JBuilder) in 2002!!
Same question here.
> Linux developers are, overall, C/C++ ones. Having a C/C++ IDE with the
> quality of BCB is going to rule in Linuxworld space.
Linux developers are not IDE guys. They like using a good text editor
to write their code and MAKE to build their projects. Moreover, they
dislike vendor locking.
Possibly BCB will have it's niche on newbie programmers and small
quick & dirty applications.
> I see similar things for Symbian.
Symbian developers will certainly receive BCB as a godsend, altough
this a really small camp.
Will they port BCB or Delphi to Symbian? Knowing the actual developer
base, Delphi would be better. Symbian OS is a (simple) C++ API based
one, although this is not a big problem if Borland wants to run Delphi
on it. (I havn't checked actual releases, but ER5 does not use C++
exceptions nor RTTI).
> I do hope they continue the migration of BCB to support other non Intel
> based systems as well.
I hope they receive some income if they keep doing it.
--
Oscar
Chris Uzdavinis (TeamB)
> Linux developers are not IDE guys. They like using a good text editor
> to write their code and MAKE to build their projects. Moreover, they
> dislike vendor locking.
>
> Possibly BCB will have it's niche on newbie programmers and small
> quick & dirty applications.
I mostly agree, but IMHO the most important thing that Borland has
that Linux currently lacks is a great debugger. That's what I think will sell
Kylix/C++. The database access will be another benefit, but IMHO, the
GUI and WebServices and all that kind of stuff... I just don't see it
being used on linux the way it is on Windows. Windows people seem to
like gadget technology. Linux/Unix people tend to like less fluff and
more meat&potatos.
--
Chris(TeamB);
Oscar Fuentes
> Oscar Fuentes <o...@wanadoo.es> writes:
>
> > Linux developers are not IDE guys. They like using a good text editor
> > to write their code and MAKE to build their projects. Moreover, they
> > dislike vendor locking.
> >
> > Possibly BCB will have it's niche on newbie programmers and small
> > quick & dirty applications.
>
> I mostly agree, but IMHO the most important thing that Borland has
> that Linux currently lacks is a great debugger.
I almost mentioned this, but thought that most developers wouldn't do
BCC builds solely because the debugger.
GDB is not so bad, though. The only thing I like more on GDB than on
Borland is the key map. There is commercial debuggers, too. [I've
tried TotalView and it is not so user-friendly as Borland's one, much
less for C++ code]
BTW, a great thing of not having a good debugger is that you work
harder on creating correct code. If you add to this the long compile
times of gcc... :-)
> Linux/Unix people tend to like less fluff and more meat&potatos.
That's true for some Windows people as well ;-)
--
Oscar
Ed Mulroy (TeamB)
OLE, COM and especially DCOM solve very real problems that are not addressed
by languages such as C++ which purport to be platform independent and by
extension totally ignorant of cooperative display as one presentation from
multiple processes and by network distributed cooperative processes. They
may give us ulcers, but those things solve real problems that no C++ spec
will ever address.
. Ed
Oscar Fuentes
For some things you mention above, the C++ community had implemented
libraries based on vendor-independent standards. See CORBA, for
example, which has several implementation available for almost every
platform, including the free TAO, highly regarded as very good one.
IMHO, it's a good thing that the C++ spec does not mandate how you
must do inter-process communication, CORBA does it nicely,
thankyou. What the C++ spec must provide is a general way that allows
you to implement that functionality on top of the C++
language or standard library.
OTOH, if we pretend to extend the language to support whatever
tecnology each vendor creates for supporting special applications, we
would end having a kind-of Visual Basic. See MSVC v7 as an example of
this.
--
Oscar
Andrue Cope
Hmmm. Builder 5 has a /good/ debugger but I wouldn't say it was
/great/.
There are several issues that I'm hoping BCB6 will address but given
my inability to report bugs from this side of the Atlantic I have my
doubts.
Andrue Cope
[Bicester, UK]
Ed Mulroy (TeamB)
I was responding to this:
"...Thereby forcing them to move to MS's newest change-the-world neat-o
technology. OCX/ActiveX/COM/DCOM/COM+/Whatever..."
In this context, CORBA is just another "neat-o technology". It is not a C++
thing, but a system thing. My point was that it was really not fair to
characterize MS' adventures with multiple processes as exclusively a way to
lock in users. There were some creative ideas in OLE, COM and especially
DCOM that were not easily available to developers and end users before MS
introduced them. You speak of extending the language, but they are not C++
language extensions, no more than Visigenics stuff is.
. Ed
paul
> "pnichols" <paul@computer-logic> writes:
>
>
>> Linux developers are, overall, C/C++ ones. Having a C/C++ IDE with the
>> quality of BCB is going to rule in Linuxworld space.
>
> Linux developers are not IDE guys. They like using a good text editor
> to write their code and MAKE to build their projects. Moreover, they
> dislike vendor locking.
>
That would depend on what and where they are working.. If no one used IDEs
in the Linux space, then why Glade, KDevelop, Ajunta, etc? True, kernel
level developers might not use IDEs, albeit even the would love the
debugging capabilities. But those working on GUI based apps and Database
apps, will simply love it, IMHO. I know I will!!
Robert Kindred
are not very reliable, and the documentation is poor. Have you tried to
read the ATL book? Also, try to run the Adaptec Easy CD Creator after you
have played a game of Quake II. You have to reboot first. As an
alternative, I am looking forward to getting better at CORBA on linux.
Robert Kindred
"Ed Mulroy (TeamB)" <e...@mulroy.org> wrote in message
news:3c576428_1@dnews...
> Come on Chris, as they say "give the devil his due".
>
> OLE, COM and especially DCOM solve very real problems that are not
addressed
> by languages such as C++
<snip>
>
> . Ed
>
>
>
Edward Diener
Evidently when one gets paid big money, and I would guess that is the
case with Mr. Lippman as it is with Mr. Hejlsberg, it seems easy to
accede to some of the wishes ( and hype ) of one's employer. Having
never experienced that situation I wouldn't know personally, however,
and knowing the type of person I am I almost certainly will never find
out <g>. However, Stan Lippman seems to be a very independent
personality so I hope for the best.
Before being officially hired by MS, Mr. Lippman had written an article
in one of the programming magazines ( DDJ probably ) about "delegates"
in C# and I wrote him an e-mail about __closures in BCB as an already
existing C++ way of doing "delegates". His friendly response was that he
didn't approve of extensions to C++. I then wrote back that evidently he
felt the same way about __declspec extensions introduced by VC++ <g>. I
didn't get a further reply from that one <g><g>.
As you say, time will tell.
Chris Uzdavinis (TeamB)
> > Linux/Unix people tend to like less fluff and more meat&potatos.
>
> That's true for some Windows people as well ;-)
Sorry. :)
I just meant that the fact that such a market even exists...
--
Chris(TeamB);
Chris Uzdavinis (TeamB)
> Oscar,
>
> I was responding to this:
>
> "...Thereby forcing them to move to MS's newest change-the-world neat-o
> technology. OCX/ActiveX/COM/DCOM/COM+/Whatever..."
There is a renaming game that MS plays, and with each revision, a lot
of change that breaks the old code. This strums you along with MS at
their pace, not yours.
This is not saying that certain problems aren't solved by the above
acronyms. It's just the way that MS strings you along, and everybody
(figuratively speaking) jumps like Pavolv's dogs in response.
> In this context, CORBA is just another "neat-o technology". It is not a C++
> thing, but a system thing. My point was that it was really not fair to
> characterize MS' adventures with multiple processes as exclusively a way to
> lock in users.
It's not the single technology. It's how the MS technology is designed with
intentional obsolescence in a short period of time.
> There were some creative ideas in OLE, COM and especially DCOM that
> were not easily available to developers and end users before MS
> introduced them.
Of course, MS techonology isn't available to developers before they're
introduced.
> You speak of extending the language, but they are not C++ language
> extensions, no more than Visigenics stuff is.
You must not follow what MS has done to C++ to make it fit into their
new technologies. (And I'm not just talking about a few extensions
with double-leading-underscores like __closure and __property, which
is the ANSI C++ sanctioned way to add things to the languages.) I'm
talking about completely different syntax, like this:
For example, what the heck is [DllImport("shell32.dll")] doing there?
That's not even an approved way to extend the langauge. They do this
kind of ugliness all the time. (Code found at
http://www.codeproject.com/)
// Declare the function that is exported from unmanaged dll (shell32.dll).
[DllImport("shell32.dll")]
extern "C" int _cdecl ShellExecute(int hwnd,
String *strVerb,
String *strFileName,
String *strParameters,
String *strDirectory,
int nShowCmd);
--
Chris(TeamB);
Chris Uzdavinis (TeamB)
> Chris,
>
> Hmmm. Builder 5 has a /good/ debugger but I wouldn't say it was
> /great/.
I'm comparing it to what is currently availabe on linux.
> There are several issues that I'm hoping BCB6 will address but given
> my inability to report bugs from this side of the Atlantic I have my
> doubts.
I agree that things can improve. It seems there are always things we
find important that Borland doesn't...
--
Chris(TeamB);
Gerard Garreau
that C++ Builder version 6 is being delayed because of the perceived need
for a "Kylix" version. If C++ Builder had more recent patches, this may not
have been a problem, but as it is, it has bugs that has no hope of being
fixed and 6 have to wait for the linux and windows version to be in sync.
On the larger subject of C++ vs "all the other", I have the impression that
other langage developper (microsoft for one) do not seem to put as much
emphasis as they used to on C++. Borland seems to be on a Delphi/Java streak
and Ms is all over C#.
Gerard Garreau
"Edward Diener" <eddi...@tropicsoft.com> wrote in message
news:3C5618D4...@tropicsoft.com...
Andrue Cope
I don't think it's a serious problem. For me (and obviously for a lot
of people) BCB5 works and works well. If it can't do both then I'd
rather have Borland concentrate on earning revenue than fixing a few
bugs that (I assume) can generally be worked around.
Andrue Cope
[Bicester, UK]
Ed Mulroy (TeamB)
> (as was the case with Delphi 6), > that C++
> Builder version 6 is being delayed because
Using the word "impression" in that fashion is often an effective technique
to imply something while avoiding taking ownership of the statement if it is
later proved wrong. Another example might be "I had the impression that in
front of my house last night Elvis was putting on a show singing to
three-headed space aliens".
Please cite a source for your assertion couched in an "impression" that C++
Builder 6 is being delayed because of a 'perceived need for a "Kylix"
version'.
. Ed
Lauchlan M
technique
> to imply something while avoiding taking ownership of the statement if it
is
> later proved wrong. Another example might be "I had the impression that
in
> front of my house last night Elvis was putting on a show singing to
> three-headed space aliens".
>
> Please cite a source for your assertion couched in an "impression" that
C++
> Builder 6 is being delayed because of a 'perceived need for a "Kylix"
> version'.
I agree that 'impression' can be (and maybe is) used in that way, but given
that
(i) There was the two year delay for D6 as they developed Kylix
(ii) There is a two year delay for C++B 6
(iii) They have stated that they are developing a Kylix type CLX for C++B
don't you think there are reasonable grounds for this 'impression'?
It certainly seems that way to me.
Lauchlan M.
William Meyer
news:3c594e93$1_2@dnews...
> What annoy me in all this is the impression (as was the case with Delphi
6),
> that C++ Builder version 6 is being delayed because of the perceived need
> for a "Kylix" version.
Well, Borland must make its own decisions as to allocation of resources,
just as does any other commercial company. What annoys me in this newsgroup
is the persistent notion that Borland's plans should be founded on the
desires of a select group of users. While there is certainly a chicken and
egg situation here (in trying to determine whether the delay has caused loss
of share, or the loss of share is responsible for the delay), it's fairly
clear that the revenue streams seem to favor more attention to JBuilder and
Delphi/Kylix.
> If C++ Builder had more recent patches, this may not
> have been a problem, but as it is, it has bugs that has no hope of being
> fixed and 6 have to wait for the linux and windows version to be in sync.
> On the larger subject of C++ vs "all the other", I have the impression
that
> other langage developper (microsoft for one) do not seem to put as much
> emphasis as they used to on C++. Borland seems to be on a Delphi/Java
streak
> and Ms is all over C#.
Your summary of Borland's and Microsoft's apparent focus matches my own
impressions. As a Delphi user, as well as a BCB user, I have also been
annoyed at the long lag in release of new versions. OTOH, my shipping apps
are in D5, and there is no compelling reason to move them to D6, whereas
moving them at this time would provoke some considerable work.
In the final analysis, what I want most is for Borland to remain profitable
and to release well designed and nearly bug-free versions of their products.
If they cater to loud voices (rather than to revenues), then they may
jeopardize their continued existence, and we all lose.
Bill
Ed Mulroy (TeamB)
Kylix version of it. You specificly worded things to not refer to the
development of Kylix, the Delphi for Linux product. Your statement referred
to development of C++ Builder 6 being delayed for a Kylix version of
Builder. You used the word "impression" in a manner commonly used to
present an assertion when the author wishes to provide a semantic way to
avoid accountability should the assertion be contested.
I do not honor that semantic scheme for avoiding accountability. I
challenge you to provide fact, not fantasy, demonstrating C++ Builder 6
development and/or release is being delayed to provide a Kylix version. If
you cannot, then your "impression" is fantasy and I submit that this
newsgroup is the wrong venue for presenting your fantasies.
. Ed
David Erbas-White
It's entirely your prerogative not to honor his impression.
But quite frankly, since Borland refuses to make definitive statements
about their products and product development cycles, people must
'interpret' what they 'pick up', instead of being able to get correct
information.
And I have to admit, that based on statements made BY BORLAND PERSONNEL
over the past couple of years (no, I'm not going to search, but I and
others have seen them in this very newsgroup), the resources for BCB
were needed for development of other products.
For someone to state that they have this 'impression' (which in and of
itself means that they are expressing their opinion), and then you jump
up and doing saying 'prove it!' doesn't do your argument any good
whatsoever.
David Erbas-White
Vesa Vanhatupa
news:g04o7i...@wanadoo.es...
> "pnichols" <paul@computer-logic> writes:
>
> Symbian developers will certainly receive BCB as a godsend, altough
> this a really small camp.
Maybe not so small camp, there's 400,000 registered developers developing
software for the Nokia devices alone, like the new communicator which is
based on Epoc 6.0.
>
> Will they port BCB or Delphi to Symbian? Knowing the actual developer
> base, Delphi would be better. Symbian OS is a (simple) C++ API based
> one, although this is not a big problem if Borland wants to run Delphi
> on it. (I havn't checked actual releases, but ER5 does not use C++
> exceptions nor RTTI).
I think that the BCB/C++ would be much more interesting for Symbian/Epoc
development, because now the developing is done in C++ already. Also, the
old development environment was indeed Borland C++ for the 9110 communicator
series, which was based on Geos operating system.
-Vesa
Andrue Cope
[snip]
My sentiments exactly.
Andrue Cope
[Bicester, UK]
Ed Mulroy (TeamB)
newsgroups stated that C++ Builder 6 for Windows would be released and that
a Kylix version would follow. Shortly after that he posted his "impression"
which fully contradicts what was posted. Even were he operating with a
dearth of information it is inappropriate to interpert the lack of
information in the most critical fashion. Borland has always actively
avoided 'vaporware' and FUD announcements.
http://community.borland.com/article/0,1410,28281,00.html
. Ed
Gerard Garreau
English is NOT my first language and I am not in a position to have a
sematic debate over the use of the word "impressions". Sorry if I used it
in the wrong context . It has to be undestand as "I have STRONG Suspicion
that...". Of course, those statements were my opinion as I am not privy to
Borland internal decisions structure. However, Borland is a business and
make business decisions based on market predictions, customer base etc.. And
I think this is quite alright and I do not see anything wromg with it.
So I stand by statement:
I think Delphi 6 was delayed so that Kylix can be released first.
Likewise, C++ Builder 6 is (I think) delayed so that a Linux version can be
released at the same time or within a 3-4 months window.
Note that if you see a need for a Linux version of both a Delphi and C++
Builder, a delay of a 3-4 months is actually a fair price to pay for that. I
personnaly do not (see a need) and is therefore a bit miffed by it. Remember
that the last patch for C++ Builder was release 2 years ago and they are
still some outstanding bugs (see elesewhere in the *.cppbuilder.*
newsgroups).
BTW, at the last Borcon, I saw a presentation of C++ Builder 6 and I though
that it was probably 3-4 months of shipment as all the major new features
seems to be included (SOAP, DATASNAP, IDE improvements etc..).
What I'm afraid is that I think we are going to see more and more of those
delays. The next time, there maybe a .NET version as well.
Gerard Garreau
Alcan Inc.
"David Erbas-White" <der...@arachneering.com> wrote in message
news:3C5A07C9...@arachneering.com...
Gerard Garreau
company. I was merelly mentionning that those delays between release and
patches may even become even more pronouncede as I can see a delphi that
would have to run under windows, Linux and the .NET framework.
I also agree that delphi 5 was pretty stable and more or less bug free. But
that is not the case for C++. The last patch was years ago and there are
remaining bugs.
Gerard Garreau
Alcan Inc.
"William Meyer" <wmhm...@earthlink.net> wrote in message
news:3c59d56c$1_2@dnews...
Oscar Fuentes
> "Oscar Fuentes" <o...@wanadoo.es> wrote in message
> news:g04o7i...@wanadoo.es...
> > "pnichols" <paul@computer-logic> writes:
> >
> > Symbian developers will certainly receive BCB as a godsend, altough
> > this a really small camp.
>
> Maybe not so small camp, there's 400,000 registered developers developing
> software for the Nokia devices alone, like the new communicator which is
> based on Epoc 6.0.
This is plain wrong. Most if the "registered developers" are people
that ony wanted to download the emulator. I keep forgetting my
registering data each time I need to access the site. So far I've
registered about a dozen times. I know from testimonials of other
users that this is a common case.
Think about it, even Linux have less than 400,000 developers.
> > Will they port BCB or Delphi to Symbian? Knowing the actual developer
> > base, Delphi would be better. Symbian OS is a (simple) C++ API based
> > one, although this is not a big problem if Borland wants to run Delphi
> > on it. (I havn't checked actual releases, but ER5 does not use C++
> > exceptions nor RTTI).
>
> I think that the BCB/C++ would be much more interesting for Symbian/Epoc
> development, because now the developing is done in C++ already. Also, the
> old development environment was indeed Borland C++ for the 9110 communicator
> series, which was based on Geos operating system.
Well, the EpocR5 development tools uses MSVC for the emulator and gcc
for the real machine. I think the Quartz & Co. tools are the
same. Correct me if I'm wrong.
--
Oscar
Lauchlan M
> Hi,
> English is NOT my first language and I am not in a position to have a
> sematic debate over the use of the word "impressions". Sorry if I used it
Your use of English was correct.
Lauchlan M.
Lauchlan M
> You clearly stated that C++ Builder 6 is being delayed in order to develop
a
> Kylix version of it. You specificly worded things to not refer to the
> development of Kylix, the Delphi for Linux product. Your statement
referred
> to development of C++ Builder 6 being delayed for a Kylix version of
> Builder. You used the word "impression" in a manner commonly used to
> present an assertion when the author wishes to provide a semantic way to
> avoid accountability should the assertion be contested.
I did not do any of these things. You are confusing me with someone else in
this thread.
With regard to C++Builder being delayed for Kylix, John Kaster said so in so
many words, and also indicated that if they were doing it again (for Delphi
and C++B) they would do it differently.
I can't be bothered chasing up the reference. Google is your tool. Post by
John Kaster, in this newsgroup, probably between a month and two months ago.
Lauchlan M.
Ed Mulroy (TeamB)
not being delayed for the Linux version.
. Ed