About GUI and IDE Interface

[Bob Calco]

Marc-Antoine:

%% > I expect the first beta version to have the following more-or-less
%% > completed
%% > features, which I'm implementing in C# (.NET):
%% >
%%
%% Sounds absolutely brilliant, and I'd hate to sound critical of what is
%% obviously a fine effort... But I'm a bit miffed that you selected .NET,
%% because I live in another platform world (Mac OS X, in my case), and
%% some linux projects notwithstanding, it's fair to say that your efforts
%% will be unaccessible to a lot of the Mozart constituency, which is also
%% unix-based.

One compelling reason why I considered C# was the presence of Mono (the
platform (www.go-mono.com), not the contagion) on Unix, which is a Unix
implementation of .NET, with emphasis on C#.

So, believe it or don't I did have cross-platform issues in mind when I
picked C#.

The C# thing is, however, a recent decision, I can still be persuaded. My
design is nearly complete - certainly at a high level I know what I'm doing
now. My goal was primarily a Windows-centric IDE, since Unix type folks
(including most Mac OS X folks I know) tend to be happy with Emacs. (I said
"tend" - I didn't mean to generalize too broadly.)

Originally I thought about Delphi, which I know well, with an eye toward
Kylix, but I just don't see the market going there. Also I like C++, so I
thought perhaps C++ Builder - which has an even smaller market.

Another related factor was the influence of Anders Heijlsberg (however his
name is spelled), he's the guy who invented Delphi and C++ Builder for
Borland before going to Microsoft to architect C# and .NET. Everything I
like about the Delphi VCL (Visual Component Library) is there, like Prego,
inside .NET's component architecture. It's actually scary how quickly I was
able to pick up the whole C# development paradigm (two nights, actually) -
I'm surprised Borland can't sue... but as a developer who really liked the
VCL, C# offers a lot on top of the same basic component development
philosophy as the VCL. So that was an inducement too.

I thought about Java but I'm not a fan of Java GUIs in general. Frankly I'm
forced to write Java GUI code now at work, and one of my motivations is to
get away from it. ;)

I could I suppose write the whole thing in Oz using Tk or GTK, or (my
preference if I went this route) implement an Oz binding to wxWindows at the
C++ level. But just that effort alone could take a while, as wxWindows is an
incredibly rich GUI framework. I don't particularly like Tk's look feel,
esp. on Windows. I really like what the Python folks did with wxPython, so a
Mozart binding to wxWindows still an appealing idea to me.

wxWindows is also implemented on Mac as well, so if I did go that route, one
plus is that you wouldn't be mad at me anymore. :)

%%
%% If/when you're done, would you consider lending us some of your code
%% base back, in case someone (...) wants to re-implement it in some
%% currently  platform-neutral language like Java?

Perish the thought! Not the "giving away the code" part, but the Java part.
;)

%% I'm thinking of an
%% Eclipse plug-in for Mozart, or something like that. (If you have not
%% looked at www.eclipse.org, stop reading this boring e-mail and jump
%% there. It's quite the dream IDE.)

The IBM'ers in charge at the PTO, under whom I'm working as a sub-sub, have
us using WebSphere Application Developer, which is based on the Eclipse
framework. It's cool. Zippy, sleek, nice looking. Not a bad idea, I'll mull
over it. But it's Java!!!! AaaarrrgggghhhH! ;)

%% Now, I fully realize that not
%% everybody wants to opensource, and I won't blame you otherwise, but I
%% had to ask.

Yeah, I'll open source the "Community Edition" of Amadeus and retain the
right to charge for Pro and Enterprise modules. Though, really, I just want
to see Mozart hit the "big time" and I think the lack of an IDE other than
Emacs is a hindrance to its acceptance as a programming system in the
commercial world.

I've got plenty of other ways to make money than to milk the currently small
Mozart community for cash. I'll feel good if I can help the community grow
and find job opportunities that involve Mozart development. For that, we
need industry acceptance. That will take some sacrifice of time and effort
for folks like myself who have ins to potentially big Early Adopters, but
not until I have something that is both Killer and Easy To Use.

Oz as a language and a development paradigm is wicked cool, but Emacs just
ain't what the industry is used to, IDE wise. I mean its OK, but having the
abstraction of a Project and the visual tools is a real plus when the
"customer" would just as soon use ... <gulp> Java or, worse,
<window-shattering-shriek> VB...

(no offense to Emacs fans intended by the implication that Emacs isn't
industry standard...)

;)

Sincerely,

Bob Calco

%%
%% Marc-Antoine Parent
%%
%% -
%% Please send submissions to users at mozart-oz.org
%% and administriva mail to users-request at mozart-oz.org.
%% The Mozart Oz web site is at http://www.mozart-oz.org/.
%%


-
Please send submissions to users at mozart-oz.org
and administriva mail to users-request at mozart-oz.org.
The Mozart Oz web site is at http://www.mozart-oz.org/.