Ben Smith CodeGear CEO webcast

[savage]

Listen to what the CEO has to say about the future of Code Gear here (http://bdn.borland.com/article/33861).

[JernejL]

I am not pleased, i want a .net-less delphi which compiles faster, 64 bit compiler, and smaller more optimized programs. :(

[WILL]

I like this Ben Smith guy. He's 'no BS' which is what CodeGear needs right now.

I'd seriously like to see what happens with them once they manage to sever Borland ties. All that separation overhead sounds like it's going to be bogging them down a fair bit for a while.

It sure sounds like they're busy as heck though. "I particularly don't like seeing the Inprise signs either... I think those should go away as well." :D

[savage]

An Email I just sent to Ben Smith and David I....


Hi Ben and David,
I recently listened to your webcast and was impressed to hear that you were keen to make the company a developer orientated company again. I believe that this can only be a good thing. Just as Apple has been brought back from the brink of death, so to can CodeGear rise from the Borland ashes.

My concern is CodeGear's long term strategy and roadmap. I, like many other lovers of Delphi ( since version 1.0 in 1995 ), am worried about the seemingly unbreakable ties with the .NET platform. I honestly believe that anyone who is serious about doing .NET development will be using C#. Not because C# is better than Delphi, but because it is always more up to date than Delphi.NET is.

.NET 3.0 ( ok it's rebranded 2.0 with a few extra bells and whistles ) is about to be released, while the CodeGear road map talks about the imminent release of Delphi for .NET 2.0. CodeGear has to stop playing catchup! Codegear needs to renegotiate better terms with Microsoft so that CodeGear is kept in the loop of any new happenings/features to .NET. I think developers would be willing to put up with a 3 month delay after Microsoft's release to get a great product, but any longer than that and I, and many developers like me, would consider it a dead horse. If you are indeed partners with Microsoft, then they need to be more forth coming with the partnership so that CodeGear can gain something out of it. Otherwise, sever the ties and look to being more innovative rather than playing catch-up and think about providing a unique selling point. Which brings me onto my second concern...

The lack of more native compiler support for various Operating Systems and Processors is also worrying. Kylix was a great move in that direction, but for what-ever internal reasons ( probably lack of money and resources ), it was left to die. Borland should have stuck it out a little bit longer. Free Pascal (aka FPK Pascal) is a 32 and 64 bit professional Object Pascal ( Delphi 7 compliant ) compiler. It is available for different processors: Intel x86, Amd64/x86_64, PowerPC, Sparc. The discontinued 1.0 version also supports the Motorola 680x0. The following operating systems are also supported: Linux, FreeBSD, Mac OS X/Darwin, Mac OS classic, DOS, Win32, OS/2, Netware (libc and classic), MorphOS and recently was made to work on a Game Boy Advance. It also allows you to develop on one OS and target another OS.

I propose that CodeGear drop it's native Delphi compiler as it is currently no where near as flexible as Free Pascal and leverage Free Pascal to target more native environments. Just as there are several brands of Linux, so to could there be a CodeGear brand of Free Pascal. CodeGear would not need to invest in making a cross-platform compiler, but could instead enhance the FreePascal compiler with various speed and language enhancements. It could then be integrated into the Delphi IDE and released ( in a similar way to how Linux is repackaged ) and CodeGear would provide commercial support for this new kickass compiler/IDE combination. Any business who is serious about Object Pascal will buy the product as soon as it is released as it will come with a state of the art IDE and would allow them to target other Operating Systems ( hmm MacOS X's popularity is steadily on the increase and Linux continues to gain ground as fewer people plan to move to Vista ). Then three to six months after the release of the product, the compiler enhancements that CodeGear has made, could be merged into the FreePascal CVS/SVN. This would benefit the open source community, but CodeGear would also benefit from any bug fixes/enhancements that the community apply to the compiler as well. One hand washes the other.

CodeGear needs to differentiate itself from the rest of the compiler pack. Just as it has leveraged Eclipse to improve it's JBuilder brand, so to can it leverage Free Pascal and provide a unique selling point in the native compiler arena.

I obviously don't have all the answers, or don't know what is going on inside CG, but I hope this email at least provides some food for thought. Please don't discount the idea, just because it involves Open Source as I believe there is room for Open Source and commercial development to work together and be benefitial for all.

I think that most Delphi developer are probably willing to give CodeGear 2-3 years, tops, to turn things around and become the great developer company that we all fell in love with. If you achieve it, you will become legends, if it fails, CodeGear will be just a footnote. I hope CodeGear becomes a legend!



Sincerely,


Dominique Louis

[FNX]

I totally agree, nothing else to say :)

I just hope that they will read and take good consideration of your mail!
The thing i don't understand is why or how people in Borland-Inprise-
Turbo-CodeGear can't figure all this out theirselves...

I mean, i think at people at Borland as the 'elected' ones, they SHOULD
be the gurus, the gods and saints together about pascal! I can't really
imagine that no guy there knows of PGD or other great communities,
forums (Free Pascal or 3rd party VCL (not INDY damnit!!!) ) out there
and finally that no one is reading the direction that us user are taking
for our development needs! WAKE UP!!

Maybe Dominique you should have included some links to posts on this
forum as well as others so that they read at least ONCE what people
outside of their corporate's walls think.

Let's see what happens next...

[Setharian]

I seriously doubt that CodeGear would drop its compiler and use FPC instead....no offense to the FPC guys, but much of the source is a mess, it will take ages till someone, who never looked at the sources before, will be familiar enough with it to implement custom language features....also they'd have to provide new implementations of the system, sysutils, classes units and of course the whole VCL (for different operating systems)....if they'd use FPC-RTL and/or LCL they could just drop all code development and only work on the IDE...so that idea is not going to happen, at most integrating support for a different compiler.....

[WILL]

Well truth is that even though PGD and the game development community inside the Pascal fold is growing we are still less than at least a sixteenth of the whole of their business' total market so our direct needs would not sway them a whole ton on our own. We have to sort of compromise some leverage as we are still a growing faction among Pascal developers.

However it would pay them well to be able to recognize a lucrative and exciting venture with a ton of potential for game dev. But in their defence it's not so easy to look outside your box to find these things. We are indeed in many ways the ultimate underdog of the software industry and programming profession.

:arrow: Heavily saturated $100,000 US Dollar market projects taking up all the spotlight on next gen consoles.

:arrow: Indie developers are new and hip and there are a ton of them around that have not and will not succeed thus making it a big hit & miss market for most.

:arrow: "Pascal is dead." and "Oh yeah, I played with that in highschool."

:arrow: Borland, Inprise & those poor sacrificial lambs.

:arrow: Education for Pascal programmers is just not there!


I think I've nailed the basics just to be brief, but this is why we are not top on their list. So much fodder to upturn. But it takes time. We are growing and even though there are a lot of challenges, Pascal will return to it's rightful place in the software market and as indie publishers and more casual gaming markets becomes available things will look brighter for us all here I believe.


On the side of the email though, I think it's good. If nothing else it will give Ben some things to think about. And if nothing else it will help to open their eyes a bit to the kinds of other options they now have now that they are on their own and have a new start of things without the Borland bog.

[savage]

I seriously doubt that CodeGear would drop its compiler and use FPC instead....


I would doubt it as well, as they have a lot invested in that IP. But they really need to rethink their strategy. Going down the same path will not yield the results most of us expect and will continue to lose them customers. I would be just as happy if they rewrite dcc32 and rename it dcc and it runs on all the platforms and environments that FreePascal does. At least that way they would not be dependent on a 3rd Party, but then then how long would it take them to make Delphi as portable as FreePascal. These are all trade-off etc that they need to assess.


no offense to the FPC guys, but much of the source is a mess, it will take ages till someone, who never looked at the sources before, will be familiar enough with it to implement custom language features....

We have not seen the dcc32 C/C++ source code either, so it could be the better or worse than FPC's source code.


also they'd have to provide new implementations of the system, sysutils, classes units and of course the whole VCL (for different operating systems)....if they'd use FPC-RTL and/or LCL they could just drop all code development and only work on the IDE...so that idea is not going to happen, at most integrating support for a different compiler.....

I never said the road less traveled would be easy :), but it would give them more options. My argument is that if Linux can be turned into a relatively successfull business model, then why can't FreePascal be used in a similar manner?

[Setharian]

if they'd do all the hard work of porting VCL/RTL to other operating systems as well as the compiler itself, it would be a huge step forward....but I think they are heading more the .NET way what is, in my opinion, not the correct one...generics, 64-bit, cross-platform, full unicode support and maybe some other language enhancements would be really cool...also many developers who do also low-level stuff including assembly would like to see new features like forcing function entry alignment, better FP code generation, automatic stack alignment for local double/extended parameters etc....the list of these could grow on...the question is if the CodeGear will decide to implement these.....but I must agree that all-in-all the seperation from Borland is a good thing as well as the new CEO looks he knows what he is talking about and has enough experience....