Board index » cppbuilder » Re: Live chat on C++Builder

2005-02-02 04:31:38 AM
cppbuilder44
Duane Hebert wrote:
There will be some work on compliance, but we are not looking at 100%
in this release.
See my running comments from the 1st half in the next thread over ;?
AlisdairM(TeamB)
 

Duane Hebert wrote:
more or less the same answer -- when it's ready / we'll tell you later.
As for the other stuff you mentioned:
* new compiler: old compiler plus bug fixes. The ultimate/long term
goal is obviously 100% compliance with the standard, but the more
realistic short term goal for whatever is released as the first version
is what I would call "practical compliance." They're focusing on
libraries like Boost and ACE and specific user-reported bugs. If you
have particular issues, get them into QC.
* new STL package: yes, they said it will be Dinkumware. Alisdair asked
at one point about support for ANSI TR1. The answer was something like,
"We aren't really doing anything about that," but I suspect that a
better answer would be that since they're using Dinkumware as the
library they should get support for a lot of those things automatically.
Anything that's just a library change ought to just work, assuming
there isn't some compiler bug that prevents it.
* new linker: I don't think anything was said about this, but I suspect
it's more or less the same situation as with the compiler. Old linker
plus bug fixes, so get your reports in now.
Short (or maybe not so short, we'll see) sidenote about QC:
I know there's a lot of skepticism about the effectiveness of QC,
particularly from the C++ customers. I think that's partly unfair.
Yes, it's true that not much has happened with C++Builder bugs that were
reported in QC, but that's not because of some failing of QC. It's
because nothing much has happened with C++Builder at Borland until
recently. Fairly obviously there was a lot of time and effort put into
C++BuilderX which was a complete departure from C++Builder.
Now that they've made it clear that they're working on C++Builder, or
some kind of C++Builder-like personality in Delphi, I think it's pretty
clear that QC is the place to make your case about bugs. Speak now or
forever hold your peace. I think if you look at Delphi and JBuilder,
it's pretty clear that QC votes do matter. Just the fact that there is
going to be a C++Builder at all is pretty strong evidence that user
votes make a difference.
--
Gillmer J. Derge [TeamB]
 

"Alisdair Meredith" <alisdair.meredith@ XXXX@XXXXX.COM >
wrote in message news: XXXX@XXXXX.COM ...
 

{smallsort}

"Gillmer J. Derge [TeamB]" < XXXX@XXXXX.COM >wrote in message
would
be progress. I have a few issues on QC.
be just a library change for BCB. It may force some compiler repair which
would be a good thing.
since BCB5. I would REALLY appreciate it if they would fix the
problem where the link stops when encountering RC files, though I don't
know if that's actually the linker at fault.
Thanks for the post Gilmer.
 

Duane Hebert wrote:
version. That suggests we are looking at early next year :(
--
Andrue Cope [TeamB]
[Bicester, Uk]
info.borland.com/newsgroups/guide.html
 

Andrue Cope [TeamB] wrote:
pattern for releases with that, but I'd guess very late 2005 or Q1 2006.
<QUOTE>
jkaster: derbas: "Has Borland not yet learned that "when it's ready" isn't a viable answer
anymore?" - We're trying to tell you as much as we can about the development we're doing. The
challenge of exact dates is a challenge we always have. We will strive to hit any dates we announce.
</QUOTE>
I think that all we ever ask for is "Were aiming for 1st quarter of 2006" or something like that. I
don't think that anyone ever asked for an exact date.
-Mike
 

Michael Gillen wrote:
They said it would be part of the next release of Delphi. If you look
back at the pattern of the timing of Delphi releases, I think that will
probably lead you to the conclusion that we're talking about 1Q2006 plus
or minus 1 quarter. Maybe 4Q2005 if we're lucky and get an early
release, maybe 2Q2006 if the timing slips, but somewhere in that ballpark.
So, yeah, they didn't give an actual date and I'll certainly concede
that it would be nice if they'd do that, but I don't think they left
customers nearly as in the dark as people are implying. And to be
honest, I'm not sure why this is any worse than Microsoft promising some
particular date, then pushing it out a year, then pushing it out another
6 months, then finally pushing it out another quarter before finally
releasing whatever it was they promised 2 years earlier.
--
Gillmer J. Derge [TeamB]
 

Michael Gillen wrote:
not faulting it, simply pointing it out). My original question included
the phrase "especially after the CBX debacle," as I was trying to point
out to them that I had hoped they had learned something from that whole
mess (i.e., telling BCB6 users that C++ was being worked on, then
releasing CBX as a pre-beta product with a resounding thud).
David Erbas-White
 

Gillmer J. Derge [TeamB] wrote:
OWE US more open communication, and particularly scheduling information.
Everyone involved with BCB knows that software schedules move
(generally slipping), but acting this way demonstrates that they really
HAVEN'T changed their stripes.
What really astonishes me is that most folks are acting as though
Borland is Miss Emily Litella, who after killing BCB and pushing CBX
went, "nevermind," and is proceeding on as though nothing happened.
The upheaval that was caused to Borland users by all of this nonsense
requires extremely STRONG, POSITIVE COMMUNICATION from Borland in order
to repair. Not communicating potential scheduling is absolutely
unacceptable.
David Erbas-White
 

David Erbas-White wrote:
starting to see that. Yesterday was an hour long interactive discussion
with a guy who was introduced as something like Lead R&D Engineer for
C++Builder. That's good communication.
I guess where I disagree is with the schedule part. I don't see why
anything that happened with CBX leads to a need for detailed schedule
information. IMHO a big part of the problem with CBX was that it wasn't
released when it was ready but rather quite a bit before it was ready.
So I would argue that the takeaway from CBX should be not that Borland
owes customers a schedule commitment but rather a renewed emphasis on
not saying anything about schedule. If they learned anything from CBX
it should be, "Wow, that whole thing about not releasing products until
they're ready is a good idea. We should actually listen to ourselves
when we say that."
--
Gillmer J. Derge [TeamB]
 

Gillmer J. Derge [TeamB] wrote:
What they were saying prior to the release of CBX (anytime anyone asked
about bug fixes for BCB6) was that they "were working on the next
release." I passed that on to my customers. We did the same thing that
they're asking us to do now, which is to 'estimate' when it would be out
based on past release cycles. Then, without any warning what was coming
down the pike (i.e., the release of CBX and the death of BCB), they
effectively kicked us in the crotch. That episode darned near cost me a
client I had for more than 15 years.
Now, YEARS after the release of WinXP, they still don't have a C++
product that works well with it, and they refuse to comment on when they
might have one. Further, it appears that any C++ product will be
happening later rather than sooner (witness all the speculation that
this will be out in 2006), so scheduling/planning information becomes
even more important.
There isn't a person on this group who can realistically believe the
Borland doesn't have an internal schedule for when this product will be
released. And at this point, that information is critical if they want
to retain the any of the remaining dwindling customer base. I'm
appalled that you think that the way that CBX was released actually
indicates that they should be keeping schedule dates secret. You're
forgetting that not only did they NOT let folks know what the CBX
release date was, but they ALSO kept saying the "when it's ready" mantra
for CBX, despite the fact that it obviously "wasn't ready." So, Borland
is saying nothing about schedule that I can do any planning on, there is
no (positive) history that I can infer from, and they have proved that
they are NOT committed to quality (i.e., for CBX "when it's ready" meant
absolutely nothing). The have also proved that promises of support and
change of direction are completely empty promises -- when CBX was
released, they made all sorts of commitments to improved product
support, providing migration paths, new open communication, and NONE of
that occurred -- in fact, it got worse (which I would have thought
impossible).
I agree that the 'chat' yesterday was a good idea, and that it was a
positive move. But when weighed against the damage that Borland has
caused, it is such a tiny move as to be almost worthless, unless Borland
has TRULY changed its corporate mind about supporting its C++ customers.
And (aside from the technical aspects of the product) the single most
important question (which was obvious from the number of times it was
asked) is SCHEDULE.
Again, I'm astounded that Borland doesn't "get it" yet -- but I guess
I'm not. Disappointed, I guess, and mad at myself for 'hoping' that
they had turned over a new leaf.
David Erbas-White
 

"Gillmer J. Derge [TeamB]" < XXXX@XXXXX.COM >writes:
[snip]
--
Oscar
 

David Erbas-White wrote:
schedule and release it when it's ready because of their dedication to
quality? You seem to want both, but I don't think they're compatible
goals. I certainly admit they don't seem to have followed then "when
it's ready" mantra with CBX, but if I have to pick one approach that I
*want* them to follow, that's the one.
--
Gillmer J. Derge [TeamB]
 

Gillmer J. Derge [TeamB] wrote:
exclusive.
I want them to release the product when it's ready.
I also want them to release information on their projected schedule.
If/when the schedule slips/changes, I want them to release information
on the schedule slip/change, as well as the reason for it.
There's not a developer alive who wouldn't understand that schedules
change -- but working completely in the dark is asinine.
David Erbas-White
 

David Erbas-White wrote:
take that with a pinch of salt due to unforseen problems, but at least I
have a good idea of the minimum time it will be before the new product
ships and I can see if it fits in with our timescales for new products
we have to produce.
But with no indication at all (and people saying Delphi will be due a
new release at the end of the year so expect it around then, isn't good
enough if it isn't from Borland) then I don't have a clue if it is
within our time scales or not.
Russell