Re: Brian Moelk guesses (wrongly) that Delphi is Going to be Killed

1. It assumes that plugging into VS.NET would be a substantial savings
over
Borland using their own IDE. Allen Bauer has already explained that
that
option was looked at, and rejected. Here's specifically why it isn't
true:
http://blogs.borland.com/abauer/archive/2005/02/02/2530.aspx. Why
assume
Borland is doing something for reasons that Borland insiders have
already
explicitly discussed to the opposite conclusion?

I don't think I assumed that directly as Allen discusses in his blog entry. The main point I made is that Borland *must* support VS.NET in their ALM strategy, so supporting BDS is additional cost.

2. It assumes that Borland HAS to support VS.NET because developers
will demand the tool they want. Funny now that argument never works
the other way.

That was only *one* aspect of the argument; specifically focused on the grass roots level. You must admit that Delphi has been adopted in many shops because of the TP heritage. That's just a fact. If you look at the grass roots support going forward, it's clear who is winning that battle. That's all.

Borland will support VS.NET because there's money in
doing so; that's a good reason in the short term. But it's not the
goal, and not at all inconsistent with them preferring their own IDE.

Yes, Borland must support VS.NET because that is where the larger market resides. The goal isn't maximizing revenue/profit? That's news to me.

3. Brian's assumption that supporting VS.NET would be just as
effective as Borland having its own IDE completely misses a major
selling point of development stacks in the first place: Companies want
single sourcing and a consistent point of contact for help and
consulting--not vendor tag.

Well, if you are choosing a *Microsoft* platform, and considering a *Borland* solution, you have already gone with a two company solution. If you truly wanted a single company, Microsofts full stack offering will look much more appealing no matter what.

4. Then too, it misses the point that MS will eventually have a stack
of its own. What happens to Borland then?

I don't know what happens to Borland's ALM strategy when MS gets in the game? I imagine they will continue to try to be better than MS' Team System. 

Being a VS.NET plug-in
vendor establishes a very low ceiling: either you'll never amount to
much, or MS will eat you. They don't do partners well.

But Borland doesn't want to be selling IDE's anymore anyway. The IDE is baggage for their ALM strategy. That's the point.

5. It argues that Borland can integrate its developer stack and the
Delphi language into VS.NET as easily as they could with their own
IDE. Frankly, I don't see how any developer who's ever been assigned
an integration task could buy that for a heartbeat: weave into a black
box you have no control over whatsoever as easy as integrating into
your own code, where the author and maintainer is a couple of
doors/cubes down? Absurd is too mild.

How many third party components and frameworks do you use? Do you integrate them together? Do you build your applications on top of database servers? All of those functions are things developers do every day. If integration was much more difficult than building your own stuff, we'd all roll our own socket libraries or GUI frameworks. But at the end of the day my customers aren't paying me to build a GUI framework, they want a usable GUI. Just as Borland's customers are going to pay them for their ALM solution, not for IDE's or compilers.

6. It argues that the integration of JBuilder over Eclipse is a model
of how Delphi could be integrated into VS.NET. That rather misses the
point that Eclipse is an open source project, controlled by a
consortium of which Borland is a member, and designed from the ground
up as an integration framework. VS is none of these. Rather, Borland's
inability to get the CF untangled from VS shows how low in priority
being pluggable remains to MS. When it comes to integration, MS's
stance is the classic "My way or the highway."

I recognize there are differences between Eclipse and VS.NET. The only question that you need to answer is that for *ALM* integration, is VS.NET open enough. ALM integration doesn't have squat to do with CF designers. Just use C#...big whoop.

7. It misses the point entirely that JBuilder IS being rewritten over
Eclipse: if having your own IDE weren't important, why wouldn't
Borland simply resell some other Eclipse-based IDE?

We don't know what form JBuilder over Eclipse will take when released. I suspect that it will be *heavily* focused on ALM integration, with most of the developer productivity stuff ported to Eclipse from PrimeTime. But hey, I could be wrong.

Clearly, there has
to be a reason that no other vendor's Java IDE would do. 

ALM integration...of course.

The money
Borland is putting into fundamentally re-building JBuilder over
Eclipse cannot be trivial, and yet, if Brian were right, Borland would
have no motivation for doing it at all.

No, they absolutely have motivation for building ALM integration into an IDE. But ALM integration is a far cry from the IDE itself.

and finally,
8. Borland's actions over the last two years are flatly inconsistent
with
Delphi being frozen. The simple fact is that Borland's strategic
movement
towards ALM strategy has been underway for several years--publicly for
at
least two.

Yep, but Borland was also hitting their revenue numbers. There is quite a difference in the way public companies react when they miss their earnings estimates.

And yet Delphi change in that time has actually
accelerated, not
diminished. Compiler changes and language extensions are still going
in--for
win32 as well as .NET; BC/C++ is being added; a massive effort is
being made
to get everything back up to snuff on the quality/reliability side
(see,
among other places, here:
http://blogs.borland.com/stevet/archive/2005/07/09/20220.aspx).
Borland is
not only working on DeXter, but on Highlander--the .NET 2.0 version to
follow DeXter.

That's because they haven't reached the point of diminishing returns for Delphi...yet. Cash cow.

--
Brian Moelk
bmoelk@xxxxxxxxxxxxxxxxxxxxxxxxxxx
http://www.brainendeavor.com


.

Relevant Pages

  • Re: Brian Moelk guesses (wrongly) that Delphi is Going to be Killed
    ... The main point I made is that Borland *must* support ... > VS.NET in their ALM strategy, ... magnitude worse by ceding them the IDE as well. ... > I recognize there are differences between Eclipse and VS.NET. ...
    (borland.public.delphi.non-technical)
  • Re: Everybodys interested, but not *that* much...
    ... For years now Borland + people here have talked about the synergy ... synergy that a split can be made that is beneficial to both sides. ... creating more and better integration between IDE and ALM tools. ...
    (borland.public.delphi.non-technical)
  • Re: What an odd move from Borland
    ... Borland has been sitting on 200+ million in cash for several years--while ... using that excess cash to acquire ALM tools. ... And because the line is profitable, the only way that selling the IDE ... that expectation. ...
    (borland.public.delphi.non-technical)
  • Guess for Borlandīs new Corporate Roadmap
    ... I donīt work for Borland nor do I have any other source of information ... While the latest JBuilder-release has a superb CaliberRM-integration, Delphi ... Other ALM-tools such as Together are available for JBuilder and even ... Java group and the .NET-business unit at borland on this unified IDE, ...
    (borland.public.delphi.non-technical)
  • Re: Opportunity passed by
    ... migrating from Delphi to C# for .NET development? ... Borland already sealed their own fate insofar as dev. ... ments offer is the fact that they are all under the same IDE, ... then there wouldn't have been a CLX for Windows and the CLX on Linux ...
    (borland.public.delphi.non-technical)