Hi all,

something i just read at http://www.firebirdnews.org and in Marco Cantu's blog ( http://blog.marcocantu.com/blog/coderage_2008_closing.html )

regards,

something i just read at http://www.firebirdnews.org and in Marco Cantu's blog ( http://blog.marcocantu.com/blog/coderage_2008_closing .html)

Sometimes I just understand those guys. Delphi (as language) by support of FreePascal and Lazarus had been already on Mac and Linux since years! Especially since Lazarus supported GTK2 on Linux and Carbon on Mac. Why are they still busy with other alternatives (Mono/.Net, Java, RealBasic, Ruby, XCode, etc) while what they had been dreaming about is right before their eyes. Yes, FPC and Lazarus is not perfect, but what is?!

FPC's FCL and Lazarus' LCL is very much similar and compatible with Delphi's (as product) VCL. Having single codebase for different platforms and OSes is not that hard, though in some special cases it can't be 100% single codebase but it can be solved easily using IFDEFs.

Some people (mostly people in this list) have been using Delphi (as language) on more than 16 platforms (including mobile devices), yet they are (Codegear/Embarcadero fan boys) still dreaming about it. My very deep condolence goes to them. :)

-Bee-

has Bee.ography at:
http://beeography.wordpress.com

Bee --

You make a very excellent point in your post and one that I think doesn't get repeated loudly enough into the ears of the Embarcadero folks. Maybe its time for an open letter posted to
these kinds of lists and sending their president a e-mail with the link to read the comments for himself.

Market presence, that elusive and ill-defined Holy Grail that Borland chased, is now luring these folks into trying to be too many things to too many people. But what we as developers with the FPC and Lazarus must do more of and be keen about is promoting the benefits you describe and coding projects that will make the case less a matter of loyalty and more a matter of results.

The mobile/embedded market is the place every software tool company wants to be right now. It's very difficult to get Embaradero to take their eyes off the .NET market since they have decided to
sell $900 development tool sets.

We have to be smarter than they are and keep pushing the language toward these new applications.

Prince

FPC's FCL and Lazarus' LCL is very much similar and compatible with Delphi's (as product) VCL. Having single codebase for different platforms and OSes is not that hard, though in some special cases it can't be 100% single codebase but it can be solved easily using IFDEFs.

With sufficient abstraction, it's even possible without IFDEFs. My applications don't run on 16 platfroms, but it does run or 4, and there isn't single IFDEF in my code.

Some people (mostly people in this list) have been using Delphi (as language) on more than 16 platforms (including mobile devices), yet they are (Codegear/Embarcadero fan boys) still dreaming about it. My very deep condolence goes to them. :)

I can't agree more! ;-)
I've mentioned it a few times in the CodeGear newsgroup as well. If they can use a different compiler and IDE to target .NET/Mono, why can't they do the same for cross-platform support. Simply use Lazarus IDE and Free Pascal Compiler - supply many bug fixes and ship that!

They keep going on about the .NET dependency justification in the IDE, saying they didn't want to reinvent the wheel! Well, they should practice what they preach. They are going to do just that, if they want there own cross-platform support. They are stupid if they don't embrace the Free Pascal Compiler and Lazarus IDE. It would save them a ton of work!!

Regards,
- Graeme -

_______________________________________________
fpGUI - a cross-platform Free Pascal GUI toolkit
http://opensoft.homeip.net/fpgui/

What kind of nonsense is that ? What does .NET dependency in the IDE have to do with reinventing the wheel ? Does .NET come with a complete code refactoring assembly, written in C# by Microsoft or what ?

Assuming they wrote the IDE (and thus code refactoring) themselves, what is the problem to compile it as native code and get rid of .NET dependencies in the IDE ?

Really...

Michael.

In our previous episode, Michael Van Canneyt said:

Afaik it is not even a direct .NET dependancy, but a Java dependancy that they chose to resolve via J#->.NET.

I didn't get into it much, but many developers are upset about the .NET dependency in the IDE in last few Delphi releases. Now they can't run the Delphi IDE via WINE, which apparently many of them did for Delphi 7 etc...

CodeGear staff mentioned reusing refactoring features of the .NET framework in the IDE and some other stuff for the editor etc...

Regards,
- Graeme -

_______________________________________________
fpGUI - a cross-platform Free Pascal GUI toolkit
http://opensoft.homeip.net/fpgui/

Marco van de Voort wrote

Afaik it is not even a direct .NET dependancy, but a Java dependancy that they chose to resolve via J#->.NET.

Is J# still being developed? I have seen or heard anything about J# in ages.

Regards,
- Graeme -

_______________________________________________
fpGUI - a cross-platform Free Pascal GUI toolkit
http://opensoft.homeip.net/fpgui/

In our previous episode, Graeme Geldenhuys said:

> Afaik it is not even a direct .NET dependancy, but a Java dependancy that > they chose to resolve via J#->.NET. Is J# still being developed? I have seen or heard anything about J# in ages.

No idea. Just know this because it was mentioned in CG ngs when they were still alive.

Sometimes I just understand those guys.

I meant: "Sometime I just *don't* understand those guys."

Sorry for the typo. :)

-Bee-

I don't believe this argument for a second.

1. .NET has no knowledge of Pascal, so how can it be used to refactor pascal ?
2. The IDE - prior to .NET - already had extensive codetools, I would be surprised to learn that they couldn't implement the 2 or 3 things that "required" .NET using their own codetools... Reversing this argument would mean that they admit that their own codetools are a lot of junk ?

Either way, the .NET dependency IS a big hindrance for a lot of their clients, and obviously they are (again) not listening. But what else is new ?

Michael.

Michael Van Canneyt wrote

Either way, the .NET dependency IS a big hindrance for a lot of their clients, and obviously they are (again) not listening. But what else is new ?

Exactly, they don't learn! And luckily the old Delphi-maintained projects in our company are getting less and less. Soon it will be Free Pascal everywhere! :-)

Regards,
- Graeme -

_______________________________________________
fpGUI - a cross-platform Free Pascal GUI toolkit
http://opensoft.homeip.net/fpgui/

On the positive side they dumped Delphi for .NET and implemented Unicode support. Maybe they are recovering from the last decade of ill management?

There is just something I don't get about some business. Delphi is similar to Palm in my oppinion. It was a very popular technology and then it's development stagnated and ill choices have put it in a near dead state. In the case of Palm really dead.

Was it really more profitable for Borland to let Delphi nearly die? Wouldn't they profit more if they kept investing (and in the correct direction) but also had a larger market share then now?

But nothing really new, I have always heard that depending on a single vendor is a bad idea. Sure, some single vendor software are doing fine now. But how long before management decides it's more profitable to milk the software and stagnate development?

Was it really more profitable for Borland to let Delphi nearly die? Wouldn't they profit more if they kept investing (and in the correct direction) but also had a larger market share then now?

I think they had to make a descision: Either against Microsoft or with them. They chose to go with Microsoft so they now have to do everything that supports M$ OSes and avoid everything that could drive users to open source. In both ways there is no guaranty for success: If they try to compete with Microsoft they have a strong enemy and if they collaborate with them they can go down with them if open source becomes stronger and stronger.

Jürgen Hestermann.

IMO, they had made a very wrong decision. Supporting open source doesn't always mean to open source everything. Besides, by supporting open source, they could also support M$ as well. Supporting open source doesn't mean against M$. Most of open source projects come with cross-platform support in mind, including M$. Even Apple indirectly supports M$ (Safari, iTunes, etc). Nobody denies that M$ still dominates the market. So, supporting open source should be the safest choice for them. If M$ grows bigger, they can grow bigger with M$. If M$ goes down, they can still grow bigger with other platforms (linux, mac, etc).

IMO, RemObjects realize this. That's why their Oxygene/Prism could work on .Net (M$) and Mono (non M$). Also that's why their other OP-based products starting to support FPC and Lazarus (to target non M$ platforms). RO is smarter than CodeGear/Embarcadero, obviously. ;)

-Bee-

So, supporting open source should be the safest choice for them. If M$ grows bigger, they can grow bigger with M$. If M$ goes down, they can still grow bigger with other platforms (linux, mac, etc).

Many years ago (when Borland was still a real competitor of M$) Borland had problems to create good compilers because they didn't have all the information about the M$ OS (while compilers from M$ had of course). I remember that some articles stated that Borland made a deal with M$ so that they will get information but on the other hand may not support other OSes (and open source). I am not sure whether that's true but it would explain the behaviour of all Delphi owners.

Jürgen Hestermann.

Jurgen and Bee made a few rich contrasting points about Borland's past mistakes. However, we must acknowledge that Borland's biggest problem was Borland persisten failure to deliver 'price competitive' and 'high value' programming tools.

No one denies Borland's well earned reputation for innovation in Windows programming, Turbo C, Turbo C++ and Delphi blew the doors off MC Visual C and MFC programming for they Windows 3.1, Windows 95, Windows NT platforms. They lead with products that were not only feature rich, they turned out solid code and made the job of wrangling the Windows Frameworks like COM, COM+, AFC, ODBC, ASP, and MFC much easier.

However, they grabbed defeat from the jaws of victory when they overextebded their development group (Inprise), rushed incomplete and inadequate products (Kylix, Delph 7/8) to market while they
buried themselves in 'Enterprise' COBRA/IIOP, and database engines.

When they slipped, like the hare, this gave MS the catch up and pass them which allowed them to solidify their base with the OS systems.

On Wed, Dec 10, 2008 at 1:01 AM, Jürgen Hestermann <