Lynn Fredricks wrote:
I have a growing fear of xCode - about its serious disadvantage for
developers from a business perspective.
Xcode is merely an IDE. I am using Eiffel, Ruby, FreePascal, and pure
C++ in Xcode. I could use Java, Python, and a few other languages, too.
We shouldn't confuse the compiler with the editor.
CodeWarrior was from a third party, and it behooved the third party to be as
backwards compatible as possible with older hardware configurations -- why?
It's what customers wanted, so they could continue to ship apps that work
with the old hardware out there. Metrowerks didn't sell the boxes to run the
software, so what did they care?
CodeWarrior had to keep adding features and reasons to upgrade. If I
used third-party libraries and frameworks within CW, I had to make sure
I was at the same release version for object files. I stopped CW at
version 5.x -- but if I wanted the latest and greatest UI elements and
features of operating systems, I had to consider upgrades. Metrowerks
didn't make their product appealing enough... and now it is limited to
Freescale products and few other side things.
Borland is in the same boat. They just couldn't make a compelling case
to enough developers that their IDE (JBuilder comes to mind) was so much
better than Eclipse that it justified extra money. Is JBuilder better?
Yes. Can I use the free tool? Yes. Therefore, unless the commercial is
10x better, I'll settle for the free tool.
Now here's xCode, which is a very nice environment. They don't make money
off of it directly. So how is it a profit center? By enhancing the sales of
other products Apple makes.
Apple needs products that work with and use their hardware. The more
productive Xcode developers are, the more products support the Mac, the
more Macs Apple can sell. It's the same thing with any operating system.
The software is what people want to use -- few of us care about the
hardware other than the quality of support.
If Joe's Handy-Dandy IDE were the best Mac tool, and blew the doors off
Xcode and the underlying free compiler sets I download from various
places, then I would use Joe's HDI. For many here, RB gives them quick
development -- which might be their priority.
Sure, you can make a decent application with RB, and probably faster
than in Objective-C. (I'd say I could do things as fast in
Delphi/FreePascal, but I'd be locked into Windows.) I wouldn't compare
RB to what the best tools on any given platform can do. As has been
stated, it would be really difficult to code an Office or Photoshop
clone in RB that matches feature-for-feature the big guys. RB is a neat
tool for some things, not for others.
Not every project will require the fastest, lowest-level compiler tools.
In fact, probably darn few applications demand low-level tools. One
reason I like PowerBASIC is that I can code x86 inline with BASIC. You
can make a *blazing* DLL in that manner. But tricks like that lock you
into a specific CPU family, much less a specific OS.
RB's selling point is portability, and some might say it's
"easy-to-learn" syntax. It's up to each developer to decide if parts of
a program demand something lower-level or if an entire program needs
another approach. I find RB lacking in a lot of areas... by the time I
budget for third-party plug-ins (written in another language, of
course), I have to decide if the productivity gains justify the costs --
or would learning Objective-C eventually result in efficiency and speedy
code.
As it stands, I remain conflicted.
- Scott
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>
Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>
