Re: Captain Jake's Top Ten List of what I'd like to see in thenextversionof Delphi

Not always. I never bought Delphi 7, for example. My current employer never
moved up from Delphi 6, for another example. I'd love to use D2006 at work,
but for that to happen we'd have to buy a new version of the entire
DevExpress set of components since we have only the DCU version, in addition
to the substantial costs of D2006 itself, as well as a few other components.
That makes an already difficult sell even harder.

Right, but even if you had the source, you'd loose but the most basic all IDE
integration due to lots of changes in the integration.

When I moved my newsreader over to D2006 I remember it taking only about a
day, because I had bought the source code to any third-party components I
was using. My philosophy in my personal programming projects is to use only
those component for which I have the source. However, I have no such control

This is imo how it should be, even if you never ever look at it. If nothing
else it's risk minimizing investment.

over firms in the market for a Delphi developer. More often than not we get
pulled into a project that was started a long time ago and built according
to a design philosophy that may be quite different than our own.

Yes that happens a lot. But imo that project then should stay in the
environment it was built with. There's nothing that stops from having multiple
Delphi versions installed. As a matter of fact I've D1, D3, D5, D6, D7 and
BDS2006 on my dev machine together with various C++, CBuilders, VS and VS.net.
D1 is there just for one, the rest is still in use in various products I have
to maintain.

I don't expect them to furnish anything past the version I purchase. I would
like for that version to work through the next several versions of Delphi. I
would also like to be able to use a new version of Delphi without having to
buy new versions of components, because, quite frankly, at most employers it
simply is not going to happen if it costs too much.

Understood.

There is a principle of good design that says you never change or delete
parts of an already-published interface, you simply add new non-conflicting
functionality. If you follow this rule, you greatly minimize the need to
re-test everything that is built on a particular interface just because you
add new functionality.

I agree with that for higher level interfaces, at the RTL level imo it's just
not practical or probably not even possible.

--

Hannes Danzl [NexusDB Developer]
Newsgroup archive at http://www.tamaracka.com/search.htm
.

Relevant Pages

  • Re: Thinking Clearly
    ... I inferred it from "But IMO, there is no tangible benefit to moving Delphi ... Win32 code forward to .NET. ... Sometimes good theories don't pan out, particularly when normal market ...
    (borland.public.delphi.non-technical)
  • Re: The alternative Delphi roadmap to success
    ... How do you interpret Borland going out of it should it be commercially viable? ... If MS had decided to use Pascal instead of C, for Windows, we wouldn't have that number of patches and hotfixes, IMO. ... Without a chance to demonstrate the advantages of Delphi, over other languages, where should new Delphi users come from? ... The development could be simplified, if Delphi.NET would become a language module in the Microsoft VisualStudio, instead of suffering from an poorly documented language and IDE of it's own. ...
    (borland.public.delphi.non-technical)
  • Re: The Delphi Survey
    ... IMO, Delphi has yet to break out successfully from it's rich GUI ... really a new OS platform of any significance. ... There are essentially 4 competitive strategies: price, differentiation, ...
    (borland.public.delphi.non-technical)
  • Re: Why arent you upgrading?
    ... IMO, ... I don't think that DevCo should be afraid of making and keeping ... Delphi different. ... Native code has its niche, managed code has its niche, ...
    (borland.public.delphi.non-technical)
  • Re: Definition of outdated...
    ... That is one of my main topics regarding Delphi. ... Its imo where Delphi would ... Map etc and no generics support to ... There are loads more functionality in Java for example for this stuff, ...
    (borland.public.delphi.non-technical)