QC: Please Vote for BDE report# 7089 on Quality Central
bcb
Many of you are aware of the old BDE bug which misreports the available hard
disk space when the actually there is more than enough. It happens when the
available disk space is a multiple of 4gb. More users are upgrading their
pc's these days so this will make the error appear more often than before.
So if you folks can go and please vote for report#: 7089 on QC this would
perhaps better the chances that Borland will fix this bug. Thanks!
This topic was originally started in the QualityCentral section of the
newsgroups:
http://newsgroups.borland.com/cgi-bin/dnewsweb?utag=&group=borland.public.bdn.qualitycentral&xrelated=4208&cmd_related=Related+Items
Download the QC client from:
http://qc.borland.com
Bill Todd (TeamB)
http://community.borland.com/article/0,1410,28688,00.html
--
Bill (TeamB)
(TeamB cannot respond to questions received via email)
Steven Green
> Before you waste your vote you should read
> http://community.borland.com/article/0,1410,28688,00.html
so they're voting on what they want fixed the most, but nobody has
offered to fix *anything*, right?
--
Steven Green - Waldorf Maryland USA
Diamond Software Group
http://www.diamondsg.com/main.htm
Paradox Support & Sales - Corel CTech Paradox
---------------------------------------------------
Diamond Sports Gems
http://www.diamondsg.com/gemsmain.htm
Trading Cards and other Sports Memorabilia
---------------------------------------------------
bcb
> http://community.borland.com/article/0,1410,28688,00.html
I understand, but this bug has existed well before May 14, 2002. Perhaps
this isn't the best comparison but if I remember correctly a Delphi6 patch
has been released even while Delphi7 has been out on the shelves, and if the
BDE is so old that they don't even want to support it, why did they ever
even waste a second of their time to make Delphi for .Net support it. If
everyone pitches in a vote and shows Borland how important this issue is,
they might well fix this annoying bug, that they should have fixed a long
time ago and save us a lot of grief. Anyhow most of the people who would
signup and vote wouldn't perhaps visit QC for any other reason, and if they
decide it's a waste of a vote they can always remove their vote for that
article.
bcb
Like they ever offer to fix some of the most annoying bugs in C++Builder ;-)
they existed in version prior to v6 and long before C++BuilderX was ever
announced. This is a major bug and there is and was no excuse not to fix it.
AFAIK this shouldn't be a major overhaul or a major investment on their
part. Win16 users can still keep running older versions of the BDE, for us
Win32 users (if I may say is the majority of applications running these
days) they should've released a fix/patch for us. Win32 and HD's over 4gb
have been around for at least 8 years , the last BDE update was approx ~2.5
years ago, so you tell me they did not find the time to fix this major bug?
P.S.
I do not mean to sound like a Borland basher, but please understand the
situation. There are thousands of developers all over the world using the
BDE, why should all of them use low level hooking techniques(which is of
course the proper way, the customer should not even see signs that this bug
even exists) to fix this while Borland could really make us all happy and
more loyal customers by just applying a small fix.
Steven Green
> If everyone pitches in a vote and shows Borland how important this issue is,
> they might well fix this annoying bug, that they should have fixed a long time
> ago
I agree with you, that the magnitude of the importance of this bug is growing
rapidly.. but that in itself isn't gonna get Borland to do anything about it, or
any other bug.. there's a big difference between the community saying "we want
this fixed" and Borland asking "what do you want fixed?"
Bill Todd (TeamB)
<gre...@diamondsg.com> wrote:
>so they're voting on what they want fixed the most, but nobody has
>offered to fix *anything*, right?
Read the article. To me the article states clearly that Borland is not
going to do any more work of any kind, including bug fixes, on the
BDE. All I am saying is that, IMHO, voting to fix a bug in a product
that Borland has said they are not going to fix is not a good use of
QC votes. It would be better to vote for the bugs that you want fixed
in D7 which Borland has indicated they are going to update.
Bill Todd (TeamB)
<gre...@diamondsg.com> wrote:
> there's a big difference between the community saying "we want
>this fixed" and Borland asking "what do you want fixed?"
Exactly, and Borland has asked the community to vote for what they
want fixed in D7.
My understanding is that the BDE bug did not get fixed when the BDE
was being maintained because it is difficult to fix in that it
involves changes in a lot of places in the BDE. That combined with
Borland's publicly stated plan for the BDE makes it very very unlikely
that this will be fixed.
dk_sz
> going to do any more work of any kind, including bug fixes, on the
> BDE. All I am saying is that, IMHO, voting to fix a bug in a product
I think that is stretching it a bit.... They dump SQL Links but are going to
maintain BDE
(with it's now "limited" functionality) for a while... That is how I
remmeber the article at least.
best regards
Thomas
Steven Green
> I think that is stretching it a bit
no, it's not.. if something as big as the "4 gig" error was gonna be fixed, it
would have been done a long time ago..
David R. Robinson
> was being maintained because it is difficult to fix in that it
> involves changes in a lot of places in the BDE. That combined with
> Borland's publicly stated plan for the BDE makes it very very unlikely
> that this will be fixed.
I agree that it's a waste of time. IMHO, there is less than a
0.00000000000000000000000000000000000000000000000000000000000001% chance
that Borland will fix this. There are many other more serious bugs in the
BDE than this. Those bugs have been around for more than 6 years. Borland
simply is not going to fix bugs like this in the BDE. If they were they
would have fixed them at one point in the past 6 years.
Yes, Borland spent some time including the BDE into D8, but that doesn't
mean that they did anything other than put some wrappers on it to make it
work under D8. It's still the same DBE that hasn't been updated in several
years. The BDE is dead technology. It hasn't been updated in at least 5
years and it isn't going to be updated again. The sooner people realize
that and replace it with something more modern the better off they will be.
I realize that it may not be feasible to update a legacy application and
IMHO that is the only reason someone should be using the BDE. However,
anyone still using the BDE should know not to expect any updates or changes
from Borland. It is not in their best interest to spend their R&D time or
effort on something that they publicly stated was dead a long time ago.
David R.
Jon Purvis
> Yes, Borland spent some time including the BDE into D8, but that doesn't
> mean that they did anything other than put some wrappers on it to make it
> work under D8. It's still the same DBE that hasn't been updated in several
> years. The BDE is dead technology. It hasn't been updated in at least 5
> years and it isn't going to be updated again. The sooner people realize
> that and replace it with something more modern the better off they will be.
I do use ADO now on a project, but only because it makes installs easier and I
wanted to play with it. I actually prefer the BDE and have never had a problem
with it. So, BDE works for me. It works fine with my copies of D5 and D7. I
can create an ODBC link to Paradox, Access, and MSSQL and BDE will work. I find
working with BDE dataset structures easier than with the ADO ones.
Maybe someone can explain to me the need to move away from the BDE. Why the
overwhelming importance in moving away from something that works just because a
new technology has come out? Why should I care about modern vs. old? All my
users care is that it works, not that the code is newer. BDE has bugs, but so
does ADO - I see no advantage in switching to modern bugs. Is ADO supposed to
be faster? Better somehow? I haven't found a single thing that I need in ADO
that BDE cannot do. Have I missed something?
Maybe I'm just resisting change. Maybe in .Net I'll be forced to move, but
that'll be several years before I have to worry about it. Is there any reason
other than "new toy" to move?
Bill Todd (TeamB)
<jon.p...@tpwd.state.tx.us> wrote:
>Have I missed something?
A couple of things. The world is moving to multi-tier data access
models for a variety of reasons and the BDE does not and cannot
support that. The BDE is very complex. It is half a million lines of
code including several thousand lines of assembler. It is very
expensive to maintain so, from Borland's perspective, something that
is less costly to maintain is very desirable. There are other reasons
but those encapsulate the major considerations.
Daniele Barbato
when 8 years ago I prefere Borland Delphi for develop my application we take
delphi for the bde. Now I have over 1000 installation and I use dBase table
and I haven't problem but if in the future the bde s compatible with
microsoft os platform?
If I use cobol with cobol database now I have a team that develop this tool,
but the borland no.
"Bill Todd (TeamB)" <n...@no.com> ha scritto nel messaggio
news:4iri2093526i8p5bs...@4ax.com...
Bill Todd (TeamB)
<newsd...@svemu.it> wrote:
>Now I have over 1000 installation and I use dBase table
>and I haven't problem but if in the future the bde s compatible with
>microsoft os platform?
Only Microsoft has any control over that.
Leslie Milburn
I agree with you and I will be voting, whats one vote? - ask whats his name
that lost to G.W. Bush - it can count!
Leslie.
"bcb" <noe...@sorry.com> wrote in message
news:40286530$1...@newsgroups.borland.com...
David R. Robinson
the
> overwhelming importance in moving away from something that works just
because a
> new technology has come out? Why should I care about modern vs. old? All
my
> users care is that it works, not that the code is newer. BDE has bugs,
but so
> does ADO - I see no advantage in switching to modern bugs. Is ADO
supposed to
> be faster? Better somehow? I haven't found a single thing that I need in
ADO
> that BDE cannot do. Have I missed something?
The point is that the BDE is no longer being supported or maintained. If
something happens where it stops working you are out of luck.
Are you not worried that the next OS from MS or the next SQL Server update
or a MS Security update is going to break the BDE? Where I used to work we
had over 100,000 users using the BDE with out application. We supported MS
SQL, Oracle, InterBase, Sybase, Informix and Paradox (we dropped Sybase,
Informix and Paradox toward the end). I was very concerned that some update
from MS or a database vendor was going to come out that broke the BDE and I
would overnight have 100,000 users that couldn't run my software. At some
point that is going to happen. It is inevitable. Granted, it may not
happen for 5 more years, but some day it will happen.
MS, Oracle, etc. are not testing their software to be sure it works with the
BDE. The BDE SQL Links haven't supported new versions of Oracle or MS SQL
Server for several years. Borland isn't testing the BDE against new
versions of SQL servers. Eventually the BDE is going to stop working. If
you don't ever make plans to get rid of the BDE and continue new development
projects using the BDE you are increasing your dependency upon unsupported,
untested and archaic technology. If you think that is in your best interest
then that is your decision. Nobody knows your business situation better
than you. You just have to realize what the ramifications will be if
something happen that causes the BDE to stop working.
David R.
Bill Todd (TeamB)
<jon.p...@tpwd.state.tx.us> wrote:
>Then I would have to read
>messages everyday saying ADO is isn't supported, is archaic, and only dinosaurs
>are using it.
They already have for the .NET environment. You do not have to worry
about ADO being superceded in the Win32 environment because MS is not
developing anything for Win32. They are entirely focused on .NET and
Longhorn.
Jon Purvis
> The point is that the BDE is no longer being supported or maintained. If
> something happens where it stops working you are out of luck.
>
> Are you not worried that the next OS from MS or the next SQL Server update
> or a MS Security update is going to break the BDE? Where I used to work we
> had over 100,000 users using the BDE with out application. We supported MS
> SQL, Oracle, InterBase, Sybase, Informix and Paradox (we dropped Sybase,
> Informix and Paradox toward the end). I was very concerned that some update
> from MS or a database vendor was going to come out that broke the BDE and I
> would overnight have 100,000 users that couldn't run my software. At some
> point that is going to happen. It is inevitable. Granted, it may not
> happen for 5 more years, but some day it will happen.
I am no more worried about MS breaking the BDE than them breaking ADO.
Actually, my bigger worry is that I switch over to ADO, and six months later
Microsoft comes out with yet another technology. Then I would have to read
messages everyday saying ADO is isn't supported, is archaic, and only dinosaurs
are using it. MS did this with DAO, so I fully expect it to happen with ADO in
the next couple years. Why switch because something may happen? I agree that
eventually it will happen, but I can wait until it does.
> MS, Oracle, etc. are not testing their software to be sure it works with the
> BDE. The BDE SQL Links haven't supported new versions of Oracle or MS SQL
> Server for several years. Borland isn't testing the BDE against new
> versions of SQL servers. Eventually the BDE is going to stop working. If
> you don't ever make plans to get rid of the BDE and continue new development
> projects using the BDE you are increasing your dependency upon unsupported,
> untested and archaic technology. If you think that is in your best interest
> then that is your decision. Nobody knows your business situation better
> than you. You just have to realize what the ramifications will be if
> something happen that causes the BDE to stop working.
I work for a government agency (Texas Parks and Wildlife) and until the current
project used Access. Create an ODBC link, and BDE works. It may not be the
optimal solution, but it worked well, and was plenty fast enough for the small
databases and 2-3 users that would connect to them. Also works great for the
single user, local database project I did. Now I'm working a bigger project
(small MSSQL database, 3 users with write access, eventually ~20 read access)
and have chosen ADO only because I wanted to play with it. I'm lucky - I don't
have to worry about selling software and what all the users are doing. When we
upgrade the MSSQL, I'll likely be the one doing the upgrade and so will know
about it well ahead of time. If it doesn't work, well, I'll put the old version
back on (I use exactly 1 server, so this isn't such a big deal) and we'll
continue with it.
My sense is that BDE is sort of like Win32. It's the old thing, and to prove
you know your stuff to paying clients, you have to go the new thing (ADO, .Net).
Both work, but clients want to hear about the new. My clients are generally so
computer illiterate that they can't spell ADO, let alone know which is the newer
way of doing things. Chances are reasonable that by time the old doesn't work,
the new will have changed greatly as well. I'd rather wait until the old
doesn't work, than learn the current new and hope it's still relevant 5 years
from now when I actually need it.
dk_sz
fixed, it
> would have been done a long time ago..
Well - in that case they aren't working on Delphi(Win32 branch) either ;-)
best regards
Thomas
Rob Taylor
I realize it's a shocker, but there are a lot of us poor slobs
still using Paradox (an love it), but are at the mercy of the
BDE. Large disks are a way of life, and surely a bug like this should be fixed. No one is asking for the world here. Just a
fix for a myopic piece of (non)foresight.
C P
number of years ago that we shouldn't continue to depend on it. I think
those of us that are in a bind because of this only have ourselves to blame.
At some point, it just doesn't make sense for a company to keep maintaining
old software - even if some users really like it. It seems Borland doesn't
want to try to compete in the desktop database market any more. With MS'es
FoxPro and Access, and a number of vendors supporting dBase access, I think
this is a reasonable choice. From what Bill (Todd) has written, this fix
would involve some big/difficult changes to the BDE code and it doesn't
really make much sense for Borland to do that given that they don't seem to
want to focus on their own desktop database system as a source of income.
"Rob Taylor" <creac...@alltel.net> wrote in message
news:402b8414$1...@newsgroups.borland.com...
Daniele Barbato
to it Open Source.
Best Regards
"C P" <nos...@nospam.com> ha scritto nel messaggio
news:402b8b29$1...@newsgroups.borland.com...
Stephen Beausang
Paradox and what are the implications.? I am not sure if it is a problem for
my company or not..
Stephen
"Rob Taylor" <creac...@alltel.net> wrote in message
news:402b8414$1...@newsgroups.borland.com...
>
Stephen Beausang
intebase Databases as well
TIA
Stephen
"Steven Green" <gre...@diamondsg.com> wrote in message
news:402B9780...@diamondsg.com...
> Stephen Beausang wrote:
>
> > Could Somebody explain this large disk problem...Is is just a problem
with
> > Paradox and what are the implications.? I am not sure if it is a problem
for
> > my company or not..
>
> it's one of those "signed integer" vs "unsigned integer" things.. and it's
> aboutt how the BDE interprets the available disk space value it's being
fed.. at
> every 4 gig cycle, it gets confused and thinks there's zero left..
>
> how much more than the 4 gig threshhold you need to have available,
depends on
> the tables and context.. if there's 4.1 gig out there, and BDE thinks
there's
> only .1, and you're working with small tables, you fly on past it.. but if
> there's 4.1 gig out there, and you're getting ready to query, restructure,
or
> fill a half-gig table, it'll barf on you..
Steven Green
> Could Somebody explain this large disk problem...Is is just a problem with
> Paradox and what are the implications.? I am not sure if it is a problem for
> my company or not..
it's one of those "signed integer" vs "unsigned integer" things.. and it's
Steven Green
> This is helpful, is it only a problem with Paradox tables or would it affect
> intebase Databases as well
I'd have to assume both, but somebody else will have to confirm.. the issue is
the BDE deciding how much space it needs to do something, and how much space it
thinks it has available..
C P
Borland dev product(s) to be able to distribute the BDE. I have seen many
posts discussing this and apparently Borland has no plans to release the
source for the BDE. I suppose some of the code used in the BDE may still be
relevant to some of Borland's other products, and they may not want to share
it.
"Daniele Barbato" <newsd...@svemu.it> wrote in message
news:402b8ec4$1...@newsgroups.borland.com...
Olivier Beltrami
> number of years ago that we shouldn't continue to depend on it. I think
> those of us that are in a bind because of this only have ourselves to
blame.
I'm sorry but some of us have to maintain old and large existing client
bases and do not have the luxury of upgrading one's "Hello World" app to the
latest and greatest version of Delphi and whatever brain-dead DB-access
technology is the currently favored. The BDE has many flaws but at least it
is stable, warts and all. What would have happened if I had selected any of
the umpteen DB access layers proposed by Borland since Delphi 5 ?
Additionally, Borland has repeatedly stated recently that it will keep on
supporting desktop DBs. To me that means Paradox.
Olivier
Liz
number of BDE users (and esp. those who use Paradox the software
application, not just Paradox the table format) would be willing to
pay a pretty large price for a developer license (that includes
unlimited distribution with an application) if certain key problems
were fixed.
I expect they could make a pretty good profit off of one more patch
(but likely, that patch would have to wait until they could make it
compatible with "longhorn" - BDE works now, all bugs can be worked
around, the biggest concern is the next version of Windows).
But Borland have been really hard-nosed on this, so debating it is
useless, I suspect.
Liz
C P wrote:
>
> It sounds like this problem will be expensive to
> fix.
C P
We were warned a few years ago about this, and we all had to weigh our
options of when/if to move away from the BDE. Moving off of the BDE will
not be trivial for me either - but I don't think that obligates Borland to
keep supporting the BDE. It sounds like this problem will be expensive to
fix. That means they'll have to pass the cost off to us that buy Borland
products in one way or another. But with so many other free/cheap options
now available (ADO for example), I'm sure it doesn't make a whole lot of
sense for them to do this, especially when the BDE's days are probably
numbered even if this bug gets fixed (I'm thinking of Longhorn and .NET
here).
As far as I know, BDE was replaced by dbExpress which is still the way to go
for Win32. .NET introduces new a data access layer, but that's a non-issue
for those of us sticking with Win32 apps. I'm not aware of any other data
access layer besides dbExpress proposed by Borland. They did allow us to
use ADO components, but that wasn't a replacement, rather an alternative
that they made easier to use. We can use ADO without Borland's ADOExpress
components. So, I don't think there's been umpteen data access layers from
Borland since the BDE.
To me desktop DB's include dBase, FoxPro, Access, Pervasive, DBISAM, etc...
not just Paradox. Borland has retained the ability for us to make use of
the BDE 'as-is' in D6, 7 and 8, but I don't think they've ever promised to
update the BDE for this. We can use any of these other non-Paradox desktop
DB's without the BDE, so it seems to me they are continuing to support
desktop db's in general.
Again, I understand the frustration that many Borland developers feel about
this - it will impact me too in a big way. But, when I put myself in
Borland's shoes, I'm sure I'd make the same decision. From what I've read
(and some assumptions I've made) it seems that continuing to develop the BDE
would not be a sufficient revenue source (or worse yet would lose Borland
money).
Chris
Liz
won't run without the BDE - BDE won't fire up, neither will Paradox.
There are tens of thousands of Paradox installations out there. Many
(most?) of them aren't developers - they're folks who use Paradox for
ad-hoc querying and reporting against various data sources.
For Paradox developers, it's not just a simple matter of switching
data platforms - it's a matter of learning a completely different
development environment and every one that I've seen (Delphi, C, VB,
what have you), is harder to learn, more complex and requires the
developer to do more to get the same results. (Paradox is far more
limited in what it "can" do than those others, but for what it does
do, it's unbeatable (in the opinion of Paradox developers).)
(Again, as mentioned, I think this is all irrelevant here anyway,
Borland aren't going to do anything about it - even if they could make
a one-time profit on a patch.)
Liz
C P wrote:
>
> Well according to Bill, the BDE involves a lot of complex (and) assembler
> code, so this is a non-trivial bug. Faced with the choice of a large BDE
> license or the cost of moving to a different platform, I'm not sure how many
> would really make the choice to pay for the BDE. Also, I don't know if
> there's that many of us using the BDE in the first place to make this
> cost-feasible.
C P
code, so this is a non-trivial bug. Faced with the choice of a large BDE
license or the cost of moving to a different platform, I'm not sure how many
would really make the choice to pay for the BDE. Also, I don't know if
there's that many of us using the BDE in the first place to make this
compatible with Longhorn. Back when I chose the BDE/Paradox it was the best
choice, but now (even if it worked with Longhorn) I don't think I'd pick the
BDE or Paradox (given all the other options we currently have). I suspect
this is true for most developers. Even if Borland did decide to re-start
support of the BDE, it would only be for a constantly dwindling number of
developers supporting legacy apps. So it doesn't really make much sense for
Borland to invest in something that realistically won't generate new sales.
It would only be for a shrinking number of existing users.
"Liz" <l...@paradoxcommunity.com> wrote in message
news:402BCD17...@paradoxcommunity.com...
Liz
For basic end-user ad-hoc stuff, Access may well be easier/better.
For development of full-featured multi-user database application,
Paradox leaves Access in the dust (from all the comparisons I've read
and from what little time I've spent with Access - you'd have to
switch to VB and MSDE to get to what Paradox can do).
Unfortunately, it's impossible to say how many installs there are out
there (neither Borland nor Corel will talk on the subject), but based
on info from the Paradox newsgroups, I don't think 10s of 1000s is
unrealistic. (There are still an awful lot of people using Paradox
for DOS applications (even performing new installs of the same),
Paradox 5 apps, Paradox 7-16bit, etc. etc.)
I think faced with paying $500 (let us say, for the sake of argument)
to continue using what one knows, or paying $500 to buy a replacement,
then spending the time (and money) to learn the replacement, and then
more time to redo everything in the replacement, most folks would
choose to stick with what they know. Especially developers with
full-featured applications, for whom the replacement product may cost
more than $500, for whom training would cost $1000s, for whom getting
up to the same skill level would take months to years, and for whom
re-doing their entire app would take weeks-months-years.
Liz
C P wrote:
>
> Ah, I missed that part about the Paradox app vs. Paradox tables. I'm
> surprised to hear there's that many Paradox (the app) users. In the 100's
> of sites I've been involved with, I've yet to see someone using Paradox (the
> app) - it's always Access, or maybe FileMaker. I would be interested to
> know how many such users there are out there. While I've never used Paradox
> itself, I'd be surprised if it has much of an advantage over Access in terms
> of simplicity or power. To me Access queries seem a lot easier for a basic
> user than a QBE query, and Access has a ton of wizards for the novice user.
> And let's not forget it is really easy to integrate with Word and Excel. Do
> you really think a user with a basic self-built Paradox app would pay for a
> BDE update rather than just move to Access?
>
> Anyway, I agree that it's probably a pointless debate... I just enjoy a good
> debate. :)
>
Steven Green
> I would be interested to know how many such users there are out there.
there are still thousands and thousands of them.. and will be for a long time..
there are still people using PdoxDOS 3.5 coming out of the woodwork all the
time.. it's frightening, actually..
> While I've never used Paradox itself, I'd be surprised if it has much of an
> advantage over Access in terms of simplicity or power.
it's not just queries.. there are so many things that you can do down-and-dirty
with Paradox, so easily, that you can't even dream of doing with other
products..
> Do you really think a user with a basic self-built Paradox app would pay for a
>
> BDE update rather than just move to Access?
geez.. and I always thought there was no such thing as a dumb question <g>..
that "user with a basic self-built app" typically runs his whole business in
that app.. he doesn't want to move to *anything* else..
C P
surprised to hear there's that many Paradox (the app) users. In the 100's
of sites I've been involved with, I've yet to see someone using Paradox (the
know how many such users there are out there. While I've never used Paradox
itself, I'd be surprised if it has much of an advantage over Access in terms
user than a QBE query, and Access has a ton of wizards for the novice user.
you really think a user with a basic self-built Paradox app would pay for a
BDE update rather than just move to Access?
Anyway, I agree that it's probably a pointless debate... I just enjoy a good
debate. :)
"Liz" <l...@paradoxcommunity.com> wrote in message
news:402BD59E...@paradoxcommunity.com...
C P
> there are still thousands and thousands of them.. and will be for a long
time..
> there are still people using PdoxDOS 3.5 coming out of the woodwork all
the
> time.. it's frightening, actually..
...well, "you learn something new every day". I didn't realize that Paradox
(the app) was so heavily used. I guess that makes the decision Borland has
made a little less simple. But even though the current BDE market may be
bigger than I had thought, it's still really a shrinking market. That being
the case it would seem wise to direct their resources elsewhere. If they
did fix this bug, I'm sure they'd really end up having to fix several things
to really make charging for an update tolerable. And, that makes fixing
this bug a lot more expensive and less appealing for Borland.
> geez.. and I always thought there was no such thing as a dumb question
<g>..
> that "user with a basic self-built app" typically runs his whole business
in
> that app.. he doesn't want to move to *anything* else..
I realize that users do feel like this, but they can't have their cake and
eat it too. I don't think people can realistically expect to change some of
their hardware (and/or software) to all kinds of newer technologies and
expect it to all work well with legacy software (or hardware). Everything
is tied together in computers (despite our efforts to avoid this). There's
always going to be a limit to the forward compatibility of old stuff, and
backwards compatibility of new stuff. And, if we want the benefits of new
technologies we'll often have to deal with the cost of updating much of our
system (of software and hardware). If a user insists on using some old
software (i.e. the BDE) then maybe they've got to accept certain limitations
(such as using older, smaller hard drives). It's not pleasant when you are
faced with this situation, but I don't see it as avoidable. The BDE is old,
and it works with old (small) hard drives. If you want to move to newer
(bigger) harddrives, then you've got to move to a new DB platform.
"Steven Green" <gre...@diamondsg.com> wrote in message
news:402BDCF2...@diamondsg.com...
Steven Green
> But even though the current BDE market may be bigger than I had thought, it's
> still really a shrinking market. That being the case it would seem wise to
> direct their resources elsewhere.
agreed.. absolutely.. like you already said, the vast majority of the market is
"legacy"..
> I don't think people can realistically expect to change some of their hardware
> (and/or software) to all kinds of newer technologies and expect it to all work
> well with legacy software (or hardware).
so far, for the most part, it *does* work.. and "DOS" works better on XP than on
2000 or NT.. in spite of what MS would like you to believe..
> The BDE is old, and it works with old (small) hard drives. If you want to
> move to newer (bigger) harddrives, then you've got to move to a new DB
> platform.
if and when the issue comes into play on a system, the term "smaller partition"
is still valid, and really easy to accomplish.. no more than a trivial annoyance
to most..
David R. Robinson
That makes sense to me. I'd probably do the same if I were in your shoes.
I agree that you have the same problem with MS dropping technology as you do
with any other company. As Bill stated, Win32 is pretty much dead at this
point and MS is all focsed on .NET. In your case you don't have many users
and you have a very controlled environment. You can probably make the BDE
last for you for a long time.
David R.
bcb
> I agree with you and I will be voting, whats one vote?
That's the spirit.
Thanks :-)
To vote doesn't take much of your time and you have nothing to lose, this
bug should have been fixed but perhaps at the time it wasn't clear that so
many people are affected, if it was understood then maybe Borland(which is a
company I still have a bit of respect left for) would have solved this
issue. One doesn't have to offer a fix to actually fix something, if you see
a major glitch you go fix it yourself, when a problem is reported I try to
fix it ASAP.
If users of a product have need for a very important bug fix and enough
users request that, the product being in "end of life" or as you folks put
it "maintenance mode" doesn't really matter. Oh why didn't BDE users gather
their requests for this to be fixed before the BDE was being declared as
being in maintenance mode , thus the problem would be more likely fixed back
then :-( , I agree on that.
Say you've released a v1 of a product which has a serious bug, then released
v2. v1 is "not" being maintained anymore. At the moment the total user base
using v1 is still considered significant or as many as those using v2, or
something similar to this case: users may have deployed their product based
on your v1 product in a combined total of at least 3 times fold of the user
base of v2. Seeing so many people are affected by the glitch/bug, I
personally would release a fix if I developed that v1 software.
And does "maintenance mode" mean something? if it means what is says here
http://dictionary.reference.com/search?q=maintenance then Borland should
consider a patch, or perhaps the word "maintenance" is used as a gimmick
where in fact it's really "end of life" , which also isn't an excuse if a
major bug exists in your software(that many people still use and "complain"
about).
Rodney Wise
--
...
`·.¸¸.·´¯`·.¸¸.·´¯`·-> rodney
Bill Todd (TeamB)
wrote:
>The BDE runs fine under Linux using WINE.
Are you saying that the BDE does not exhibit the 4 gig free space
error when running under WINE on Linux? That does not seem possible.
Phil Hassid
> models
Pardon the ignorance but what does that mean exactly Bill?
> for a variety of reasons
What are they, briefly?
Phil
Phil Hassid
> <g>..
> > that "user with a basic self-built app" typically runs his whole
business
> in
> > that app.. he doesn't want to move to *anything* else..
In my case (as with many others no doubt) the app is so extensive and
customised that it has virtually *become* the business. I will do a *great*
deal to avoid having to change programmes and I do not care one bit what the
conventional wisdom is in the IT world.
> I realize that users do feel like this, but they can't have their cake and
> eat it too. I don't think people can realistically expect to change some
of
> their hardware (and/or software) to all kinds of newer technologies and
> expect it to all work well with legacy software (or hardware). Everything
> is tied together in computers (despite our efforts to avoid this).
There's
> always going to be a limit to the forward compatibility of old stuff, and
> backwards compatibility of new stuff. And, if we want the benefits of new
> technologies we'll often have to deal with the cost of updating much of
our
> system (of software and hardware). If a user insists on using some old
> software (i.e. the BDE) then maybe they've got to accept certain
limitations
> (such as using older, smaller hard drives). It's not pleasant when you
are
> faced with this situation, but I don't see it as avoidable. The BDE is
old,
> and it works with old (small) hard drives. If you want to move to newer
> (bigger) harddrives, then you've got to move to a new DB platform.
No offence, but this argument typifies the glaring extent to which the whole
IT world has come adrift from much of small business thinking. Moore's Law
can go take a hike as far as I am concerned. My business runs as well as it
will *ever* need to with the technology I have. Put me up 15 years from now
against a competitor with the latest and greatest and me with what I have
now (based on Paradox the programme) and I would do then what I do now - eat
them alive.
Phil
C P
I don't think we disagree here. If your software/hardware works for you,
then it would not make sense to upgrade just because something is new. What
I am saying though is that it's probably not reasonable to expect to do
significant upgrades to one part of your system (i.e. your hardware), and
then not have to change other parts (i.e. the software - Paradox). This is
because (unfortunately) these things are more dependent on each other than
we might like. If Paradox is working for you, then great - stick with it,
but also realize that means sticking with smaller drives/partitions. This
shouldn't be a problem if your solution is working like you say it is. I'm
not advocating at all that people should always be upgrading to the latest
and greatest. I'm just saying often you can expect to have to upgrade
several things when they're part of a dependent system, even if all you
really want is to upgrade one part of that system. I'm also not saying this
is a good thing, but rather a current reality of computers (and not just MS
stuff).
Bill Todd (TeamB)
<philh...@ozemail.com.au> wrote:
>> multi-tier data access
>> models
>
>Pardon the ignorance but what does that mean exactly Bill?
DataSnap and ADO.NET. It is an architecture where the client
application talks to a middle tier applicaiton server and the middle
tier application server handles all database access, business rule
enforcement and other functions not related to the U/I. One of the
advantages of this is that you only have to install the database
client software on the application server, not on every machine.
Another advantage is short running transactions. Another is low
network traffic which is important for distributed apps. It is commone
for the application server to be in a different location from the
clients.
>
>> for a variety of reasons
>
>What are they, briefly?
The BDE is large, complex and costly to maintain.
It is difficult to deploy and complex to configure.
Because it does so much is is a relatively thick middleware layer and
imposes a performance penalty.
It never became a standard so no one but Borland ever supported it or
developed drivers for it.
It does not support multi-tier architecture and that is what most
organizations are interested in today.
Phil Hassid
> then it would not make sense to upgrade just because something is new.
What
> I am saying though is that it's probably not reasonable to expect to do
> significant upgrades to one part of your system (i.e. your hardware), and
> then not have to change other parts (i.e. the software - Paradox). This
is
> because (unfortunately) these things are more dependent on each other than
> we might like. If Paradox is working for you, then great - stick with it,
> but also realize that means sticking with smaller drives/partitions. This
> shouldn't be a problem if your solution is working like you say it is.
I'm
> not advocating at all that people should always be upgrading to the latest
> and greatest. I'm just saying often you can expect to have to upgrade
> several things when they're part of a dependent system, even if all you
> really want is to upgrade one part of that system. I'm also not saying
this
> is a good thing, but rather a current reality of computers (and not just
MS
> stuff).
And what I am saying does not just speak to the manner of someone in my
position dealing with my predicament which we both understand, what I am
saying is that from *one* small business perspective (and I realise there
will be many others especially those who intrinsically gain more from the
latest and greatest) the IT tail has been wagging the business dog for too
long. That is just fine for big business, but not for many of the millions
of small businesses around the world. An IT industry formula needs to be
found where small businesses should be able to avoid the risks you explain
well for a longer time (at some reasonable cost to those businesses). The
problem is that much of the IT industry appears to be either disinterested
or outright antagonistic. To me something has to give at some point of time.
Phil
Phil Hassid
> application talks to a middle tier applicaiton server and the middle
> tier application server handles all database access, business rule
> enforcement and other functions not related to the U/I. One of the
> advantages of this is that you only have to install the database
> client software on the application server, not on every machine.
> Another advantage is short running transactions. Another is low
> network traffic which is important for distributed apps. It is commone
> for the application server to be in a different location from the
> clients.
Oh OK you were referring to client-server (so multi-tier is the more generic
name). I thought you might have been meaning there was yet more layers other
than clients and servers.
> The BDE is large, complex and costly to maintain.
> It is difficult to deploy and complex to configure.
> Because it does so much is is a relatively thick middleware layer and
> imposes a performance penalty.
> It never became a standard so no one but Borland ever supported it or
> developed drivers for it.
> It does not support multi-tier architecture and that is what most
> organizations are interested in today.
Thank you for your explanations.
Phil
C P
> position dealing with my predicament which we both understand, what I am
> saying is that from *one* small business perspective (and I realise there
> will be many others especially those who intrinsically gain more from the
> latest and greatest) the IT tail has been wagging the business dog for too
> long. That is just fine for big business, but not for many of the millions
> of small businesses around the world. An IT industry formula needs to be
> found where small businesses should be able to avoid the risks you explain
> well for a longer time (at some reasonable cost to those businesses). The
> problem is that much of the IT industry appears to be either disinterested
> or outright antagonistic. To me something has to give at some point of
time.
>
> Phil
I hear what you're saying... But, I'm not sure if IT is disinterested in
improving things. Rather, I think it's in IT's interest to keep things
working this way - it's more money and business for those of us in IT when a
small upgrade to one component of a system forces an upgrade to many other
parts. I don't think this is unique to IT either. No manufacturer wants to
build a car or toaster or whatever that never breaks down or becomes (too)
obsolete. If they did, then you'd never need to buy anything from them
again, and there goes their future income. I think much of the world's
economy is dependent on the fact that stuff gets replaced - goods that last
forever aren't good for business. We (especially in the Western world) are
a throw away society in general. And I think you're right, this is not a
sustainable approach to business (or anything else). However, a lot more
than just IT needs to change. IT is simply following the common pattern of
our society: "out with the old, in with the new" (and not just for new's
sake, but because we make sure that "old" can't persist).
Bill Todd (TeamB)
<philh...@ozemail.com.au> wrote:
> I thought you might have been meaning there was yet more layers other
>than clients and servers.
That is exactly what I mean. A typical client/server application is a
two tier environment with the client application as one tier and the
database server as the second. What I described is a three tier
application with the client responsible for the user interface, a
middle tier application server that handles all database access and a
database server to store the data. Sorry that I was not clear.
Leslie Milburn
Well my vote makes it a huge three :-(
This is really PATHETIC.
I know for a fact that many of you on the PNEWS groups have suffered from
this problem - pull your finger out folks and lodge a vote. So many have
whinged about the BDE availability etc but the reality is that this is a
test to give Borland at least some idea of how many of us still require it
out there.
So, all you Paradox for Windows Users VOTE NOW.
Leslie.
"bcb" <noe...@sorry.com> wrote in message
news:4027...@newsgroups.borland.com...
> Hello,
>
> Many of you are aware of the old BDE bug which misreports the available
hard
> disk space when the actually there is more than enough. It happens when
the
> available disk space is a multiple of 4gb. More users are upgrading their
> pc's these days so this will make the error appear more often than before.
> So if you folks can go and please vote for report#: 7089 on QC this would
> perhaps better the chances that Borland will fix this bug. Thanks!
>
> This topic was originally started in the QualityCentral section of the
> newsgroups:
>
http://newsgroups.borland.com/cgi-bin/dnewsweb?utag=&group=borland.public.bd
n.qualitycentral&xrelated=4208&cmd_related=Related+Items
>
> Download the QC client from:
> http://qc.borland.com
>
>
Paul Barfoot
So, all you Paradox for Windows Users VOTE NOW.
It has my vote'
I have always considered Paradox to be the best.
Lets get that bug fixed.
Cheers
Paul
Leslie Milburn
Great to hear, please lodge your vote on Quality Central to make sure that
it counts.
Link is http://qc.borland.com and the report # is 7089, you do not need to
donwload the Windows client, choosing the web interface is easiest (and
quickest).
Good on you.
Leslie.
"Paul Barfoot" <paul...@sentex.net> wrote in message
news:4034b594$1...@newsgroups.borland.com...
Dominic Dumée
Cheers
Dominic
Liz
that hard to set up an HTML page where people log on and vote -
why on earth do we have to download software to do it....
FYI, for those who haven't got the software, there's no install,
just a single exe that seems to be self contained. (Reports >
Goto Report... 7089; then Activities > Vote On Report... enter 5;
or click on the Votes hyperlink)
If I understand correctly, one can vote from 0 to 5 on this one,
but if you vote the 5, you can't vote on any other BDE issues
(should there even be any), is that right? (FYI, looks like you
can remove votes later if you want.)
If any of the pnews folk want to set up an account and send me
their login, I'll vote for them so they don't have to install the
software.
I'll repost this on pnews for everyone's convenience.
Liz
Bill Todd (TeamB)
wrote:
>OK, I did it, but do they have to make it such a pain - is it
>that hard to set up an HTML page where people log on and vote -
>why on earth do we have to download software to do it....
There is a Web client you can use without downloading anything.
Liz
really don't feel like giving. Pick your evil, I guess.
Liz
Phil Hassid
some fuzzy Zen considerations.
Phil
Phil Hassid
> > I thought you might have been meaning there was yet more layers other
> >than clients and servers.
>
> That is exactly what I mean. A typical client/server application is a
> two tier environment with the client application as one tier and the
> database server as the second. What I described is a three tier
> application with the client responsible for the user interface, a
> middle tier application server that handles all database access and a
> database server to store the data. Sorry that I was not clear.
Thanks for persisting with me. I get you now. So at what size of
organisation / network do these advantages kick in?
Phil
C P
the multi-tier approach. The idea is that while you can put these different
tiers on different machines if you want, you could put them all on one
machine if you needed to. That gives you a lot of scalability. Even
planned you app to be a standalone app running on a single machine, there
are still some advantages to the multi-tier approach. By splitting your UI,
business logic, and storage into separate tiers, it should make them more
independent. Changing the database backend should be an easier job than if
your UI, business logic etc. was all in one. It also opens up the
opportunity to add new/alternate UI's without having to worry about breaking
business rules. And so on...
"Phil Hassid" <philh...@ozemail.com.au> wrote in message
news:4035...@newsgroups.borland.com...
Bill Todd (TeamB)
<philh...@ozemail.com.au> wrote:
>So at what size of
>organisation / network do these advantages kick in?
There is no easy answer to that. The main advantages are less network
traffic between the client and the middle tier than you would have
between the client and the database server in a two tier application
so multi-tier applications are an advantage when clients will be
connecting over a relatively low speed connection such as the
Internet. With multi-tier applications you do not have to install the
database client software on the client PC's. The database access
software is only required on the application server machine so
deployment is easier. Once again this is a particular advantage when
there are many clients and/or they are displersed geographically.
In some cases two or more client applications may be able to share a
single application server. That means code can be centalized in one
place which avoids duplication and makes maintenance easier.
Multi-tier applications can support a stateless connection between the
client and the middle tier application server. This allows many
clients to share a pool of connections reducing the load on the
application server and the database server. Others may suggest other
advantages. You need to look at each of these factors in relation to
your needs and decide if a multi-tier solution is best for you. HTH.
bcb
Also thanks to Liz for helping spread the word.
Phil Hassid
from
> the multi-tier approach. The idea is that while you can put these
different
> tiers on different machines if you want, you could put them all on one
> machine if you needed to. That gives you a lot of scalability. Even
> planned you app to be a standalone app running on a single machine, there
> are still some advantages to the multi-tier approach. By splitting your
UI,
> business logic, and storage into separate tiers, it should make them more
> independent. Changing the database backend should be an easier job than
if
> your UI, business logic etc. was all in one. It also opens up the
> opportunity to add new/alternate UI's without having to worry about
breaking
> business rules. And so on...
OK I see the advantages. Thank you very much for that explanation. What
would be the disadvantages (especially on the extreme case of single
workstation you referred to)?
Phil
Phil Hassid
Phil
C P
typically means a bit of extra overhead (how much depends on how you build
everything). However, if the multi-tier solution is built well, I don't
think this should be too much of an issue.
Reinaldo Yañez
http://codecentral.borland.com/codecentral/ccWeb.exe/listing?id=21475
Regards,
Reinaldo
Steven Green
> Hi, please see a patch posted in:
excellent.. you've most likely made a lot of people really, really happy..
but.. will this only work within a Delphi app, or can it be used as a wrapper so that that other BDE platforms can use it, or *can* it
be made into a wrapper that other apps can use?
the beauty of Lesspace was that you'd load it, then load Paradox or whatever, then unload it on the way out..
--
Steven Green - Waldorf Maryland USA
Diamond Software Group
http://www.diamondsg.com/main.htm
Paradox Support & Sales - Corel CTech Paradox
---------------------------------------------------
Diamond Sports Gems
http://www.diamondsg.com/gemsmain.htm
Trading Cards and other Sports Memorabilia
---------------------------------------------------
Reinaldo Yañez
http://codecentral.borland.com/codecentral/ccWeb.exe/listing?id=21475
Regards,
Reinaldo
"Bill Todd (TeamB)" <n...@no.com> wrote in message
news:q4eg205hdbgqslvtb...@4ax.com...
> Before you waste your vote you should read
> http://community.borland.com/article/0,1410,28688,00.html
Steven Green
> Hi, please see a patch posted in:
you should post this again, as a new thread.. buried in this thread, most
people won't even see it..
Reinaldo Yañez
I created a small DLL that is a wrapper to the code fix. To use it
you just have to call
LoadLibray('FIX4GBug.dll').
If you want to "unload the patch" you just call
FreeLibrary(handle)
If somebody wants the DLL please ask me for it and I will send it by email.
Regards,
Reinaldo
"Steven Green" <gre...@diamondsg.com> wrote in message
news:403CE612...@diamondsg.com...
Reinaldo Yañez
This is the link:
http://codecentral.borland.com/codecentral/ccWeb.exe/listing?id=21475
Regards,
Reinaldo Yañez
Reinaldo Yañez
This is the link:
http://codecentral.borland.com/codecentral/ccWeb.exe/listing?id=21475
Regards,
Reinaldo Yañez
"Liz" <l...@paradoxcommunity.com> wrote in message
news:4034E053...@paradoxcommunity.com...
Dominic Dumée
(95/98/2000/XP?).
By the way the file you posted at codecentral is an html file but I'm sure
whoever downloads it will figure out what to do with it to get the code
out... :)
Cheers
Dominic
Reinaldo Yañez
Thank you. In truth it's a pascal file, that you can
access going to the option "List Files" or "Download". As far I can tell,
this is the usual procedure used in Code Central to access a file.
Regards,
Reinaldo
"Dominic Dumée" <nospam_dominic@nospam_psas.co.za> wrote in message
news:403d...@newsgroups.borland.com...
Steven Green
Dominic Dumée
the program when using it. Had no problems with Win XP/2000 so far though.
Reinaldo Yañez
AV's on finalization ?? Or during program operation ??
The patch applies and stays in memory, and redirects the calls to
GetDiskFreeSpace to a new routine. Are you sure that your AV are produced
by my code ??
Are you using Delphi packages ?? I found a problem on finalization when
using packages and will post a new patch soon.
Regards,
Reinaldo
"Dominic Dumée" <nospam_dominic@nospam_psas.co.za> wrote in message
news:4048...@newsgroups.borland.com...
Reinaldo Yañez
corrected. I also posted a small DLL that makes the patch when loaded and
"unpatchs" when unloaded.
http://codecentral.borland.com/codecentral/ccWeb.exe/listing?id=21475
Best Regards,
Reinaldo
"Dominic Dumée" <nospam_dominic@nospam_psas.co.za> wrote in message
news:4048...@newsgroups.borland.com...
Steven Green
> Hi, I tested the unit on Win98 and did not see any problem yet.
a friend of mine who is an excellent Delphi programmer wrapped it into a little
exe, so I could try to test it generically.. works fine on XP/2000, but on my 98
machine, it first gives an error "you must use at least Win95 R2" then it bombs
out..
Reinaldo Yañez
Win 95 OSR2 is indeed required because of the use of function
GetDiskFreeSpaceEx, but Win98 also support this function...
From Windows API documentation (MSDN):
GetDiskFreeSpaceEx Function
a.. Windows 95: Requires OEM Service Release 2 (OSR2) or later.
a.. Windows 98: Supported.
a.. Windows NT: Requires Windows NT 4.0 or later.
a.. Windows 2000: Supported.
a.. Windows CE: Not Supported.
Best Regards,
Reinaldo
"Steven Green" <gre...@diamondsg.com> wrote in message
news:40492552...@diamondsg.com...
Steven Green
> What you report is really strange !
> Win 95 OSR2 is indeed required because of the use of function
> GetDiskFreeSpaceEx, but Win98 also support this function...
exactly.. so the real question is, why did it think my 98 was 95 R1?.. and then
AV when it decided it was wrong and/or because of that?