CrossKylix discontinued
Rod
http://crosskylix.untergrund.net/
"Leonardo M. Ramé"
like PHP and Ruby and doesn't hearing the ever growing community of
FreePascal.
A big step in the correct direction may be allowing Delphi to compile
for both DCC32, FPC and Mono. In the future they can integrate some
other aspects of the IDE.
Leonardo M. Ramé
http://leonardorame.blogspot.com
Rod escribió:
Dennis Landi
Dean Hill
> bummer
I am not so sure. I think that the CrossFPC project is a better one in
the long term anyway. I can't see how it would be financially viable
for CG to work on a Linux compiler when there is already one being
actively worked on.
--
Dean
Dean Hill
> A big step in the correct direction may be allowing Delphi to compile
> for both DCC32, FPC and Mono. In the future they can integrate some
> other aspects of the IDE.
Maybe that is something CG can look at when the do the new version of
VCL for use in Tiberon. Try and make the code FPC compatible where
possible especially in the non visual stuff.
--
Dean
Brian Moelk
> Dennis Landi wrote:
>
>> bummer
>
> I am not so sure. I think that the CrossFPC project is a better one in
> the long term anyway.
Agreed 100%. Besides, if CodeGear isn't dedicated to Kylix, why should
there be any more effort in CrossKylix? It just doesn't make any sense;
it's best to build off something that is growing, not withering away.
> I can't see how it would be financially viable
> for CG to work on a Linux compiler when there is already one being
> actively worked on.
IMO, CG has two choices: revitalize/build their own or use FPC. If it
isn't already, in the next few years, cross platform development will
simply be a requirement for most development shops.
The best hedge they've got in the .NET space is to support mono 100%
(official support, development enhancements, IDE integration, etc). But
I think a better hedge is to leverage FPC.
--
Brian Moelk
Brain Endeavor LLC
bmo...@NObrainSPAMendeavorFOR.MEcom
Peter Morris
> totally alienated from Codegear and vice versa. Its weird.
They just don't seem to share a common goal.
Dennis Landi
"Dean Hill" <no...@none.com> wrote in message
news:4702...@newsgroups.borland.com...
> Dennis Landi wrote:
>
>> bummer
>
> I am not so sure.
Its a bummer that another major Delphi Developer in the community is totally
alienated from Codegear and vice versa. Its weird.
-d
Dean Hill
> Its a bummer that another major Delphi Developer in the community is
> totally alienated from Codegear and vice versa. Its weird.
That is disappointing. That said, I remember the unfortunate public
discussion around this matter and it's easy to see how these things
quickly get out of hand. I am sure it is something that communications
could have sorted but, IMO, it was over before it even became a public
discussion. As Peter said, they have a different goals/requirements.
These things are bound to happen from time to time.
--
Dean
Dennis Landi
"Peter Morris" <sup...@NOdroopySPAMeyes.com> wrote in message
news:4702...@newsgroups.borland.com...
>> Its a bummer that another major Delphi Developer in the community is
>> totally alienated from Codegear and vice versa. Its weird.
>
> They just don't seem to share a common goal.
>
What is so exceptional about the goal of supporting the Linux platform?
That's weird, too.
-d
Dean Hill
> What is so exceptional about the goal of supporting the Linux
> platform? That's weird, too.
From the sounds of it, the majority of CG's clients are Windows
developers. Any Linux development would be a risk. IMO, it's a
worthwhile risk but at the end of the day, any development has to make
money. I believe that the FPC route is better as it provides not just
Linux support, but Mac and various others as well.
The only thing lacking for me right now in FPC is runtime package
support. Delphi will always be my IDE of choice for pascal
development, whether using the Delphi compiler or the FPC compiler.
IMO, CG will have to go multi-platform (And I am not talking .NET :) )
eventually as the Native Windows space is decreasing. More and more
Windows developers are being hoodwinked by MS. That will leave a whole
bunch of developers who need to start supporting the Linux's and Mac
OSX's of the world. I have still not seen anything that even comes
close to Delphi in terms of easy of use, power and speed.
--
Dean
marc hoffman
>> http://crosskylix.untergrund.net/
>
> bummer
if it means a renewed focus on CrossFPC - which i can't wait for - i
think this is actually great news. Given the choice of using a years-old
unmaintained back-end compiler vs a moder up-to-date one that also
targets x64, OSX and other platforms, i think FPC will be by far the
better choice to use now-a-days, than Kylix.
--
marc hoffman
RemObjects Software
The Infrastructure Company
http://www.remobjects.com
bee
>It's funny to see CodeGear making IDE's for some Open Source projects
>like PHP and Ruby and doesn't hearing the ever growing community of
>FreePascal.
Arrogance? Shame? Dignity? :P
>A big step in the correct direction may be allowing Delphi to compile
>for both DCC32, FPC and Mono. In the future they can integrate some
>other aspects of the IDE.
http://lazarus-dev.blogspot.com/
If Lazarus (Delphi-like IDE for FPC) is getting more and more complete and stable, we don't need Delphi anymore, right? :P
If CG keep acting like this to Delphi community, we will happily move to FPC/Lazarus and leave CG with its arrogance. Congratulations! ;)
-Bee-
Dennis Landi
"Dean Hill" <no...@none.com> wrote in message
news:4702...@newsgroups.borland.com...
IMO, the only thing missing from the current strategy is a X-platform
vision. I, too, could really make use of this.
We see x-platform PHP and Ruby initiatives and a revitalized JBuilder. It
could be that Codegear intends to try to capture x-platform market share
with these products and just let the Delphi developers who want x-platform
switch to those products while using Delphi for its "traditional" purpose,
which is Win32/64 development.
If I look at that dispassionately enough, I guess that is fine. At this
point in my career I am multi-lingual because I saw the writing on the wall
five years ago. Basically If I want x-platform tools, I'll need to master
other toolsets to make it happen. Either way, I won't be harmed, I guess.
But as a "fan" of the Delphi toolset, it saddens me to see squandered
opportunities for partnerships in the Linux arena. Chad Hower (aka Saint
Chadwick) and Simon Kissel (soon to be known as Saint Simon) should have
been staunch leaders in the revitalized Delphi Linux community instead of
what? Ostracized?
And what of opportunities to bring FPC into a larger Delphi community? And
what about RemObjects who nailed their .NET Chrome compiler strategy from
the start?
Is Codegear really making the right choices by eschewing all of these
players? I think the alternative would have been better for us the customer
base.
But, I am finally beginning to care less.
I am currently developing new products for Win32 using Delphi. For what it
is, its damned good. I like it. Well, actually, I love it. And I guess
that's enough for now. Being multi-lingual is a pain in the arse; but on
the other hand it makes me more competitive in the market place and helps me
put on all the young whipper-snappers in their place.
The one bright spot on the horizon for me in the upcoming Unicode and Win64
support. This at least ensures the viability for the native windows
compiler into the future. So at least my current efforts on the Win32
platform has a supported future. Things could be worse.
So my new advice is this: Go out and learn a new language or two. You'll be
stronger, faster, smarter and better looking as a result...
-d
Tom Corey
> If CG keep acting like this to Delphi community, we will happily move
> to FPC/Lazarus and leave CG with its arrogance. Congratulations! ;)
Yeah, I really hate how CG is offering new, improved products lately.
Their insistence of increasing my productivity is really pissing me off.
Dean Hill
> If Lazarus (Delphi-like IDE for FPC) is getting more and more
> complete and stable, we don't need Delphi anymore, right? :P
I disagree, the difference between Lazarus and Delphi is astronomical.
Delphi is probably 5 years ahead of Lazarus and will keep going. That
does not mean one should not be able to use Delphi to build FPC
applications. Lazarus has the benefit of being multi-platform. This
is also, to a certain extent, it's achilles heel. The Delphi IDE is
focussed on Windows with a dedicated team of developers. This makes
changes to Delphi far easier to effect and will make it very difficult
for Lazarus to even come close in functionality and polish.
--
Dean
Dean Hill
> So my new advice is this: Go out and learn a new language or two.
> You'll be stronger, faster, smarter and better looking as a result...
I already know far too many languages :). Despite looking and working
with all of these alternatives, I still believe Delphi (both
environment and language) is the best choice for the majority of
Windows development. The problem is, that every time I use another
tool, I just finding myself keep wishing that I was able to use Delphi
instead of being stuck with something that is limited in comparison.
--
Dean
Brian Moelk
> The problem is, that every time I use another
> tool, I just finding myself keep wishing that I was able to use Delphi
> instead of being stuck with something that is limited in comparison.
IMO, unless you have really paid your dues with the other tool/language,
this is called a "learning curve". :)
GrandmasterB
> Delphi is probably 5 years ahead of Lazarus and will keep going. That
Not so much a problem for those of us still using a 6 year old Delphi 7 ;-)
WillR
The perhaps you should switch to Visual Studio and program in BASIC. lol
Just kidding -- really!
I tried VS -- really -- just leaves me wishing MS could do as well as
CodeGear (Borland)
--
Will R
PMC Consulting
Dean Hill
I should probably have said 10 years then because even Delphi 7 is
quite a bit ahead of Lazarus. That said, if you are still using Delphi
7, you should take a look at Delphi 2007. There are a number of
productivity enhancements that could save you some development time.
--
Dean
Nick Hodges (CodeGear)
> Their insistence of increasing my productivity is really pissing me
> off.
Yeah, sorry about that. ;-)
--
Nick Hodges
Delphi Product Manager - CodeGear
http://blogs.codegear.com/nickhodges
Jolyon Smith
says...
> It's funny to see CodeGear making IDE's for some Open Source projects
> like PHP and Ruby and doesn't hearing the ever growing community of
> FreePascal.
You have a funny sense of humour, is all I can say.
;)
--
JS
www.deltics.co.nz
Dean Hill
> IMO, unless you have really paid your dues with the other
> tool/language, this is called a "learning curve". :)
I have paid my dues, it still doesn't help :)
It's like driving a fancy car and then a rust bucket. In your mind,
you keep comparing the rust bucket to the fancy car, regardless of how
long you drive the rust bucket.
--
Dean
Nick Hodges (CodeGear)
> That said, if you are still using Delphi
> 7, you should take a look at Delphi 2007. There are a number of
> productivity enhancements that could save you some development time.
Definitely.
GrandmasterB
> I should probably have said 10 years then because even Delphi 7 is
> quite a bit ahead of Lazarus. That said, if you are still using Delphi
It really depends on how you measure that, doesnt it? FPC/Lazarus, being
cross platform, is thus many years ahead of Delphi. It all depends on what
you are looking for in an IDE.
> 7, you should take a look at Delphi 2007. There are a number of
> productivity enhancements that could save you some development time.
Been there, done that, bought the collectable shot glass. While its
certainly better than D2006 & D2005, I still prefer my sleeker D7. I'm not
interested in IDE enhancements and most of the new language/component
additions just arent of any interest/use to me. I probably wont upgrade
again until they have either a unicode VCL or cross-platform development
capabilities.
Peter Morris
Where did I say anything was exceptional? I just said that they don't share
the same goals.
Dean Hill
> It really depends on how you measure that, doesnt it? FPC/Lazarus,
> being cross platform, is thus many years ahead of Delphi. It all
> depends on what you are looking for in an IDE.
The question is whether the IDE needs to be cross platform. I am happy
to develop in a Windows VM for Linux. While a cross platform IDE is
nice, it's not a necessity for cross-platform development.
> I'm not interested in IDE enhancements and most of the new
> language/component additions just arent of any interest/use to me. I
> probably wont upgrade again until they have either a unicode VCL or
> cross-platform development capabilities.
It took me about 1 week to stop using Delphi 6 once I had BDS 2006. I
have Delphi 2007 but can't switch until after our current release.
Yes, there are problems, but the productivity gains outstripped those.
The VCL changes are probably something that you should seriously look
at, even if it is just using Delphi 2007 to do your builds. There have
been important changes there too.
--
Dean
Marius
> The only thing lacking for me right now in FPC is runtime package
> support. Delphi will always be my IDE of choice for pascal
> development, whether using the Delphi compiler or the FPC compiler.
As soon i got a *decent* ide together with fpc + runtime packages i'm
also out. And we will discover a whole new world (+bugs) ;-)
> IMO, CG will have to go multi-platform (And I am not talking .NET :) )
> eventually as the Native Windows space is decreasing. More and more
> Windows developers are being hoodwinked by MS. That will leave a whole
> bunch of developers who need to start supporting the Linux's and Mac
> OSX's of the world. I have still not seen anything that even comes
> close to Delphi in terms of easy of use, power and speed.
I agree 100%. The OS market is opening up again, MS is rapidly losing
its grip on software world.
Marius
ms.net to run smoothly under linux ;-)
Marius
that largefile problem, so when the hack is upd#4 coming :P)
But to get back ontopic, out of curiosity, what do you think was such a
major enhancement in 2007.win32? As far i could see they only did minor
improvement on win32 itself.
-Web2(Ok)
-New IDE (the same as d2005/6)
-Blackfish (To new for heavy duty)
-Together (Nah, we not gonna switch again dudes)
-Starteam (never used it, an old product anywhay)
-DbExpress4 (Nothing magical, new and updated drivers, cheers)
-Delphi Language Enhancements? (no, we need compatibility for fpc)
-Refactoring? (greplace still does a much better job, project wide!)
Basicly i find our development is turing away from the general direction
codegear is taking (we will never need .not., a major untertaking and a
stupid catchup game with ms), however we do need to be native
multiplatform in the future (for now win32, later linux and some arm/wince)
I want to break the VCL code many times like FPC does :P
Nick Hodges (CodeGear)
> But to get back ontopic, out of curiosity, what do you think was such
> a major enhancement in 2007.win32?
Over what product...?
Dean Hill
> But to get back ontopic, out of curiosity, what do you think was such
> a major enhancement in 2007.win32? As far i could see they only did
> minor improvement on win32 itself.
From D7 or BDS2006?
--
Dean
Marius
Dean Hill
> As soon i got a decent ide together with fpc + runtime packages i'm
> also out. And we will discover a whole new world (+bugs) ;-)
I would not stop using the Delphi compiler but rather use the FPC
compiler where Delphi support is not available. For Windows client
side apps, FPC just does not have the third party support (DevEx etc).
--
Dean
James David
> It's funny to see CodeGear making IDE's for some Open Source projects
> like PHP and Ruby and doesn't hearing the ever growing community of
> FreePascal.
It is indeed surprising that Borland/Codegear has completely abandoned
or ignored any type or size of effort to put pascal on *nix. Oh well,
they must know their business better than I do :-)
Anyway, we've ported A LOT of our analytics libraries from
Delphi/Windows to fpc/Linux not because of Delphi, but because Linux
offers so much more flexibility in fulfilling the demands of our
business niche.
--
Dennis Landi
"Peter Morris" <sup...@NOdroopySPAMeyes.com> wrote in message
news:4702b856$1...@newsgroups.borland.com...
>> What is so exceptional about the goal of supporting the Linux platform?
>
> Where did I say anything was exceptional? I just said that they don't
> share the same goals.
>
>
My point is that the x-platform goal should not be a difficult one to share.
And, not having such a goal at all (and not sharing it with your customer
base publically) is W E I R D.
-d
Ed
> "Leonardo M. Ramé" wrote:
>
>> It's funny to see CodeGear making IDE's for some Open Source projects
>> like PHP and Ruby and doesn't hearing the ever growing community of
>> FreePascal.
>
> It is indeed surprising that Borland/Codegear has completely abandoned
> or ignored any type or size of effort to put pascal on *nix. Oh well,
> they must know their business better than I do :-)
>
I also feel CG should reconsider their stance with the *nix
community. Borland's attempt was laudible at best; but they
just didn't put their best feet forward. I believe the
initial version left a very bad taste in those who bought
it.
However, with the current IDEs so much tied to the .NET
platform, it would take a miracle for CG to either decouple
the dependency and/or change the platform such that it
works both under Windows and *nix.
The obstacles are definitely there (in terms of platform
support, (db) driver support, etc). Will CG bother?
Edmund
Ed
> Marius wrote:
>
>> But to get back ontopic, out of curiosity, what do you think was such
>> a major enhancement in 2007.win32?
>
> Over what product...?
>
I would assume previous products, but then again, that wouldn't be
considered 'on topic' since CrossKylix and 2007.win32 are different
products. Yes. Brilliant art of stating the obvious. Thank you. :)
Edmund
Ed
> good old delphi7
I don't have 2007, but between D7 and BDS2006, I would say the
enhancements are pretty good, though the only one I
use daily is the code folding. Code refactoring? I haven't
figured that out yet.
The bad thing is... the help.
Edmund
Brian Evans
Refactoring at the basic level is the editing of code with a tool
that understands it's structure and syntax.
The easiest one to understand is renaming. A project has
lots of controls with default numbered names. With refactoring
you can give them meaningful names with a lot less work
compared to normal text editing with things like search
and replace. Pick someplace where say Button45 is referenced,
right click and select Refactoring, Rename field "Button45".
Enter a better name like ButtonNewInvoice and click OK.
You now get a list of changes it will make, apply them
all and the control in question has been intelligently
renamed throughout. So you now have (taken from random lines):
ButtonNewInvoice: TButton;
procedure ButtonNewInvoiceClick(Sender: TObject);
ButtonNewInvoice.Enabled := false;
The other refactoring options are more specialized but not
too complex once you figure out what the end goal of each is.
It takes some playing around to understand what each does
but it's worth the effort.
Brian
Anders Isaksson
> The easiest one to understand is renaming.
Or maybe "declare variable"? Saves some (flow-breaking) scrolling
around in the source for finding the right place to insert the variable
(assuming local scope is what you want).
--
Anders Isaksson, Sweden
BlockCAD: http://web.telia.com/~u16122508/proglego.htm
Gallery: http://web.telia.com/~u16122508/gallery/index.htm
Kryvich
> I tried VS -- really -- just leaves me wishing MS could do as well as
> CodeGear (Borland)
+1
Delphi 2007 IDE much better then Visual Studio 2005 IDE in stability
and usability.
Anders Isaksson
> What is so exceptional about the goal of supporting the Linux
> platform?
I don't think that's a *goal* for a company. The *goal* for companies
is making money, the *means* may be to support Windows program
development, linux program development etc.
Kryvich
>> It is indeed surprising that Borland/Codegear has completely abandoned
>> or ignored any type or size of effort to put pascal on *nix. Oh well,
>> they must know their business better than I do :-)
>>
> I also feel CG should reconsider their stance with the *nix
> community. Borland's attempt was laudible at best; but they
> just didn't put their best feet forward. I believe the
> initial version left a very bad taste in those who bought
> it.
All I need from future Delphi:
1. Unicode support (planned for 2008, thanks!)
2. Linux (cross)compiling
Anders Isaksson
> My point is that the x-platform goal should not be a difficult one to
> share.
Well, there are *millions* of programmers out there who *don't* share
that goal. What makes you think you are part of a majority? Or even a
*substantial* minority?
Arthur Hoornweg
> The easiest one to understand is renaming. A project has
> lots of controls with default numbered names. With refactoring
> you can give them meaningful names with a lot less work
> compared to normal text editing with things like search
> and replace. Pick someplace where say Button45 is referenced,
> right click and select Refactoring, Rename field "Button45".
This feature is sometimes problematic.
Take the following code for example:
Procedure tform1.Button1Click (sender: tObject);
begin
// Do something
end;
Procedure tform1.Timer1Timer (sender:tObject);
Begin
Button1Click(NIL);
End;
Now if you use refactoring to rename "Button1" to "btnCheck"
the project will no longer compile and requires manual work.
This once got me into real trouble when I "cleaned up" a huge
form and renamed some 50 controls using refactoring.
--
Arthur Hoornweg
(In order to reply per e-mail, please just remove the ".net"
from my e-mail address. Leave the rest of the address intact
including the "antispam" part. I had to take this measure to
counteract unsollicited mail.)
Brian Moelk
> Well, there are *millions* of programmers out there who *don't* share
> that goal.
Yes, Visual Studio has many users. :)
--
Brian Moelk
Brain Endeavor LLC
bmo...@NObrainSPAMendeavorFOR.MEcom
Marius
things (specially those "with statements").
Try refactoring a public property, you _might_ expect it to adjust the
dependent calls, but it doesn't (and its not that easy)
I miss the most usefull refactoring of the unit-name (project wide
refactoring of the uses clause)
Marius
(and its not even funny)
Nick Hodges (CodeGear)
> I would assume previous products,
Ed --
Brilliant -- yes, of course, /previous/ products! ;-)
I was angling more for which specific previous product. If it is, say,
Delphi 7, then the list of advantages is pretty long.
Tom Corey
>> Their insistence on increasing my productivity is really pissing me
>> off.
> Yeah, sorry about that. ;-)
It's ok. I compensate by only typing with my left hand.
Nick Hodges (CodeGear)
> It's ok. I compensate by only typing with my left hand.
;-)
Nick Hodges (CodeGear)
> The other refactoring options are more specialized but not
> too complex once you figure out what the end goal of each is.
The two I use most frequently are Rename and Extract Method.
I love Extract Method -- you end up with "clean" code and nice small
routines.
Nick Hodges (CodeGear)
>
> Excuse me, what help?
> (and its not even funny)
Have you checked out the help in Delphi 2007 lately?
Marius
2007.Rad or win32?
Marius
or less stopped on the large file problem.
Pete Fraser wrote:
> They are both the same with the latest update (3) - at least
> the help is...
> HTH Pete
>
> "Marius" <mar...@somewhere.nl> wrote in message
> news:4703bf2d$1...@newsgroups.borland.com...
Pete Fraser
the help is...
HTH Pete
"Marius" <mar...@somewhere.nl> wrote in message
news:4703bf2d$1...@newsgroups.borland.com...
Richie B.
> on a Linux compiler when there is already one being actively
> worked on.
Sorry, for a moment I thought you wrote:
"I can't see how it would be financially viable for CG to work on a 64-bit
Windows compiler when there is already one being actively worked on."
or
"I can't see how it would be financially viable for CG to work on a .NET
3.0+ compiler when there is already one being actively worked on."
or even
"I can't see how it would be financially viable for CG to work on a
crossplatform 32-bit Windows compiler when there is already one being
actively worked on."
RB
Marius
> Ed wrote:
>
>> I would assume previous products,
>
> Ed --
>
> Brilliant -- yes, of course, /previous/ products! ;-)
>
> I was angling more for which specific previous product. If it is, say,
> Delphi 7, then the list of advantages is pretty long.
to say it blunt: delphi has not changed much from delphi5 up to now.
Sure the're the logical improvements, vb-like ide, bug fixes, addons,
newer drivers etc., but it all minor, nothing impressive for win32
legacy. Important to know in this case is that we are absolutely not
waiting or depending on .net or any of the other new gadgets.
i'm interested in generics.win32, cpu64 (Commodore '2009) and
multiplatform, at least those changes would make sense to us.
Nick Hodges (CodeGear)
>
> 2007.Rad or win32?
Delphi 2007 UPdate 3, and RAD STudio 2007, have the same help.
It's also now much easier for us to do Help updates.
Nick Hodges (CodeGear)
> to say it blunt: delphi has not changed much from delphi5 up to now.
Well, to put it bluntly, that's totally wrong.
This is not a short list:
http://blogs.codegear.com/nickhodges/2006/11/27/29891
and it's not even complete.
This post points to demos of many of those features:
http://blogs.codegear.com/nickhodges/2007/03/28/33579
> i'm interested in generics.win32, cpu64 (Commodore '2009) and
> multiplatform, at least those changes would make sense to us.
Okay, that's fine. But just because the long, thorough, and full list
of new features since Delphi 5 aren't things that /you/ might
particularly like or want doesn't mean that the list isn't long, full,
and replete with useful enhancements and features.
Pete Fraser
market for *commercial* linxu IDEs is even smaller.
It therefore makes little sense to work on a compiler for a
small market when someone is working on a free one.
The windows platform is a completely different situation as
CGs revenue is much larger.
Rgds Pete
"Richie B." <firs...@lastname.com> wrote in message
news:4703c314$1...@newsgroups.borland.com...
Dean Hill
> "I can't see how it would be financially viable for CG to work on a
> 64-bit Windows compiler when there is already one being actively
> worked on."
They have no choice here. Also, I would not imagine that the changes
are as much as would be required for a multi-platform solution.
> "I can't see how it would be financially viable for CG to work on a
> .NET 3.0+ compiler when there is already one being actively worked
> on."
This is a gray area. I am not sure how many people are using Delphi
.NET but it would seem that there are more Delphi .NET developers then
Delphi cross platform developers. I would personally prefer to see the
effort going into modifying the VCL for cross platform FPC support.
> "I can't see how it would be financially viable for CG to work on a
> crossplatform 32-bit Windows compiler when there is already one being
> actively worked on."
This would be closest to what I said. I can't see CG doing any effort
towards this as the work would be astronomical. I would prefer CG to
start moving towards supporting the FPC compiler along with the Delphi
compiler. That on it's own would be a massive undertaking, VCL and
hosting of components. This is why I posted in another (same?) thread
about how it would be nice for CG to add FPC support to the VCL for non
visual stuff and perhaps even the server stuff. Don't think it will
happen but who knows, the world is a strange place.
--
Dean
Marius
>> If Lazarus (Delphi-like IDE for FPC) is getting more and more complete
>> and stable, we don't need Delphi anymore, right? :P
>
> Would be right if it were so. But it isn't.
> Installed it recently for developping PDA software. Installed FPC first
> then FPC WinCE arm compiler and then lazarus. Tried to pint lazarus to
> the arm compiler but he always whines that it can't find the compiler
> and I should direct him to a directory where probably such and such
> files are present as this would be the right directory. But I had done
> that? Can't it just stop whining and do the bloody compile?
>
> (opkay will look at it again now with a tutorial, but this seems a bad
> start!)
We had the same experiences a year ago, but succeeded in developing it
(load of unusual unexpected stuff on wince, geesh, but ok, even indy
worked in the end). We saved the installation of fpc/lazarus, since then
i have been updating but can't get lazarus stable for arm/ce
Markus.Humm
>
> I miss the most usefull refactoring of the unit-name (project wide
> refactoring of the uses clause)
>
Hello,
just go ahead and QC this as feature request.
Greetings
Markus
Rolf Lampa [RIL]
> Marius wrote:
>
>> to say it blunt: delphi has not changed much from delphi5 up to now.
>
> Well, to put it bluntly, that's totally wrong.
>
> This is not a short list:
>
> http://blogs.codegear.com/nickhodges/2006/11/27/29891
>
Got exited, pardon me.
Being involved as a business consultant, I started designing a
enterprise business system using Delphi, in -98. Former acquaintance
with OOP was Turbo Pascal 5.5. You can imagine. We started using
Rose98 for modeling. We planned to use Bold Architecture, and we did.
But UML what is that?
However, we started using Delphi 4, then 5, and finally, D6 was the
version when I left the project in 2004, when a ~350 class system had
been up running for a year or so. And yes, we heard of .Net and D8,
and we gladly skipped it since the only thing that really had changed
since the days of TP5.5 was MD, that is, Bold Architecture, or now,
Eco. Which we were already using.
So what am I trying to say? Yes, that semantically not much have
changed. Since TP5.5. When was that anyway...?
But yes, Delphi HAS changed. But from the perspective of the "end
customer", which includes company managers, investing in enterprise
tools, for realizing (administrating) real business goals (apart from
imaginary such achieved only on power point presentations), not much
have changed.
Sorry, not much have changed. And still quite a bit has changed. But
it depends on perspective how you quantify "much".
In the design project (I took active part also in the detailed coding
of the enterprise system aforementioned) I actually used my skills in
OOP from the days of TP5.5. That's what I am saying.
And I had fun, almost all the time. I think the reserved word
"deprecated" has been introduced since TP5.5. And properties. And some
more. But you get the idea.
From the perspective of actually achieving an end goal, for a paying
customer, perhaps not very much have changed from TP5.5 to D(n),
except for forms and the graphical interface.
OK, some things have changed. I noticed that in D4,5 & 6 I no longer
had to fit code blocks into 64K chunks...
To be serious, very serious, no, it's not enough to sell the very good
tool Delphi to developers - only. It is potentially more useful than
that. I love Delphi. I use it right now (making a Mediawiki dump
processing tool), but in so doing I do well with D7. I could as well
have used D4, well, but at least D5, D6 or D2005 would do the trick
for me. Or D2006. I have all those, but I use D7.
Semantically not much have changed.
Code folding? As said, it depends on perspective. If you want to get a
better/different grasp of the semantics of the system you
build/design, and if the system only slightly complex, then Model
Driven is a significant change, not code folding. I'm serious.
But MD is new, ans special, and not all will grasp the idea and go for
that for yet a while. But imagine if I could at least (re)compile both
the Mediawiki processor I'm working on to run on the X-platform which
Mediawiki (preferably) runs on - AND the aforementioned enterprise
business system too! Both these apps needs to scale, with a factor of
x20 - x10+ in respective case, in the near future. They need to scale
both regarding data storage capacity, data throughput, and capacity to
serve clients. That means a lot of hardware. Stacks of hardware. And
execution platforms (OS). In short, lots of $, or M$ to be more precise.
What if it only took recompiling it into X-platform... or at least
parts of it. And think about all the "zillions" of other small
"services" out there, which are getting built with PHP etc. It scales.
At least the people sitting on the wallet cares in arguments like
this. Many of them wants to scale, significantly, and .net or dot
this, or a fix in the syntax/protocol won't help them a bit in getting
"more competitive".
The next generation decision makers might notice what really makes
them "more competitive". Spinning the number after D won't do the trick.
Yes Dennis, it's weird, it really is. I would try to come up with and
argument which addresses the people sitting on the wallet, too, if I
had something I wanted to sell. Arguments with true substance.
I have seen all this it from both sides, from the wallet side and from
the developer side. Delphi is an excellent tool. But former versions
were excellent too. =)
Regards,
// Rolf Lampa
Markus.Humm
> Indeed :P, and i'm afraid they stuffed the ide it with to many dependent
> ms.net to run smoothly under linux ;-)
>
> Jolyon Smith wrote:
>>> It's funny to see CodeGear making IDE's for some Open Source projects
>>> like PHP and Ruby and doesn't hearing the ever growing community of
>>> FreePascal.
>>
>> You have a funny sense of humour, is all I can say.
>>
>> ;)
Hello,
could you think about changing your posting style from top qoute to
appending your replys at the end? (no pun intended or so, just 99,999%
use that style and always switching when reading your posts, which are
welcome here, is a tiny little bit annoying...)
Greetings
Markus
Markus.Humm
>> It's funny to see CodeGear making IDE's for some Open Source projects
>> like PHP and Ruby and doesn't hearing the ever growing community of
>> FreePascal.
>
>
>> A big step in the correct direction may be allowing Delphi to compile
>> for both DCC32, FPC and Mono. In the future they can integrate some
>> other aspects of the IDE.
>
> http://lazarus-dev.blogspot.com/
>
> If Lazarus (Delphi-like IDE for FPC) is getting more and more complete and stable, we don't need Delphi anymore, right? :P
Would be right if it were so. But it isn't.
Installed it recently for developping PDA software. Installed FPC first
then FPC WinCE arm compiler and then lazarus. Tried to pint lazarus to
the arm compiler but he always whines that it can't find the compiler
and I should direct him to a directory where probably such and such
files are present as this would be the right directory. But I had done
that? Can't it just stop whining and do the bloody compile?
(opkay will look at it again now with a tutorial, but this seems a bad
start!)
Greetings
Markus
Markus.Humm
Unicode VCL is planned for the next release already!
Greetings
Markus
Markus.Humm
I don't think a *nix IDE would be required, a cross compiler would be
enough (maybe coupled with a remote debugger) for a good start!
Greetings
Markus
Marius
> Marius wrote:
>
>> 2007.Rad or win32?
>
> Delphi 2007 UPdate 3, and RAD STudio 2007, have the same help.
>
> It's also now much easier for us to do Help updates.
>
Question, why the same help? Is CG trying to make a new record on
pushing unwanted programs and now even the .net help to 2007.win32? (jk) ;-)
Rolf Lampa [RIL]
> Hello,
>
> I don't think a *nix IDE would be required, a cross compiler would be
> enough (maybe coupled with a remote debugger) for a good start!
>
Exactly. It would have the potential of changing the world. Also for
designers of business enterprise systems.
Giving access to an almost free execution platform for applications
developed with Delphi doesn't undermine the market for Delphi, it just
makes the arguments stronger - for using Delphi.
Regards,
// Rolf Lampa
Nick Hodges (CodeGear)
> Sorry, not much have changed. And still quite a bit has changed. But
> it depends on perspective how you quantify "much".
I think that is pretty much my point.
From one perspective, /nothing/ has changed since the days of John Von
Neumann (http://en.wikipedia.org/wiki/John_von_Neumann), right? I
mean, way back then, programmers had a problem, they coded and ran a
solution. How is that any different from what we do today? Well, it's
not.
But of course it is --- the tools are radically different, and orders
of magnitude of orders of magnitude more powerful. So are the machines.
So, in the end, it's hard to argue that things haven't changed since,
say, D7. I guess it just matters where you look at it from. I'm
perfectly happy to accept that for some developers, there's little to
be excited about. But that doesn't mean that there is universally
nothing to be excited about.
Rolf Lampa [RIL]
> Except that the market for Linux IDEs is very small and the
> market for *commercial* linxu IDEs is even smaller.
What is the *commercial* value of having an option to scale with your
existing software by swapping to a different execution platform (OS)
for, well, lets say "significantly less money" than you are now forced
to spend if you need to scale?
The argument seems to be that because none is using your product you
shouldn't bother trying to sell it. Hm.
> It therefore makes little sense to work on a compiler for a
> small market when someone is working on a free one.
The execution platform is not a "market". The user, and the usability
of a tool, that is, the true value of the tool, is what's significant
in determining the "market". For the tool.
This might, or might not, apply also for Delphi... um.
Mission critical business W32 applications could (or should) scale,
running on X-platforms. Yes, why not?
The "size of the market" depends also on whether anyone CAN utilize
the almost free X-platforms out there.
Marius
> Rolf Lampa [RIL] wrote:
>
>> Sorry, not much have changed. And still quite a bit has changed. But
>> it depends on perspective how you quantify "much".
>
> I think that is pretty much my point.
Agreed, Rolf refactored it in better diplomatic words..
Rolf Lampa [RIL]
>
> So, in the end, it's hard to argue that things haven't changed since,
> say, D7. I guess it just matters where you look at it from. I'm
> perfectly happy to accept that for some developers, there's little to
> be excited about. But that doesn't mean that there is universally
> nothing to be excited about.
Agree. But always agreeing on the developer perspective is what I find
a bit troublesome. The developer perspective is not irrelevant, nor is
it less important, but it's not exclusive the only perspective.
I mean, how excited are mature IT managers about ever new syntax and
garbage collectors? They not only have to defend their spending (or
not spending) before developers, they also have to give arguments for
why/how a real world enterprise would benefit from this or that (in
perspective) low level syntax instead of the other syntax. What I am
saying is that they really really do need help with the arguments.
Well, X is one. And you can make figures based on it. Real figures. =)
Regards,
// Rolf Lampa
Marius
> Well, to put it bluntly, that's totally wrong.
>
> This is not a short list:
>
> http://blogs.codegear.com/nickhodges/2006/11/27/29891
>
> and it's not even complete.
>
> This post points to demos of many of those features:
>
> http://blogs.codegear.com/nickhodges/2007/03/28/33579
>
>> i'm interested in generics.win32, cpu64 (Commodore '2009) and
>> multiplatform, at least those changes would make sense to us.
>
> Okay, that's fine. But just because the long, thorough, and full list
> of new features since Delphi 5 aren't things that /you/ might
> particularly like or want doesn't mean that the list isn't long, full,
> and replete with useful enhancements and features.
LOL last post on this one, we are talking here about win32 so please
strip away anything with the .net stuff, leave out the stuff that could
be implemented otherwise like Fastcode, Rave, Together, dUnit etc. (I
cannot even find ECO in the 2007.win32 feature matrix so strip that also).
Leave out the language improvements because whe want to be absolutely
compatibility with other versions of delphi (delphi7 mainly but also fpc
if possible) and also in case delphi is becoming unusable and we need to
downgrade (like i found out recently)
I have not seen some of the new VCL components, but i'm not excited with
any of them, i'm sure there are third party components who can replace
them easely.
Leave me with only with this list of some IDE Features, so why should i
as a win32 programmer be excited with what borland/codegear has been
doing the last 6 years?
IDE Features Since Delphi 7
* Live Templates
* Block Completion
* History Tab
* Code Healer Borland Edition
* Castalia Borland Edition
* Better debugging (enhanced locals, call stack, expandable
tooltips and watchs, etc.)
* VCL Guidelines
* Code Folding
* SyncEdit
* Editor Line Numbers
* Line Change indicators
* Refactoring
* Flexible docking IDE
* Audits and Metrics
And some of these aren't even new, some of the new ide enhancements
looks like it has been (excuse me for putting it this way) "stolen" from
coderush who had some of these enlightened ideas 7 years ago (you know,
that guy who offered to correct some ide problems back then for free).
So its a very short list for win32 programmers, and just the ide
improvements won't do it for me (certainly not for that price and bugs).
Bob Dawson
> Leave out the language improvements because whe want to be
> absolutely compatibility with other versions of delphi (delphi7
> mainly but also fpc if possible)
So your complaint is that you don't want any new language features, and so
there's nothing but IDE features to look at?
I can see where that would be a problem. <g>
bobD
Marius
>> I can't see how it would be financially viable for CG to work
>> on a Linux compiler when there is already one being actively
>> worked on.
>
> Sorry, for a moment I thought you wrote:
>
> "I can't see how it would be financially viable for CG to work on a 64-bit
> Windows compiler when there is already one being actively worked on."
>
> or
>
> "I can't see how it would be financially viable for CG to work on a .NET
> 3.0+ compiler when there is already one being actively worked on."
>
> or even
>
> "I can't see how it would be financially viable for CG to work on a
> crossplatform 32-bit Windows compiler when there is already one being
> actively worked on."
>
> RB
>
>
some money :P
Rolf Lampa [RIL]
>
> Leave me with only with this list of some IDE Features, so why should i
> as a win32 programmer be excited with what borland/codegear has been
> doing the last 6 years?
>
> IDE Features Since Delphi 7
> * Live Templates
> * Block Completion
> * History Tab
> * Code Healer Borland Edition
> * Castalia Borland Edition
> * Better debugging (enhanced locals, call stack, expandable tooltips
> and watchs, etc.)
> * VCL Guidelines
> * Code Folding
> * SyncEdit
> * Editor Line Numbers
> * Line Change indicators
> * Refactoring
> * Flexible docking IDE
> * Audits and Metrics
>
> And some of these aren't even new, some of the new ide enhancements
> looks like it has been (excuse me for putting it this way) "stolen" from
> coderush who had some of these enlightened ideas 7 years ago ...
I felt very productive using D6 with latest version of CodeRush. Now I
use ModelMaker 8.2 with D7, and I refuse to start the installed D2005,
or D2006.
What has real "substantial value" in the list above is, IMO,
> * Better debugging (enhanced locals, call stack, ...
> * Refactoring
I appreciate the call stack, and refactoring comes with ModelMaker so
D7 will do.
In conclusion; Adding support for X would add to the substantial list,
actually increasing it with 50%... :)
Regards,
// Rolf Lampa
Marius
> Marius wrote:
>> leave out the stuff that
>> could be implemented otherwise like Fastcode, Rave, Together, dUnit
>> etc. (I cannot even find ECO in the 2007.win32 feature matrix so
>> strip that also).
>> Leave out the language improvements
>
> has changed I guess.
>
Well, think we must agree to disagree :P
I really think you should start a whats new for win32 developers. Then
also see if they can get the same functionality otherwise (plugns,
experts etc.), what left is a list that will be rather short.
Wayne Niddery [TeamB]
> leave out the stuff that
> could be implemented otherwise like Fastcode, Rave, Together, dUnit
> etc. (I cannot even find ECO in the 2007.win32 feature matrix so
> strip that also).
> Leave out the language improvements
Yes, if you leave out a lot of stuff that has changed, then, well, not much
has changed I guess.
--
Wayne Niddery - Winwright, Inc (www.winwright.ca)
"We've all heard that a million monkeys banging on a million
typewriters will eventually reproduce the entire works of Shakespeare.
Now, thanks to the Internet, we know this is not true." - Robert
Wilensky
Daniël Mantione
"Dean Hill" <no...@none.com> wrote:
>Dennis Landi wrote:
>
>> bummer
>
>I am not so sure. I think that the CrossFPC project is a better one in
>the long term anyway. I can't see how it would be financially viable
>for CG to work on a Linux compiler when there is already one being
>actively worked on.
There is a good C compiler for Linux: GCC.
Intel, Pathscale and PGI sell C compilers for Linux and make
good money with it. Why are they financially viable, and why
would a second Pascal compiler not be financially viable?
Daniël Mantione
Marius
Bob Dawson
> >
> Its not what i said.. ;-)
You used a lot more words, but AFAICS that's what it boiled down to.
Personally, I have no need whatsoever for backward compatibility with D7 or
fpc (why aren't you demanding that fpc keep up with Delphi?).
And the notion that anything that can be done with a 3rd party add-in or
wizard adds no worth to the box is preposterous. I run D2007 about as close
to out-of-the-box as I can. It greatly simplifies configuration management.
bobD
Dean Hill
> There is a good C compiler for Linux: GCC.
>
> Intel, Pathscale and PGI sell C compilers for Linux and make
> good money with it. Why are they financially viable, and why
> would a second Pascal compiler not be financially viable?
Wild guess, but here is a good reason. If we add C, C++ and Pascal
developers together, we would get something like:
Pascal Market 2%
C Market 98%
The pascal market is just not presently big enough to support massive
amounts of R&D.
There is also a lot of device drivers etc written in C which people
like Intel produce their compiler for.
--
Dean
Tom
> market for *commercial* linxu IDEs is even smaller.
Interesting... What facts do you base this information on?
Tom Corey
> Leave out the language improvements because whe want to be absolutely
> compatibility with other versions of delphi (delphi7 mainly but also
> fpc if possible) and also in case delphi is becoming unusable and we
> need to downgrade (like i found out recently)
Excuse me, but that's completely absurd. In every way.
Nick Hodges (CodeGear)
> So its a very short list for win32 programmers, and just the ide
> improvements won't do it for me (certainly not for that price and
> bugs).
Marius ---
Okay, so what you are saying is that there haven't been any
improvements, except for the long list of improvements.
Did I get that right? ;-)
Nick Hodges (CodeGear)
> Interesting... What facts do you base this information on?
Can you name a commercial Linux IDE of any success?
Marius
> Okay, so what you are saying is that there haven't been any
> improvements, except for the long list of improvements.
>
> Did I get that right? ;-)
I might have over reacted (sorry about that), its not that bad Nick,
there are loads of improvements, but nothing a hardcore win32 developer
are waiting on.
Most exciting changes are on the roadmap and look far away at this moment.
Dean Hill
> Can you name a commercial Linux IDE of any success?
vi :)
--
Dean
Nick Hodges (CodeGear)
> vi :)
Does anyone actually pay for that? ;-)
Rolf Lampa [RIL]
> Tom wrote:
>
>> Interesting... What facts do you base this information on?
>
> Can you name a commercial Linux IDE of any success?
>
Put the effort on the Windows IDE, bout also provide compiler for
Linux. That's what would give maximal value.
It's entirely irrelevant for a developer using a
"several-thousand-dollar-development-tool" (Delphi) what platform he
uses for developing. Again, entirely irrelevant.
The point is instead to be able to deploy and run the applications he
produces on an X platform. It's not very difficult to figure out that
that's where the value lays.
Too much value perhaps? It might rather soon become a "political"
question. MS should be kept happy at all times.
Perhaps I need to point out that I'm perfectly happy with my Windows
machine. But multiplying application servers and web servers, based on
M$ isn't exactly as pleasant, no surprise.
It might not pass unnoticed in Redmond though, nor amongst spending IT
managers, to support Linux executables by the click of a button
directly from the Win-Delphi-IDE. <looking blushed>
Regards,
// Rolf Lampa
Dean Hill
> It's entirely irrelevant for a developer using a
> "several-thousand-dollar-development-tool" (Delphi) what platform he
> uses for developing. Again, entirely irrelevant.
Just as everyone says that .NET is not relevant in the IDE provided the
resulting code is free of it. It amounts to the same thing. I just
need the multi-platform support for the resulting executables and not
the IDE.
> Perhaps I need to point out that I'm perfectly happy with my Windows
> machine. But multiplying application servers and web servers, based
> on M$ isn't exactly as pleasant, no surprise.
These costs really hurt. Even putting Windows SBS on the server cuts
down what I can charge my clients. They need software and that
software requires a server. The server, unfortunately, adds a bundle
to the end price which pushes up the solution cost. The end result is
that I need to keep the software pricing at a level where the total
solution is reasonable and the only pricing I can effectively cut is my
own. This is worse where clients are looking for a solution that
includes something such as email. For that, I either need a Windows
Mail solutions (more boodle) or I need to supply two servers.
--
Dean
Dean Hill
> Does anyone actually pay for that? ;-)
I think the pain would be payment enough :-)
--
Dean
Andreas Hausladen
> Does anyone actually pay for that? ;-)
The UK people :-)
http://www.heise.de/bilder/96895/0/1
--
Regards,
Andreas Hausladen
Dean Hill
> Does anyone actually pay for that? ;-)
Seriously though, about the only one that springs to mind is Qt.
There are a lot here
http://linuxmafia.com/faq/Devtools/ides.html
although only some of them are commercial such as CodeWarrior and
WebSphere.
--
Dean
Andreas Hausladen
> And are they really an IDE, or a framework?
A framework + a form designer. For editing you can use kdevelop or vi
:-)
--
Regards,
Andreas Hausladen