--
Nick Hodges
Delphi/C# Product Manager - Borland DTG
http://blogs.borland.com/nickhodges
> If the newest car were bigger and slower and less reliable, would you
> buy it? Don't be surprised if the same criteria applies to software.
I'm not surprised. Is that why you haven't upgraded?
Why I haven't upgraded:
There isn't an upgrade available for me to TurboDelphi from D6.
I worry about the expense of upgrading my extensive code library from 3rd
parties.
I am unsure about DevCo and/or Borland.
I have past resentment toward Borland.
What I have now works for Win32 and works well.
I have other tools for .NET and have no interest in VCL.NET and Delphi.NET.
"I am not representative of Borland's primary market."
-Johnnie
> I'm not surprised. Is that why you haven't upgraded?
Yes, mainly.
> "I am not representative of Borland's primary market."
Johnnie --
I'm not busting your chops here -- not at all, I really want to know.
Please forgive me ahead of time if this question is perceived as
hostile, as it isn't even remotely meant to be:
Do you think /one/ developer should be representative of a whole
market? Why does it bother you?
Thanks for all the disclaimers above. <g>
I read that line (or a variation) so many times in these newsgroups, and
every time I read it I was not pleased. I always took it as, "we don't need
the little guy anymore".
I wish you the best of luck to getting that notion out of everyone's head.
-Johnnie
If I have any of the above in the wrong order then its my age.
So why havent I upgraded coz Borland stinks and I cant wait until Devco
gets into harnass I will see if u can get me back but BORL I sold mine.
HTH
Rita
"Nick Hodges (Borland/DTG)" <nickh...@gmail.com> wrote in message
news:4524361f$1...@newsgroups.borland.com...
> Do you think one developer should be representative of a whole
> market? Why does it bother you?
The issue is that nobody should get that response from a TeamB member,
even if it is true.
When you are shopping and you make an opinion, the vendor never tells
you that you are irrelevant (unless he is a complete idiot).
> I wish you the best of luck to getting that notion out of everyone's
> head.
The little guy is /all we are focused on/ here at DTG. We don't sell
anything that doesn't, in some way, support the single developer
sitting in a single chair, wherever that chair may be.
I agree that shedding a lot of baggage will be tough. Sometimes a name
change can do that. And sometimes not. ;-)
To D8.
Decided that if that is what .NET offers, I don't want any part of it.
Am currently eagerly awaiting Delphi 64 native.
And waiting.
And waiting.
And waiting.
And waiting.
And waiting.
And waiting.
And waiting.
And waiting.
And getting bored.
And while waiting am having a real close look at Lazarus.
"Mat Ballard" <m...@chemwares.com> wrote in message
news:45244d0d$1...@newsgroups.borland.com...
> http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
Well, we've upgraded. But what would have kept us from upgrading (and
almost did) is when we can no longer develop our application. What I
mean is. When a new version comes out, we get one copy and load up our
projects and make sure they work. Sometimes there are minor compiler
changes, new warnings, deprecations etc. No problem. We make the
corresponding changes in our code. Basically, we check to make sure
all is well and nothing is majorly broken. Once that is proven, we do
the full update for all developers.
Now what almost kept us from upgrading (and did for D8 and D2005), was
the numerous problems we had with our projects. D8 first off didnt have
w32. D2005 had major problems with our projects. Then came D2006.
This version was almost a no go for one reason. We have some activeX
stuff and typelibs etc. We had some issues. But finally got it
working. Its balancing on a string though. We fear anytime we need to
get into that area of our code.
And finally, the other major thing that will halt an upgrade for us, is
the 3rd party ball and chain we have. Some vendors are much better at
getting their stuff upgraded. Others are not. So, we will not upgrade
until we can get our 3rd party stuff upgraded or removed.
hope this helps.
--
> Why aren't you upgrading?
Because you didn't offer an upgrade path to the product I was interested
in upgrading too. Which has spectacularly back-fired.
I started out not needing .net currently, and couldn't justify the price
tag of a BDS upgrade in order to enjoy only 25% of the languages it
supports (Delphi.32)
Never-the-less, hearing "good things" about BDS 2006 (vs the mess that
had been D8 and 2005) I downloaded and installed the BDS Trial, the
operation of which was ridiculous - of the 30 days granted to try it, I
only had 2 or 3 useful days. Overall the experience was merely a huge
waste of my broadband download data limit.
I did however discover FastMM as a result of the experience, and am now
happily using that as a matter of course in D7 and indeed D5 projects!
Turbo Delphi.32 Pro looked promising, especially given the announced
price point, giving rise to an expectation that an upgrade from D7 would
be significantly cheaper than the upgrade to BDS, and hence obviously
more attractive (especially given the stated re-juvenated interest in
caring about "the little guys").
The Turbo Explorer gave me time to play properly that the BDS Trial did
not.
Bottom line, I found very little to excite me. The Form Designer
Guidelines are neat, but not compelling. The Component Palette stinks
to high heaven, and overall in even the relatively short time I had
playing in the IDE I encountered errors and instability issues and poor
performance and so went running home to D7.
More importantly perhaps, the overall impression that I got from the
"Turbo" experience was that it reeked of slapping a new label on the
box, doing some clumsy (and incomplete) search and replace rebranding,
and hurriedly launching a product in a desperate attempt (yes,
"desperate" is how it seemed) to recover something from a hopelessly
messed up situation arising from mistakes of previous years.
So overall confidence in Borland and Delphi going forward was shaken.
In the meantime however, since I now had .net installed anyway, I
started doing some more digging into the options for .net development
and stumbled across Chrome.
Immediately excitement reached levels not felt since first playing with
Delphi 1.0 - all of the concerns and niggling doubts I had had every
time I had sat down intent on finding out how Delphi.net hung together
crystallised in my mind.
Where Delphi 1.0 was a revolution (and revelation), providing a
fantastic way to deliver Windows appications, Delphi.net was more
concerned with providing a vehicle for porting Win32 to .net, not a
great new vehicle for delivering new .net apps.
I wasn't looking to Delphi.net to see how I could keep doing the same
old things, I wanted to see how Delphi made .net more exciting and
easier/more fun/more powerful to work with than the alternatives.
Chrome on the other hand showed how Delphi could have been.
If nothing elses, Delphi.net is stuck in .net history (1.1 framework),
rather than being at the cutting edge, which in itself spake volumes
when considering Delphi for the long term (i.e. .net - for Win32 D7 was
keeping me more than satisfied).
The game plan that had evolved to this point HAD been:
- Stick with D7 for now and see what the Turbo Edition of
Highlander has to offer (assuming there is one), and maybe
by then I will have need of the .net stuff too, so an
upgrade to BDS will make sense if I still am not offered
an upgrade to Turbo.
But, having experienced Turbo and discovering Chrome, the gameplan NOW
is:
- Stick with D7 for Win32
- Move to Chrome for .net when necessary
I can do some useful work (i.e. primarily "learning") in Chrome for free
with the command line compiler, and when the time comes can purchase
Chrome with VS (bizarrely I can take advantage of upgrade pricing with
Chrome, yet not with Borland!) for E199.
Note that even /without/ the competitive upgrade pricing, a wholly new
Chrome & Visual Studio license would cost me only 249 Euro. Compared to
384 Euros for the Turbos (using the UK online shop site as price
reference, rahter than the perhaps more apt NZ/Oz size - don't ask me
why, there was no particular reason for that it's just how it ended up)
+0.02
--
Jolyon Smith
imho This statement /really/ doesn't sit well alongside the decision not
to offer upgrade pricing to Turbo from D7 (etc).
Too late for me mind you (see separate post).
Had there been an upgrade at a "no brainer" price I would almost
certainly have stumped up without thinking. But I had to stop to think
and that's where the wheels fell off the "Jolyon as a Borland/DTG
Customer" wagon.
:(
--
Jolyon Smith
No-one thinks that "irrelevant", and "not representative of the *entire* market" are
the same thing (unless he is a complete idiot).
--
Dan
> I did however discover FastMM as a result of the experience, and am now
> happily using that as a matter of course in D7 and indeed D5 projects!
have you also discovered:
GExperts - especially the well-hidden little gem of multi-line component tabs ?
DelphiSpeedup - which inserts FastMM into Delphi, inter alia.
DDevExtensions - by the same Andreas Hausladen, which has the nifty Component
Searcher.
cheers,
Mat
> No-one thinks that "irrelevant", and "not representative of the
> entire market" are the same thing.
Of course they aren't, but when someone gives that response he is
implying that.
I guess I probably shouldn't be answering this question...
I have upgraded. :-)
And I'm patiently waiting for Highlander. I am really looking forward
to generics support. And I will again upgrade when Highlander comes out
or shortly after that. I'm hoping that is at the latest early 2007
because I need BDS 2007 now.
Cheers,
Kevin.
>http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
>
I'm interested in native code generation and a snappy development
environment - D6 fits these requirements better than BDS. I realize
there are some benefits and improvements in BDS, but, I guess, it's
just not as fun to work with a fat tool. If D64 comes bundled with a
fat IDE, I probably won't upgrade to that either.
If it is possible to make a win32/64 'lite' product (not lite on core
functionality or licensing, though) I think you'd have a lot of happy
customers. You could still sell a 'kitchen sink' version to those who
want it :)
Thanks for asking.
-ckd
I was keen on learning C# when I saw Chrome. It seemed to have the best of
Pascal and C#. I really like it. I am a licensed user, but have not done any
real development in Chrome yet. I have only gone as far as a few C# ASP.NET
applications in .NET so far. Chrome seems to be able to keep up with the
.NET framework too, which is more than a "nice to have" feature.
-Johnnie
However, it might be better to stay with BDS 2006 and get the
2006-compatibile components, because who knows when there will be
2007-compatible components?
Nick Hodges (Borland/DTG) wrote:
> http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
--
Well, we have yet to see, since the name change has not yet happened and now
the rumor has it that it may never happen. It's a tough position to be in,
you want to announce things ahead of time to keep the buzz in the market
place, but if you can't deliver on these announcements, that buzz starts to
become negative :(
-BKN
I know I shouldn't say anything, but yet here's another one...
-BKN
So I have to ask, why are you visiting these newsgroups? Is there a glimmer
of hope or something?
-BKN
Would you consider upgrading to a non-dotnet version? (i.e. Turbo D)
-BKN
I've seen a few posters state that if they go back to D7 for an old
maintenance app, they get very irritated by the lack of this feature or that
feature. By zeroing in on what makes D2006 *better* is what the marketing
department should be focusing more on. If there are compelling reasons to
use the app, then users will sometimes ignore the reasons to not use the
app. The reasons why people aren't moving to D2006 could be something for
the dev team to focus on for hotfixes, "new" features, etc.
-BKN
> GExperts - especially the well-hidden little gem of multi-line component tabs ?
Have been using GE for many years now.
> DelphiSpeedup - which inserts FastMM into Delphi, inter alia.
Erm, FastMM is already part of my Delphi landscape (5 thru 7), design-
time and runtime.
If DelphiSpeedUp doesn't make the BDS/Turbo IDE /faster/ than the D7 IDE
I fail to see how it could be seen as anything other than a sticking
plaster solution to a fundamental problem that is more easily solved by,
erm, not inviting the problem to the party in the first place.
> DDevExtensions - by the same Andreas Hausladen, which has the nifty Component
> Searcher.
No. Point: I shouldn't have to find (and/or wait for) 3rd party
solutions that just about make a new IDE as useful/pleasant to use as an
older one.
Pay for the privilege of having to put in some effort to get back to
where I was before?
No spank you very much.
;D
--
Jolyon Smith
Delphi 1
Delphi 2
Delphi 5
Delphi 7
The upgrade to Delphi 7 was driven by the need to get a second developer
license.
I'm only doing Win32 development, so all the .NET fuss is of no need to me.
As far as I remember, there was no big benefit of moving from D5 to D7.
What could make me upgrade?
Good unicode support.
x64 compiler
Regards
Meikel
Rita
"Bryce K. Nielsen" <br...@sysonyx.com> wrote in message
news:45247006$1...@newsgroups.borland.com...
Another reason I don't upgrade is because of quality issues. Until
recently you had to purchase an annual copy of Delphi to get bug fixes.
I've seen so many people left out in the cold because of this. For
example, "I can't get Delphi 2005 to work" and "Well you need to buy
Delphi 2006". I don't feel like Delphi is supported like other
applications.
I feel Borland has thumbed their nose at customers and the community. I
almost bet the farm on Kylix and was left out in the cold. You've got
guys like Andreas Hauslauden(sp?) who are stellar community members but
ignored.
The final reason is we were promised a regime change with the DevCo
sale/spinoff.
Advantages are:
the line up tools in the forms editor are a great help
tools pallette is cool
improved editing features are good, especially with the castalia addon
Cbuilder compiler faster
"Jolyon Smith" <jsm...@deltics.co.newzealand> wrote in message
news:MPG.1f8f19a95...@newsgroups.borland.com...
.Net side belongs to Visual Studio (sorry), D8 getting broken by a .Net service
pack contributed greatly to scaring all .Net development away from Borland in a
rather definitive fashion.
D2005 scared Win32 developers away from BDS.
D2006 trial wasn't convincing, critical drawbacks like the help system more than
negated all the positive impressions.
Turbo 2006 was the way to reintroduce D2006 here, but if more trials were made,
the IDE still feels "shaky" despite the hotfixes: flickering and sluggishness,
glitchy behaviours in undocked mode, occasional unreproducible crashes, dodgy
component palette, modeling functionality untrustworthy. It's definetely better
than D2005, but doesn't feel solid/comfortable compared to D7.
It all boils down to D7 still doing the job, and our D7 being souped up with
experts and wizards, the built-in BDS improvements aren't such a big leap ahead
(it's a leap back in some area, since we don't have all our experts to D2006).
We have started relooking/testing our apps for Vista, D7 doing fine enough so
far, so no pressure (leap from D5 to D7 was to get WinXP fixes here).
Unicode, Win64, improved multithreading programming paradigms/support would
likely get us moving to a new version in short time by providing added value
that D7 couldn't compete with (assuming critical issues like help have been
taken care of, and quality is at least BDS2006+hotfix level at release time).
However, as time passes, the likelyness of us moving to the next version doesn't
increase, it only increases the amount of code that gets done in VS.
Eric
I would upgrade if Borland would...
- sell a IDE which is as fast and stable as D7
- sell a IDE with a better Project Group Manager.
- sell a IDE with a TreeView in the Component Palette.
- sell a IDE with a Memory-Leak checker like FastMM included.
- sell a IDE which supports the build process for Develop,Testing and
Release.
- sell a IDE with Unicode for Win32.
- release the Bold Source to opensource.
Just one hint for Borland.
Why do people start writing tools in their spare time ? (e.g.
FastMM,GpProfile,GExperts,Sourcecode-Formater,DelphiPackageTool,UnusedUnits,
MadExcept,ExceptionMagic)
Because there a still things which could be improved!
"Nick Hodges (Borland/DTG)" <nickh...@gmail.com> wrote in message
news:4524361f$1...@newsgroups.borland.com...
> http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
I've been waiting especially for Delphi to support .Net 2.0. Since Delphi
2006 doesn't support .Net 2.0, I'm prepared to wait, seeing as our product
is still developed with Delphi 7 - although I bought it, I'm not using
Delphi 2005 for many reasons. We've targetted the next Delphi version as
our next version to upgrade to.
The rest of our organisation is Microsoft and has gone down the .Net 2.0
path. If Delphi 2006 had have supported .Net 2.0, we would have upgraded to
it.
a) I don't want anyone else than DTG to have my money, like the ALM-Borland
b) The addititional uncertainty about the future of Delphi posed by
announcing a spin-out without having a buyer (shouldn't I play with C#
instead on these pet-projects, so that, in the unlikly case I don't get
silly rich from making them, I have at least gained some new skills):
I work for a small ISV, and we havent upgraded since Delphi 6. In this case
I don't have to care about additional uncertainty posed by the spin-out; we
only make one application, and it's going to stay on native Delphi for a
long time. In this case I'm waiting because:
a) Given the size of the application we make and the number of third-party
components we use, upgrading will make our developer team unproductive for
about a month. So, we can't upgrade more often than, say, every third year,
and hence, when we do upgrade, it has to be a really good Delphi-version. We
can't spend a month just to find out we can't use the upgrade, so we have to
go by reputation, and Delphi 2006's reputation is still not convincing me.
b) I guess there will be changes to the VCL to make it work well on Vista,
similar to the changes made for XP in Delphi 7. If we upgrade to Delphi 2006
now, we won't have time to upgrade again until maybe 2009. So, we will
probably be much better of by waiting for the first version after Vista is
released.
On the server side, we are more and more moving to Linux.
That's why we are having a thorough look at Qt right now.
I would upgrade, if Linux/OSX was on the roadmap again. As long as you guys
keep running after MS I'm not going to upgrade again.
we did, but from D5 to FPC/Lazarus.
Regards, Bernd.
Delphi 5 Enterprise to Delphi 8 Architect
Kylix 2 to Kylix 3
More questions?
> Because there a still things which could be improved!
In fact, someone will probably always think of something else to add,
regardless of how much Borland adds <g>
--
Dave Nottage [TeamB]
Not sure what that's got to do with it. You appear to be implying that
Johniie is somehow in a small minority. Otherwise I can't make heads or
tales of your question.
I am convinced from using Delphi since version 1 as a professional
consultant out in the market place, that Delphi/Win32 is what the
overwhelming majority of Delphi developers are using. This places Johnnie
firmly in the mainstream, and Johnnies' sentiments are pretty much the same
story I hear from all the Delphi developers I meet. Except for here in this
newsgroup, I have yet to stumble across a Delphi for .NET developer out in
"the field".
What I see now is the same strategy from Borland/Devco since D8. Which is
the plan to herd the existing customer base into .NET; and that its only a
matter of time until that happens. The only problem with the strategy is
that you run the risk of aggressively herding them towards Visual Studio,
(especially if the strategy is executed badly) which is exactly what has
been happening.
-d
> You've got guys like Andreas Hauslauden(sp?) who are stellar
> community members but ignored.
I can say that Andreas and his stuff is most definitely /not/ being
ignored.
I appreciate your other comments.
> Then the whole NET thing phewwwww Sun already did that with Java and
> MS couldnt get its own way, so out comes NET the next big thing, yea
> right.
Yep, software dev tools history in a nutshell...
-d
One of the reasons I can think of is that if you're a small development
company you just don't have the time to spend days of valuable time just
"modifying" sources to make sure it works with a new version. Why doesnt
Borland write some kind of parser which upgrades your sourcecode to be
compatible? Or at least make a change-suggestion. I'm quite sure Borland has
a lot more employees than most of their customers.
Another thing is: if I was a non Delphi-developer looking to move to a
development tool, I'd be looking for the support of those tools. I'd be
looking into what the people in the newsgroups talk about, how are bugs
managed and resolved, how many updates are released... Ofcourse you have to
keep in mind there's always whine from people. But quite frankly, you'd be
shocked if you look into Delphi. Borland keeps releasing new versions
instead of releasing a few upgrades which resolve bugs. Have you ever run a
query like "select count(*) from table_bugs where bug_status = unresolved"
on Borland's bug management database? A small example: we've made a
bugreport about a bug which set your form to binary when you opened it using
the .dfm file (and then F12'ed to get the form) instead of the .pas file.
This was reported while working with D3. This bug was resolved in BDS 2005.
For us - and many others - this was an annoying bug as a lot of people use
Sourcesafe to maintain source history, but Sourcesafe was unable to store
binary files. Why wasn't this bug resolved in a product update? or in Delphi
4,5,6 or 7? why does it take 5 versions before a bug like this is fixed? I
think Borland/DevCo needs to focus on making the product better, not bigger
and more unstable.
I would pay £100 for the upgrade just to feel like I helped you guys out.
> Why aren't you upgrading?
To (mis)quote Steve Ballmer:
"Help system, Help system, Help system, Help system, Help system, Help
system, Help system, Help system, Help system, Help system, Help system,
Help system, Help system, Help system, Help system, Help system, Help
system, Help system, ...."
Etc... :-)
Dave
> The only problem with the strategy is that you run the risk of
> aggressively herding them towards Visual Studio, (especially if the
> strategy is executed
> badly) which is exactly what has been happening.
I suspect there's a lot of truth in that. Over the last 18 months or so,
Borland has done far more to advance the cause of Visual Studio than
Microsoft marketing could ever do, and it hasn't cost Microsoft a cent.
Dave
Absolutely. The more I think about it - DTG might consider literally
dropping .NET. They are so far behind and with WPF, WCF etc coming out,
I dont see how they will have the resources to even catch that, let
alone what else they have to do.
Maybe the strategy should be:
-Dump winforms and all .NET designers.
-Allow for export of managed interfaces so Delphi code can be given to
other .NET users.
-Allow easy import of .NET assemblies and use from unmanaged code.
-Focus on native code compilation
-Extend compiler with new features, such as generics (native) and so on.
-Get a lean and mean IDE again. Features, but not a pig, and should not
require user or web tweaks to make it run reasonable.
--
"Programming is an art form that fights back"
http://www.KudzuWorld.com/
Need a professional technical speaker at your event? See www.woo-hoo.net
At least among the Borland users, yes.
I predicted this would happen over 2 years ago, and I was attacked and
reviled by TeamB and their hangers-on.
-d
> The Component Palette stinks to high heaven
Andreas Hausladen has written a replacement component selector which you can
find here:
http://andy.jgknet.de/dspeedup/index.php?page=DDevExtensions
It takes up almost no screen space and is extremely fast. It works fine
with my ....<cough>... slightly enhanced version of Turbo Delphi Explorer
and should work fine with the Pro versions. According to the documentation,
it will even restore the old-style Component Palette if used with BDS 2006
(i.e. not the Turbo stuff), although I haven't tried this out.
> Chrome on the other hand showed how Delphi could have been.
For some time, I argued that if Borland were going to get into .NET, then a
fully paid-up VSIP add-in for VS.NET (what Chrome is) would be the way to do
it. This would have had several advantages:
1) Would have enabled Borland to leverage the huge (relatively speaking)
number of VS.NET developers out there by - for example - providing a
time-limited trial of the add-in.
2) Would have eliminated the need to sink huge amounts of resources into
developing a completely new IDE - i.e. BDS. With the time and money saved,
we might even have would up with a decent help system. ;-)
3) Would have provided a very strong differentiation between Borland's Win32
products on the one hand (successors to D7 using the "classic" IDE) and
their .NET tools on the other hand.
4) Would have eliminated all the whinging about needing to install .NET in
order to do Win32 development. ;-)
5) Would have enabled Borland to pursue new markets (.NET development)
without pissing off the existing core user base (Win32 developers).
Dave
> At least among the Borland users, yes.
Yes - that's what I meant. Sorry to be unclear.
Dave
> "Dave Jewell" <Dave....@btclick.com> wrote in message
> > I suspect there's a lot of truth in that. Over the last 18 months
> > or so, Borland has done far more to advance the cause of Visual
> > Studio than Microsoft marketing could ever do, and it hasn't cost
> > Microsoft a cent.
>
> I predicted this would happen over 2 years ago, and I was attacked
> and reviled by TeamB and their hangers-on.
It started longer ago than that. Any discussion questioning the roadmap
and choices where aggressively attacked by certain TeamB (not here
anymore for some reason) and classified as "Borland bashing" and "TeamB
bashing", FUD and so on. Messages were even cancelled.
While Dave Jewell certainly is right, myself I found out without any
"help" that I had to change development tool for various tasks,
although I still use Delphi a lot for maintaining older projects, and
for some new projects in areas where Delphi excels.
--
Ingvar Nilsen
http://www.ingvarius.com
> I predicted this would happen over 2 years ago, and I was attacked and
> reviled by TeamB and their hangers-on.
BTDTGTTS. ;-)
Dave
Please record this and upload it to YouTUBE :)
> - because the task like "give a project to another
> developer/workplace" is still not solved.
I'm not sure what you mean here.
--
Regards,
Bruce McGee
Glooscap Software
Some reasons, in no particular order.
1. D2006 was much slower than D7.
2. Some 3rd party vendors (Digital metaphors in that case) saw an
occasion to charge more for an upgrade to a version of their components
that would work with D2006, adding to the cost.
3. No compact framework. The only .Net project I have for now is CF and
is done with C#/VS2003.
4. Tested ECO, under D2005 and D2006. Great technology, a steep learning
curve, close to zero documentation or examples. Plus, being a .NET
technology, you need to know the platform, wich adds to the learning effort.
5. The help system is a huge step backwards. I could live with a lousy
help system when working with a tool/framework/platform I know well. But
it's too frustrating to try to get some help for something you don't
have a clue about and not being able to.
6. I've bought every new delphi version from 1.0 automatically. It was
always worth the money. With D2005, for the first time I had the feeling
that I had thrown the money down the drain. I took some time to evaluate
BDS2006 and will do the same with BDS2007.
7. The IDE was slow, a memory hog and buggy as hell.
8. Together modelling support was slow, buggy and worse than ModelMaker.
9. For Win32, D7 works very well, and Win32 is what I do now.
Regards,
Gerard.
OMG Rita used the "S" word and surely the wrath of Zeus will unleash a bolt
of anger that will strike her down. ;-)
Yeah your right. I just googled myself in 2003 and that was really the start
of the "conversation".
-d
I am a little sad with Delphi 2006 after using for 3-4 months. I upgraded my
P4 computer to 1G ram for Delphi 2006.
1. IDE is too fat, too slow, each time load the things I do not need. (Rave,
Component One, IntraWeb etc)
2. Form designer is very slow. Switching from one property to another is
very slow.
3.Toolbar problem: (In a real project) When I click on a button on a
toolbar, the toolbar will be blank for 1 second and lock the mouse cursor,
then back to normal.
4. Structure view: it will collapse all the expanded node when you click on
a node, in fact I just want to jump to that source code.
5. Bad Help Function: can not find the Help about one component or property
in context.
6. FORCE the user use the fixed embed layout by default. not friendly.
7. No IDE Layout style options.
8. So few new VCL components, not make the developer's job easier, why
upgrade?
9. When editing code, the cursor position is often not correct with
Chinese comment when typing Chinese using MS Chinese IME.
10. ECO is too complex to understand and maintain so it seems it is no use
to me
11. DataSnap's Unicode support can not work well with nText field in MS SQL
server.
12. No Unicode VCL(but have WideStrings.pas). (have to use TNT VCL)
13. Still no code generation function. I think it should be a "must have"
function in a modern IDE. Why do we have to write the code that could be
generated from db or text?
14. StringReplace is faster than Delphi 7 since it come from FastCode
project. We need more functions from FastCode project in Delphi and please
put them in HELP system.
15. Each time when I compile my project I will see most of the IDE area
(80%) are white screen, It seem that there is no CPU resource to paint the
IDE windows.
16. it seems that QC does not work.
Sorry, I wrote so many small problems I meet with Delphi since I use it
everyday.
It is the first time I have to do so.
Bear
"Nick Hodges (Borland/DTG)" <nickh...@gmail.com>
wrote:4524361f$1...@newsgroups.borland.com...
> http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
> Yeah your right. I just googled myself in 2003 and that was really the
> start of the "conversation".
>
> -d
But a little overy 2 years ago I was saying this:
"All dressed up with no place to go": http://tinyurl.com/jx6x6
<quote>
... I mean the particular case of herding an entrenched customer base from a
rich, robust, natively compiled code base to an entirely new framework with
no particular advantage on Win32. In the face of clear alternatives. I
think it would have been far easier to simply upgrade the customer base to
64-bit since its
natively compiled; and the upgrade path was crystal clear. Then the .NET
stuff would logically come next. As it stands, there is no compelling
reason to leave the Win32 code base at all. As it stands, Delphi.NET is all
dressed up with no place to go. Longhorn is years away.... This is one of
the worst cases of timing I have ever seen. Yeah, .NET is the next windows
API..... ON LONGHORN.... In the mean time we could be honing our tools on
64-bit (in an alternate universe)
</quote>
Of course the "next windows API on Longhorn" was mere bogosity that I quoted
without believing. We had plenty of lively threads debunking that myth as
well.
-d
> Why aren't you upgrading?
D7 does what I need for Win32
VS 2005 does what I need for web development
I have analysed my time consumption and found out that apart from this
news group <g> I spend most time figering out what to program, not
programming itself. IOW, design and planning and documentation consumes
most time.
Nifty add-ins like refactoring and (what it is called) is not
important. I am a carpenter who use a hammer instead of a nail-gun
because I also am an architect and buying a nail-gun would not justify
the investment.
For Delphi, this would be decisive:
Unicode support
64 bit support
BDS 2006 is a super tool, source code management/history and all, very
very nice to have. But not must have.
To sum it up, I upgrade when Delphi comes with must-have features.
Hmm, more than a few of this where predicting that this
would be the case right from the first sighting of D8...
<sigh>
> Maybe the strategy should be:
> <snip excellent suggestions>
I can only agree 100% - and when .NET advocates and
native advocates can agree on this, it must be a sign
of the coming apocalypse :-)
> For some time, I argued that if Borland were going to get into .NET,
> then a fully paid-up VSIP add-in for VS.NET (what Chrome is) would be
> the way to do it.
If you have noticed, I have referred to this on several occasions. I
think you aired this opinion already in 2001/2002, was it Computer
Shopper Magazine or PC Pro?
> This would have had several advantages:
Agreed, I mentioned this too yesterday in another post.
You telling me that Dave can Monkey Dance? ;-)
> Of course the "next windows API on Longhorn" was mere bogosity that I
> quoted without believing. We had plenty of lively threads debunking that
> myth as well.
I remember all that stuff. Some people genuinely believed that Longhorn was
going to be written in managed code with the ability to support legacy x86
code apps through emulation. LOL.
Dave
This may well be water over the damn since IIRC Microsoft will not allow
someone to use VSIP if they market a competing IDE.
> If you have noticed, I have referred to this on several occasions. I
> think you aired this opinion already in 2001/2002, was it Computer
> Shopper Magazine or PC Pro?
Not sure it was that long ago, but quite a while. I seem to remember there
were legal issues with Borland going down the VSIP route...
Dave
> Please record this and upload it to YouTUBE :)
Mr. Ballmer is flying out here even as we speak... :-)
Dave
My explanation for your observation: The Delphi for .NET developers that
seems to be in this newsgroup are actually Delphi Win32 users who use a
different language for .NET development, but would like to use Delphi for
.NET if it was done better. There messages here, nice or not, are about
motivating Borland to improve their .NET strategy.
Rick
> http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
i have "upgraded", but it pains me to say that the biggest problem at
this point still is stability :(. This may be an unpopular opinion, and
it may be an isolated case, but i am these days trying to use BDS
exclusively for the development of one of our new Delphi library
products (Hydra 3), and stability is a huge problem for me - from
crashes, over long hangs to issues such as the IDE not picking up code
changes when i press F9 and running an old .exe (i need to manually
Ctrl+F9 to build).
For example, i have not managed literally a *single* time, to shut down
BDS properly, without getting a series of AV dialogs or the BDS.exe
process lingering around and having to be killed in task manager (or
worse, me finding later that i still have 5 of them running). Sure -
unclean shutdown is not a showstopper - but it does not breed confidence
in the stability of the product, either.
I'm also seeing bad resource (gdi objects and user handles) leakage.
after having BDS open for a longer period of time, i notice that other
"hungry" applications (such as for example Photoshop) will refuse to
start - until i shut BDS down. This is on a 4GB-RAM system. And task
manager confirms that BDS.exe is at the top, being the only application
with 4-digit numbers of handles in use - chased only by a few others in
the low-hundreds.
These are the easy-to-pinpoint, but less annoying problems. Bigger
problems are the debugger crashing with C++ assertions regularly - not
every time, not even in specific scenarios - but often enough to be a
hassle. Or the IDE simply locking up and having to be killed in task
manager (losing code in the progress, if i didn't have an F2
trigger-happy finger).
All of this happens, for me, without any third party packages installed
(not even our own).
I don't mean to rant, but you did ask why people are hesitant about
upgrading to BDS2006, and these issues i mentioned are issues that
frustrate me *every* single time i use BDS. :-(
Why do i keep using it? Good question. i suppose it's a mixture of (a)
me actually liking some of the IDE advancements over Delphi 7 (so no, i
am not bending over backwards to make BDS bad, i am *trying* to like
it), (b) a case of eating dog food - if our customers will be using it,
it seems like a good idea i try to use it myself as well, rather then
staying in the comfort zone of D7 and (c) some subconscious hope that
maybe /today/ it will work stable.
hth,
--
marc hoffman
Chief Architect
RemObjects Software
http://www.remobjects.com
and the fifty-two daughters of the revolution
turn the gold to chrome
A fair amount, maybe a majority, are probably rather about motivating
Borland to provide an alternative, superior development environment that
doesn't involve the tie-in to MS dev tools like .Net does.
In other words, if there was a Delphi compiler and VCL for ARM, we
wouldn't need no .Net... just like VCL meant we had no need for MFCs.
And there is no opposition between using an MS OS, and not wanting to
depend on MS dev tools: when relying MS dev tools, your course of action
is purely dictated by MS, the good, the bad, the ugly, you don't have a
choice. But when relying on MS OS, MS actions are constrained by the
needs of supporting existing and past software, giving you a much more
stable environment for your applications to live in.
Eric
Just in case you don't know, you do can use Qt 4 with Pascal right now:
http://www.freepascal.org/wiki/index.php/Qt4_binding
http://users.pandora.be/Jan.Van.hijfte/qtforfpc/fpcqt4.html
And it works for Windows, Unixes and Mac OS X
And the bindings are open source, so you never need to worry about it
becoming obsolete. They are also well maintained.
There is a unit that makes the bindings trully object-oriented. It's
called qtprivate.pas and available on the Qt interface of Lazarus.
Agreed !!! All of your suggestions would make a new version of Delphi a
true upgrade. Unless Delphi shifts it's focus back to native code, I
can't see a reason to upgrade, as far as .NET goes, I thought we'd be
using it by now, but haven't found a compelling reason to develop no
native applications, however if Delphi allowed easy of importing .NET
assemblies to WIN32 unmanaged code we would be using them right now.
Eric
It would not compete, since you, to use it, would need to purchase VS
in any case, on the contrary, it would increase VS sales.
It cost a whole bunch of money, much money for mere mortals, not for
Borland :)
Thanks for the question.
First I guess most who answered here care about Delphi. I do.
1. Personally the cost does not warrant the new features.
2. I am angry with Borland for abandoning Kylix developers.
3. Borland has earned a negative reputation for fixing bugs and
delivering half baked solutions.
4. Borland keeps changing their direction/focus. xplatform then .Net now
win32/win64 again. There should have been different ide's as different
products and all regularly maintained. (Delphi language as core)
5. Whatever happened to the Kylix community project?
I think the developer market is tough at the moment - corporates only
upgrade if they have to and the rest when they can afford it. Where I
work we are currently on D7 and are waiting for Delphi unicode support
before upgrading again.
I would hope most Delphi developers would like to see their project
deployable on other platforms. There was a lot of buzz during the Kylix
era. Most likely most 'developers' did not directly see a need to buy
Kylix at the time, but importantly it made Delphi more strategic in the
sense that you could migrate your current applications to Linux if need
be. And Linux is even much HOTTER now. I fail to understand how you can
not see this. Kylix was more a long term tool a few years ago it should
have been ready now.. if only..
Personally I would only pay for a new Kylix version if it comes out I am
happy with D7 on windows at the moment.
Listen to your users and support the ones that help you.
siegfried
Your customers are asking for
- Win64
- Unicode
- Linux
- VCL/RTL improvements
- A working help system
- A stable IDE
For some of those things, the customers are asking for it for YEARS now.
Instead you keep investing into things your customers are NOT asking for.
You keep focussing only on .NET while the big majority of your customers
and 3rd party tool partners want native development instead. You know
this from the feedback in these newsgroups, you know it from the feedback
in your Delphi hours, you know it from your sales figures, and right now you
can easily see it from the Turbo download counters. No matter which Delphi
website, community or forum website you go to, in real life those customers
who haven't left you for Visual Studio do Win32/VCL.
The whole question is silly, too. You KNOW all of this, as you are ignoring
it all the time. "We'll put it on the roadmap, it's on the roadmap for the
year
2030, this is very important to us" yadayadayada. Just finally DO it.
Finally get your act together and do what your paying customers are asking
you
for. Problem solved. Obviously you can't do this with the few people left
in the Delphi development team. So, get the spin-off done, get a huge group
of skilled developers employed so you finally are able to catch up again
with your developers needs instead of falling behind more and more each
day.
Simon
Ok, reading your post again I.P. I see what you mean. Hm..
Well said!
Mark
I was one of that who loss me money on D8. Nice box, just that.
And STILL cannot upgrade from D8 and pre debacle versions, to Turbo
delphi win32. WHY?
--
Donald Shimoda.
And after disabling most of the new features and using a few of the 3rd
party tools like SpeedUp it is a pretty good alternative to D7 for Delphi
win32, except for the help system ofcourse.
But basicly the only thing we got for this hassle if the new 'for'
statement, the documentation xml and 'strict private'. hardly worth it.
And for the doucmentation xml we needed to write our own tools to make it
usefull too.
I would not advise an upgrade to the next version unless it would fix a
serious amount of the bugs currently listed on QC (personally added 14
allready), or at the least adds generics and unicode/utf8 strings and use
these through out the rtl/vcl. (adding a collections framework would not
hurt either).
We rarely use more then 2G ram in our apps but we are getting closer so we
would need 64bits too if the current trend continues.
grt, Bart
So,
- What upgrade would you recommend ?
- How much does it cost ?
- What will my cash buy me ?
With versions up to 7, Borland made it reasonably clear to me why I should
part with my cash; they wrote to me and put forward their case. Mostly I
bit the bullet strait away, but occasionally I put it off for a while
(though I kept the blurb, and took it out of its envelope every so often to
see if I could be now be persuaded).
Peter Kane
> The whole question is silly, too. You KNOW all of this, as you are
> ignoring it all the time. "We'll put it on the roadmap, it's on the
> roadmap for the year 2030, this is very important to us"
> yadayadayada. Just finally DO it.
Strongly Agree. Nick, just postule that post to the sensitive post
please.
--
Donald Shimoda.
9. For Win32, D7 works very well, and Win32
> is what I do now.
Just imagine D7 IDE and HELP system with the improvement in teh
compiler and RTL of D2006...
--
Donald Shimoda.
G.
> http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
Hi,
my upgrade history:
D2
(some years spent doing other things)
D7
BDS2005
BDS2006
I really enjoyed D2 and D7. Did not go for D8 because I thought the
.NET world was not mature enough and IMHO still isn't, but this is just
my opinion. Until Update 2 came out for BDS2005 I could not really use
it, although I had much less problems than the majority of people. Now
I use BDS2006 and I am very happy with it.
My point is that, as stated by other people, paying a lot of money for
a buggy upgrade does not help in them upgrading further. While I had
luck and did not encounter too much problems in BDS2005 (after the
updates) I can sincerely understand that other people are angry about
D8 and BDS2005. What does help having all those new features and/or
components and languages if they don't work properly? Or if the IDE
just hangs or closes?
Furthermore, why do they have to pay for 4 languages when they want
only 1 or 2? Please don't misunderstand me, I *do* think that putting
D, D.NET, C# and C++ into the same IDE is a great idea. I , for my
part, I love it. But I also think you should have given the people the
option to choose whether they wanted it or not.
The Turbos are a good step in this direction, but I would suggest you
to consider the possibility of making it even more modular. AFAIK the
BDS IDE can be used to host any number of languages, right? Then why
not make possible to:
- start with 1 to 4 languages a-la-carte. Maybe I would choose D and
D.NET, another one would choose C#, another D, D.NET and C#, etc.
- give the possibility to upgrade number of languages. Say, for
example, I paid for Turbo Delphi Pro, and I now want to get into
Delphi.NET but do not wish to use neither C++ nor C#. You could offer
me to purchase a Delphi.NET language add-on.
Not sure if this is so or not, so just in case I'll say it: allow the
people to upgrade to whatever they want from whatever they have at the
same cost per modul, i.e., if I would upgrade from D7 Pro to TD2006 Pro
let me pay x. If I want to upgrade to BDS2006 Pro with 4 languages let
me pay 4*x, but if I want to upgrade jus to 2 languages then let me pay
2*x. This is just an example of *clear* pricing.
Now, that *is* flexibility in both cost and power and I'm sure it's
something a lot of people would happily consider as an additional
reason to upgrade.
Another point I would like to speak of is the IDE language. Nowadays
Borland/DTG offers their tools in English, French, Japanese and German.
What is the reason for having 4 language IDEs? We could discuss a lot
about it, but in the end it's just one: comfort. Comfort which makes
the programmer more productive as he/she does not need to think in
another language that is not his own.
While personally I have no problem to use an English or German IDE
there are a *lot* of people who would surely consider to buy an IDE if
it would be in their own language. I don't know if it's made like this,
but I would suggest to change the IDE to somehow use "language
packages", so that changing languages is only a matter of a few
minutes. I would then either (if it's possible) make the "language
package" structure public so that anyone can modify it, or search for
collaborators to help you translate the language package to other
languages.
If this is not possible then I would consider to add versions in the
major languages that are not present: Spanish, Chinese, Portuguese,
Hindi (I think this is the major language in India), etc.
I really think focusing more (but not only) on these 3 points (IDE
stability, flexibility in cost and power, and languages), plus of
course Unicode, Win64 and .NET 2.0 would help you boost your sales.
Just my 2 cents... although a little long ;-)
--
Best regards :)
Guillem Vicens Meier
Dep. Informatica Green Service S.A.
www.clubgreenoasis.com
--
Contribute to the Indy Docs project: http://docs.indyproject.org
--
In order to contact me remove the -nospam
It was madness to spend all that time developing BDS, because all they've
done is produce a VS look-alike that's slightly more clunky and slightly
more buggy. What was the point?
I urge DevCo to be brave and do something really radical:
1/ STOP further development of BDS, and ABANDON your roadmap.
2/ Go back to your D7 code base and develop it until it's the best W32
development platform in the world. For instance, refactoring and code
rollup, to name just two things that have appeared since D7, are wonderful.
If you need to use some .NET stuff under the hood to do that, that's fine -
I don't need to know about it. I don't care if you write it in Forth, so
long as it works great. You've got a product life of a good ten years,
according to all the professional estimates, with puts you in a GOOD
position. Not many vendors of anything can hope for that.
3/ Take Dave's advice and fork off a completely separate product stream to
develop "a fully paid-up VSIP add-in for VS.NET." Gird your loins for a
competition with RemObjects (the Chrome people), because that's a fight you
could win (or at least arrive at a mutually beneficial draw).
4/ Read Dave's contribution again.
As Dave says, that way you'll have two clearly differentiated product
streams, with none of this dreadful messy coupling between them you've got
at the moment. Your W32 customers will be delighted, and you stand a good
chance of attracting a host of new customers who would appreciate a fresh
new contribution to the VS .NET environment.
If I promise to shower you with gifts and buy you gallons of the best UK
cask-conditioned ale, could you be persuaded? I'm not above a bit of
bribery... ;-)
Thack
The roadmap(s) say Win64 and Unicode VCL will be in Highlander, which
will be released in early 2007:
http://bdn.borland.com/article/0,1410,33383,00.html#4DelphiHighlander
http://bdn.borland.com/article/33519
Can anyone at Borland/DevCo/etc. confirm this is still the plan or are
those roadmaps out-of-date? Sorry if this has been asked and answered in
other posts, but there aren't enough hours in the day to read all the
posts in this group :)
Thanks
Thanks, I'll check it it out. Qt feels like a better investment than Delphi
right now.
Lazarus seems to be coming along nicely too. It looks kind of ugly though...
--
Pete
Blessed are the geek, for they shall public class GeekEarth : Earth {}
====
Audio compression components, DIB graphics controls, ECO extensions,
FastStrings
http://www.droopyeyes.com
====
Slow, buggy, Have 6000 lines of code and have to reboot every 1.5 hours.
Main reasons for not recommending and probably never upgrading again.
No longer a fast nibble native code development tool
No innovations in the VCL for over a decade
No 64bit compiler
No UNICODE
dbExpress continues to be buggy at best. Undocumented as well.
Delphi's current strength in corporate is server side / middle tear
programming. Yet Web Service are arcane to program compared to .Net
.Net, you are way, way too far behind MS.
I would prefer an upgraded Delphi 7 with massive amounts of work done on
the VCL, Unicode, DB, and things like Web Services, then what you have
done now. The product is practically useless except for maintaining old
code until it is moved to C# or Java (which is what is happening on the
project I am working on now).
Nick Hodges (Borland/DTG) wrote:
> http://blogs.borland.com/nickhodges/archive/2006/10/04/27823.aspx
>
--
Thomas Miller
Chrome Portal Project Manager
CPCUG Programmers SIG Chairperson (formally Delphi)
Delphi Client/Server Certified Developer
BSS Accounting & Distribution Software
BSS Enterprise Accounting FrameWork
http://www.bss-software.com
http://programmers.cpcug.org/
http://sourceforge.net/projects/chromeportal/
http://sourceforge.net/projects/uopl/
http://sourceforge.net/projects/dbexpressplus
I just asked Nick Hodges a direct question about the Win64 compiler is
scheduled and he said it is scheduled for release in 2008. Now, considering
that it is Oct, 2006, and its going to take about 2 years to build, that
means that if they start building it now it will be ready at the end of
2008, not the beginning.
-d
I wouldn't put too much credence in the roadmap, as after the sale,
priorities are probably going to change more than poeple expect.
1. Problems to port my framework and related applications to D2005 and
D2006. Due to few (but unresolved) OTA bugs and/or modificated behaeviours
my framework was without design-time support and without this there was no
reason to upgrade. With the latest Hotfixes the problem seems to be resolved
so i can finally test my framework using D2006 and then upgrade. (i've D2005
and D2006 licenses so you can be happy but i'm in contrast very
disappointed...)
2. I'm no .NET developer so i'm not interested to .NET feature that i pay (i
spend money for licences and RAM for .NET) without using it. Under this
prespective D7 and D2006 are not so different. Some nice IDE features, but
no key features are present for Win32 developer (IMHO). This is why i've not
spent to much time to upgrade my framework. BTW. i will not buy D2007 if
this release just contains .NET enhancements (2.0 or 3.0 support).
3. No new ORM Tool for Win32 is present. (My dream: a backport of
NHibernate - the .NET version of the popular java ORM Hibernate - to
Win32/64, with class helpers should be possible...)
I use Delphi because it's an indipendent way do develop software for the
windows platform. Delphi.NET is for me not an option because using a
language or an IDE with a different and proprietary framework like the .NET
framework make no sense, i prefer to use M$ products with M$ technology). I
need an updatet, native VCL, a 64bit compile an ORM Tool...
May be i'm not in your primary market target :-(
Best regards
Sergio Sette
Exactly. We do Win32 development and there is way to little language
extensions to convince my boss to buy the new version.
Rick
> It cost a whole bunch of money, much money for mere mortals, not for
> Borland :)
Yeah, it did originally cost lots of spondulix but Microsoft eventually gave
it away for free.
Dave
> Hear, hear!
One day they hopefully will!
<g>