Re: Borland, Let Delphi Live !

From: C4D - Kim Madsen (kbm_at_components4developers.com)
Date: 10/24/03


Date: Fri, 24 Oct 2003 00:13:43 +0200


> There is no doubt about that. I think that what the original poster
> fears is that Delphi for Win32 development soon will be gone and that
> some of his knowledge thereby will be rendered obsolete.

I dont doubt that it may be his fear... but as long as Windows NT/2K/XP and for that matter Longhorn exists, a Win32 API
will exist, and his code will run.
For next gen. processors, MS has not been choosing... hey... this is a completely new HW platform... so this is the
right time to go dotNet only. No, they have ported their Win32 API to the platform and named it Win64 API.
I estimate that for at least 10 more years, Win32 API and Win64 API will exist.
Heck, dotNet is dependant on the API... Windows itself is dependant on it.

The only thing Im not confident about is creation of a Delphi for Win64 at any time soon.
The reason is quite simple... how many users will run on 64 bit processors the next couple of years?
Itanium have existed for around 3 years, and how many are using that at their workplace or at home?
Thus I think that the moment there is a large enough population of 64 bit servers/desktops, Borland will
confirm that they will support it, also natively. This is ofcourse completely wild guessing from my side, but
it seems natural that is the way things goes.

The VCL will exist on both Win and dotNet platforms and since its not in Borland's interest to scare away their
userbase, its
only logical if the two platforms are going to be relatively compatible. Thats how it went with Kylix. Delphi apps
could, with some tweeking, be compiled under Kylix and run on Linux. Why not expect the same will happen with Octane?

In fact I see the Octane development as rather exciting, not because of the dotNet platform, but also because whatever
new stuff in terms of class libraries and compiler syntax there will be in Octane is likely to be backported to the
Win32 Delphi to make it possible to have a common code base.

>
> The source for "conspiration theories" as you call it is what some may
> feel is vague information from Borland about the company's direction and
> future plans regarding this. Yes - don't say I said it is like this, I
> just believe that many have the feeling it is vague. Which actually
> makes no difference <g>.

true... thats usually the reason people use when putting those theories together.
But if one would use ones common sense, and think in terms of Borland having to earn money to their stockholders, its
not terrible logical for them to do lots of stuff that is designed to scare their existing customer base away.

I think its also important to remember that publishing a roadmap that shows plans for the next 3 years is almost
unprecedented in the sw industry. Even IBM doesnt do that. They are actually often worse in that sense... I have
experienced that on first hand only a couple of years ago with their ComponentBroker technology which suddenly got
obsolete 3 months after I was on a 5 day IBM course about it.

Its actually rather obvious that a 3 year plan wont work... that would require Borland to know exactly what MS, and
Sun's and IBM's and who ever else major player in this industry is planning at least for the next 3 years. And thats not
going to happen - tradesecrets.

Thus Borland is only able to tell whats going to happen so far ahead. If telling about stuff too far in the future, they
would most likely have to revise those plans many times before that period of time comes. I think Borland belive that
this situation is going to be more confusing to customers and will make more customers loose faith than not telling too
much too early.

That means that its important for people to use their common sense remembering that Borland needs to earn money. This
way I think most people can predict to some extent what the future most likely will bring.

-- 
best regards
Kim Madsen
kbm@components4developers.com
www.components4developers.com
The best components for the best developers
kbmMW          - RAD n-tier
kbmMemTable - High performance memory table
kbmWABD      - RAD web development
kbmX10           - RAD house automation


Relevant Pages

  • Re: Moving from Delphi to C++
    ... "As we celebrate the 20th anniversary of Borland, ... Each release of Delphi has ... Octane will provide full support for both Windows 2003 and the ... of the VCL framework to enable the migration of existing Win32 ...
    (borland.public.delphi.non-technical)
  • Re: Thinking Clearly
    ... keep D7 for win32 and go VS for .Net. ... way for Borland to get more money from me (not me personally, ... option than to remain with Delphi; possibly adding VS if I had new ... They still support DOS ...
    (borland.public.delphi.non-technical)
  • Re: Why was win32 killed?
    ... here, I will continue to use Delphi, there's no question on that. ... Really was looking forward to a win32 + dotNet release though. ... Borland enterprise customer. ... Borland for so long to disclose their plans in advance. ...
    (borland.public.delphi.non-technical)
  • Re: Delphi 2005 Win32 development stopped
    ... improve Win32 development or make a significantly better product for the end-user. ... Delphi, they wouldn't even merit a foot-note in marketting. ... someone from Borland or TeamB mentions for-in loops. ... and inherit all the bugs. ...
    (borland.public.delphi.non-technical)
  • missing out the vendorīs perspective!
    ... apiīs such as win32 and win64 around, is correct, you are missing out the ... it less attractive for IDE vendors such as Borland to support them. ... be the killer app here for enterprise customers). ... > API is death and in the future of Windows there will be only .NET. ...
    (borland.public.delphi.non-technical)